Description
A z990 Technology Overview
z990 — The Platform Right-Sized for Server Consolidation:
The design of z990 represents a major change for the zSeries platform. zSeries 990 introduces
a new superscalar microprocessor architecture exploiting the CMOS9S-SOI technology and improving
uniprocessor performance. A significant capacity increase has been achieved with the introduction
of:
-
Up to 256 gigabytes of memory
-
Up to 96 gigabytes of bandwidth for data communication via up to 48 Self-Timed Interconnect (STI)
host buses
-
A new Channel SubSystem (CSS) — two Logical Channel SubSystems (LCSS's) can exist for horizontal
growth, supporting up to 256 channels per LCSS for a total of 512 channels per system
-
Double the number of Logical Partitions (LPARs) — now 30 LPARs
-
Increased channel maximums for ESCON, FICON Express, and OSA-Express
-
Quadruple the number of HiperSockets — now 16 internal LANs
-
Three cryptographic options:
-
New CP Assist for Cryptographic Function (CPACF) on every Processor Unit (PU)
-
New PCIX Cryptographic Coprocessor (PCIXCC) feature
-
PCI Cryptographic Accelerator (PCICA) feature
Integrated Cluster Bus-4 (ICB-4), more than two times faster than an ICB-3
Coupling Facility Control Code (CFCC) Level 12 supporting:
-
64-bit addressing for larger structure sizes
-
Up to 48 internal tasks
-
Message Time Ordering
-
System-Managed Coupling Facility Structure Duplexing
Up to 32 Processor Units (PUs)
-
Four models — each model indicates the maximum number of PUs available for purchase
(Models A08, B16, C24, D32)
-
Each PU can be individually purchased
A Processor Unit is the generic term for the z/Architecture processor on the MultiChip Module (MCM)
that can be characterized as a:
-
Central Processor (CP) to be used by the operating system
-
Integrated Coupling Facility (ICF) to be used by the Coupling Facility Control Code (CFCC)
-
Integrated Facility for Linux (IFL)
-
Additional System Assist Processor (SAP) to be used by the Channel SubSystem (CSS)
With the introduction of z990 and the four-model structure, only one CP, ICF, or IFL must be
purchased and activated for any model. PUs can be purchased in single PU increments and are
orderable by feature number.
Refer to the
Customer Initiated Capacity
section of this announcement for information on permanent and temporary capacity.
Server Consolidation Facilitated:
With the expanded capacity of z990, and its new Channel SubSystem, there is a significant increase
in system scalability, facilitating consolidation of multiple servers into one z990.
With z/OS V1.4 z990 Exploitation Support, you will be able to install more operating system images
and up to a total of 512 channels (up to 256 channel maximum per operating system) allowing each
z990 to access more I/O and networking channels — ESCON (up to 512 channels), FICON Express (up to
120 channels), and OSA-Express (up to 48 ports).
Maximum
Models PUs Memory SAPs Spares STIs I/O Cages CHPIDs
A08 1 to 8 8 to 64 GB 2 2 12 1 to 3 512
B16 1 to 16 8 to 128 GB 4 4 24 1 to 3 512
C24 1 to 24 8 to 192 GB 6 6 36 1 to 3 512
D32 1 to 32 8 to 256 GB 8 8 48 1 to 3 512
Note:
There are 256 CHPIDs per LCSS.
The Performance Advantage
Performance Estimates with z/OS V1.4:
IBM's Large Systems Performance Reference method provides comprehensive z/Architecture
processor capacity data for different configurations of Central Processing Units across a wide
variety of system control program and workload environments. For zSeries z990, z/Architecture
processor capacity is defined with a 3XX notation, where XX is the number of installed Central
Processing Units.
Compared to the z900, z990 is designed to offer the following z/Architecture Central Processing Unit
performance advantage:
Expected performance relative to the 2064 z900 216:
2084-D32:
332: 2.46 -- 2.98 times
331: 2.40 -- 2.89 times
330: 2.35 -- 2.80 times
329: 2.29 -- 2.71 times
328: 2.23 -- 2.62 times
327: 2.17 -- 2.53 times
326: 2.11 -- 2.44 times
325: 2.05 -- 2.35 times
2084-C24, 2084-D32:
324: 2.00 -- 2.26 times
323: 1.94 -- 2.17 times
322: 1.88 -- 2.08 times
321: 1.82 -- 1.99 times
320: 1.76 -- 1.90 times
319: 1.71 -- 1.82 times
318: 1.65 -- 1.73 times
317: 1.59 -- 1.64 times
Relative to the corresponding 2064 z900 models, the expected performance of the following z990
configurations is:
2084-B16, 2084-C24, 2084-D32:
316: 1.51 to 1.55 times 2064-216
315: 1.52 to 1.55 times 2064-215
314: 1.52 to 1.55 times 2064-214
313: 1.52 to 1.55 times 2064-213
312: 1.52 to 1.55 times 2064-212
311: 1.52 to 1.56 times 2064-211
310: 1.52 to 1.56 times 2064-210
309: 1.52 to 1.56 times 2064-2C9
2084-A08, 2084-B16, 2084-C24, 2084-D32:
308: 1.52 to 1.56 times 2064-2C8
307: 1.52 to 1.57 times 2064-2C7
306: 1.52 to 1.57 times 2064-2C6
305: 1.53 to 1.57 times 2064-2C5
304: 1.53 to 1.58 times 2064-2C4
303: 1.53 to 1.58 times 2064-2C3
302: 1.54 to 1.59 times 2064-2C2
301: 1.54 to 1.61 times 2064-2C1
The above performance estimates are Internal Throughput Rate Ratios (ITRR) based on measurements and
projections using standard IBM benchmarks in a controlled environment. The actual throughput that
any user will experience will vary depending upon considerations such as the amount of
multiprogramming in the user's job stream, the I/O configuration, and the workload processed.
Therefore, no assurance can be given that an individual user will achieve throughput improvements
equivalent to the performance ratios stated here. For more detailed performance information,
consult the Large Systems Performance Reference (LSPR). Customers interested in LSPR should contact
their IBM representative.
Performance Assists for Linux and z/VM
z990 Adapter Interruptions for Linux and z/VM
Reducing Host Programming and Adapter Overhead for FCP and OSD CHPID Types:
The new z990, Linux for zSeries, and z/VM V4.4 are working together to provide performance
improvements by exploiting extensions to the Queued Direct Input/Output (QDIO) architecture.
Adapter interruptions,
first added to z/Architecture with HiperSockets, is designed to provide an efficient,
high-performance technique for I/O interruptions to reduce path lengths and overhead in both the
host operating system and the adapter — FICON Express when using the FCP CHPID type, and
OSA-Express when using the OSD CHPID type.
In extending the use of adapter interruptions to FCP and OSD (QDIO) channels, the programming
overhead to process a traditional I/O interruption is reduced. This benefits OSA-Express TCP/IP
support in both Linux for zSeries and z/VM, and FCP support in Linux for zSeries.
Adapter interruptions apply to a z990 FICON Express channel when in FCP mode (FCP CHPID type), which
supports attachment of SCSI devices in a Linux for zSeries environment. Adapter interruptions apply
to all of the OSA-Express features available on z990, whether offered as a new build or on an
upgrade from z900 when in QDIO mode (OSD CHPID type). This support is exclusive to z990.
Note:
HiperSockets supported adapter interruptions at introduction.
For Linux for zSeries support, refer to the following Web site for further information:
This support is available in z/VM V4.4. Refer to Software Announcement
203-128
, dated May 13, 2003.
Refer also to the following Web site:
The Queued Direct Input/Output (QDIO) architecture is utilized by:
-
FICON Express when configured to support SCSI devices using the Fibre
Channel Protocol (FCP CHPID type) in a Linux for zSeries environment
-
HiperSockets (IQD CHPID type) for IP communication between LPARs, between
an LPAR and a VM guest, or between VM guests
-
All of the OSA-Express features in QDIO mode (OSD CHPID type)
QDIO continues to be the preferred architecture for high-speed
communication, reducing host interruptions and improving response times.
z990 Performance Assist for V=V Guests in the z/VM Environment
Passthrough of Adapter Interruptions for FCP, IQD, and OSD CHPID Types:
z990's support of virtual machine technology has been enhanced to include
a new performance assist for virtualization of
adapter interruptions.
This new z990 performance assist is available to V=V guests (pageable
guests) that support QDIO on z/VM V4.4. The deployment of adapter
interruptions improves efficiency and performance by reducing overhead.
The
z990 performance assist for V=V guests
is a passthrough architecture that reduces host programming overhead by
avoiding the need to stop guest processing when adapter interruptions are
presented. Without the assist, the z/VM control program must intercede
to process and route the adapter interruptions.
The z990 performance assist improves Linux for zSeries performance under
z/VM by allowing guest I/O — FICON (FCP CHPID type), HiperSockets (IQD
CHPID type), and OSA-Express (OSD CHPID type) — to be handled with
minimal z/VM overhead through direct presentation of adapter
interruptions by the server to a pageable guest, boosting I/O
performance.
This support is available in z/VM V4.4. Refer to Software Announcement
203-128
, dated May 13, 2003.
Refer also to the following Web site:
Breaking the Barrier — A New Channel SubSystem
Breaking the Barrier of 256 Channels:
With the introduction of a new system structure and all of its scalability benefits, it is
essential that the Channel SubSystem (CSS) also be scalable and permit "horizontal" growth. This is
facilitated by allowing more than one Logical Channel SubSystem (LCSS) on a single z990. Consistent
with this premise, IBM is introducing increased connectivity by offering the following channel/port
maximums. For reference, the new maximums are contrasted with the z900 maximums.
Channel z990 z900
Type Maximum Maximum
ESCON 512 256
FICON Express 120 96
OSA-Express 48 24
The maximums for all other features remain the same as with z900.
The new Channel SubSystem structure offers the following:
-
Two LCSS's.
-
Each LCSS can have up to 256 channels defined.
-
Each LCSS can be configured with one to 15 Logical Partitions (LPARs).
-
Cannot exceed 30 LPARs per system.
Spanned channels which can be shared among LPARs across LCSS's. The following may be spanned:
-
Internal Coupling Channels (ICP CHPID type) for Parallel Sysplex®.
-
HiperSockets (IQD CHPID type).
Note:
There is no change to the operating system maximums. One operating system image continues to
support up to a maximum of 256 Channel Path Identifiers (CHPIDs).
Support for dynamic I/O configuration, with z/OS V1.4 z990 Exploitation Support and z/VM V4.4, will
be extended to allow channel paths, control units, and devices to be dynamically added, changed, and
deleted in multiple LCSS's. However, dynamic I/O configuration is only supported in LCSS0 until
October 2003.
Until z/OS V1.4 z990 Exploitation Support is available, z/OS can run only in a single LCSS.
Therefore, z/OS may only execute in LCSS0. z/VM V4.4 may execute in LCSS0 and LCSS1. Refer to the
Software Requirements
section for more information on availability dates.
When z/VM is the controlling LPAR, z/VM's dynamic I/O support is designed to handle all of the new
elements of the multiple Channel SubSystem facility for dynamic I/O configuration changes. To
dynamically change the I/O configuration one of two techniques can be employed:
-
z/VM Control Program (CP) suite of interactive dynamic I/O commands
-
HCM/HCD — New configuration management tools
The I/O SubSystem (IOSS) continues to be viewed as a single Input/Output Configuration Data Set
(IOCDS) across the entire system with up to two LCSS's. Only one Hardware System Area (HSA) is
used.
In the past, a two-digit identifier, a CHPID, was associated with a physical I/O connection. Now a
three-digit Physical Channel Identifier (PCHID) is being introduced to accommodate the mapping of
512 channels to two LCSS's with up to 256 CHPIDs each. CHPIDs continue to exist and will be
associated with PCHIDs.
As a result of the introduction of PCHIDs, a new CHPID Mapping Tool (CMT) is being introduced and
the CHPID report from the IBM Configurator For e-business (eConfig) is replaced by a PCHID report.
CMT is available from Resource Link™ as a standalone PC-based program.
Refer to z990 documentation for additional information on the new mapping
tool, CHPIDs, and PCHIDs.
Spanned Channels
Transparent Sharing of ICs and HiperSockets:
With the introduction of the new Channel SubSystem, transparent sharing
of Internal Coupling Channels (ICs) and HiperSockets is possible. The
Multiple Image Facility (MIF) allows sharing of channel resources across
LPARs. ICs and HiperSockets can be configured as MIF
spanning channels.
Spanning channels is the ability for Internal Coupling Channels and
HiperSockets channels to be configured to multiple Channel SubSystems,
and be transparently shared by any or all of the configured LPARs without
regard to the Logical Channel SubSystem to which the LPAR is configured.
This support is applicable to Internal Coupling Channels (ICP CHPID type)
for Parallel Sysplex and to HiperSockets (IQD CHPID type).
IC spanned channels has no prerequisite software requirements.
HiperSockets spanned channels is supported in z/OS V1.2, z/OS V1.3, and
z/OS V1.4 with PTFs. HiperSockets spanned channels is supported in z/VM
V3.1, z/VM V4.2, z/VM V4.3, and z/VM 4.4. Refer to the
Software Requirements
section of this announcement for more information.
Another Breakthrough — More Subchannels for Increased Connectivity:
With two Logical Channel SubSystems come more subchannels. There has been a 63k subchannel
architectural limitation. Since there was one CSS, there was a maximum of 63k subchannels. With
two Logical Channel SubSystems, each LCSS can have its own set of 63k subchannels.
With two Logical Channel SubSystems you can have:
This capability relieves the I/O device configuration constraints experienced by large system
configurations.
A Giant Step — Support for 30 Logical Partitions
Looking back in time, IBM introduced the Processor Resource/Systems Manager™ (PR/SM™) feature in
February of 1988. Prior to that time, IBM mainframes could only be physically partitioned.
When IBM introduced Logical Partitions (LPARs), a maximum of
four LPARs
was available, offering the flexibility to allocate the hardware resources.
In June of 1992, IBM introduced support for a maximum of
10 LPARs
and announced the Multiple Image Facility (MIF, also known as EMIF), which allowed sharing of ESCON
channels across LPARs, and since that time has allowed sharing of all channels across LPARs (such as
ISC-3, FICON Express, and OSA-Express).
In June of 1997, IBM announced increased support — up to
15 LPARs.
The evolution continues with a giant step —
doubling
the number of LPARs supported! IBM is announcing support for
30 LPARs,
15 LPARs per LCSS. This support is exclusive to z990 and is planned to be available in
October 2003.
Customer Initiated Capacity — Technology On Demand
Customer Initiated Upgrade Enhancement:
When your business needs additional capacity quickly, Customer Initiated Upgrade (CIU) with the
Express option is here to deliver it.
CIU is designed to allow you to respond to sudden increased capacity requirements by downloading and
applying an IBM
zSeries
Processor Unit (PU) and/or memory upgrade via the Web, using IBM Resource Link and the Remote
Support Facility. IBM now has a faster process for upgrading your server. With the Express option
on CIU, an upgrade may be installed within a few hours after order submission.
Permanent Upgrades
Orders (MES's) of Processor Units (PUs) and memory for IBM
zSeries systems that can be delivered by Licensed Internal
Code, Control Code (LIC CC) are eligible for CIU delivery. This includes the upgrade of PUs for
z800, z900, and z990 as well as memory for z900 and z990 systems up to the maximum available on the
installed system.
Temporary Capacity
IBM eServer™ On/Off Capacity on Demand:
When your business needs short-term additional capacity, On/Off Capacity on Demand (On/Off CoD)
is designed to deliver it. On/Off CoD is designed to temporarily turn on as Central Processors
(CPs), previously uncharacterized PUs, unassigned CPs, and unassigned Integrated Facilities for
Linux (IFLs) that are available within the current model. This capability cannot coexist with
Capacity Backup Upgrade (CBU). On/Off CoD is delivered through the function of CIU. To participate
in this offering, you must have installed CIU Enablement (#9898) and On/Off CoD Enablement (#9896).
Subsequently, you may concurrently install temporary capacity by ordering On/Off CoD Active CP
(#9897) up to the number of current CPs, and use it for an indeterminate time.
You will be billed on a 24-hour basis. Each month your bill will be calculated for the sum of all
orders installed within the prior month. Monitoring will occur through the server Call Home
facility and a bill will be generated if the capacity has been enabled for any portion of a calendar
month. You will continue to be billed for use of temporary capacity until you return the server to
the original state. After concurrently returning to the original state, you may choose to activate
a new On/Off CoD upgrade which can be different from the previous upgrade. When you dispose of the
server, or decide that you want to disable future temporary upgrades, you are required to remove the
enablement feature, On/Off CoD Enablement (#9896).
IBM eServer On/Off Capacity on Demand from IBM Global Financing:
Let IBM Global Financing, the world's leading provider of IT financing, assist in your
acquisition of IBM leading-edge technology.
Through Total Solution Financing we provide a single source that allows you to consolidate your
solution including hardware, software, and services. You have a single contract — an innovative
and tailored solution that helps make acquiring your IBM
fast, easy, and affordable.
Through our Total Usage Financing offering we can provide financing for your on demand needs. Match
your investment to your usage with competitive financing for your fixed and variable costs for
On/Off Capacity on Demand.
Connectivity Abounds
What's New with HiperSockets?:
HiperSockets, first introduced in October 2001, employs the Queued Direct Input/Output
(QDIO) architecture for very high speed internal TCP/IP communication (network within the box — an
independent internal LAN) between programs running on z/OS, on z/VM, on Linux for zSeries, or as
guests under z/VM, whether in the same or in different LPARs on the same system. A synchronous data
mover function transfers data at memory access speed with very low latency.
HiperSockets Now Supported by VSE/ESA:
In VSE/ESA V2.7, IBM extended the support for HiperSockets to include the VSE/ESA
environment. TCP/IP for VSE/ESA V1.5 (a new release) provides support for HiperSockets.
Refer to Software Announcement
203-043
, dated February 18, 2003.
HiperSockets VLAN Support in a Linux Environment:
Virtual Local Area Networks (VLANs), IEEE standard 802.1q, is now being offered for HiperSockets
in a Linux for zSeries environment. VLANs can reduce overhead by allowing networks to be organized
for optimum traffic flow; the network is organized by traffic patterns rather than physical
location. This enhancement permits traffic to flow on a VLAN connection both over HiperSockets and
between HiperSockets and an OSA-Express GbE, 1000BASE-T Ethernet, or Fast Ethernet feature.
In April of 2002, this support was offered on z800 and z900 for the OSA-Express Ethernet
features (GbE, Fast Ethernet) in a Linux for zSeries environment.
HiperSockets — Up To 16 Internal LANs and Up To 4096 TCP/IP Stacks:
When HiperSockets was introduced, up to
four internal
Local Area Networks (LANs) could be configured. That number is now being increased to up to
16 internal LANs.
In addition, the number of communication queues is being increased to 4096 communication queues
instead of the current 1024 communication queues. Since each TCP/IP stack requires one
communication queue, this means 4096 TCP/IP stacks are now supported (instead of 1024 TCP/IP
stacks). A HiperSockets channel must be spanned in order to communicate between LPARs in different
LCSS's.
This support is exclusive to z990. There are no prerequisite software requirements.
HiperSockets Broadcast Support for IPv4 Packets — Linux, z/OS, z/VM:
Internet Protocol Version 4 (IPv4) broadcast packets are now supported over HiperSockets
internal LANs. TCP/IP applications that support IPv4 broadcast, such as OMPROUTE when running
Routing Information Protocol Version 1 (RIPv1), can send and receive broadcast packets over
HiperSockets interfaces.
This support is exclusive to z990. Broadcast for IPv4 packets is supported by Linux for zSeries.
Support is planned to be available in z/OS V1.5. Support is also offered in z/VM V4.4. Refer to
Software Announcement
203-131
, dated May 13, 2003. Refer to Software Announcement
203-128
, dated May 13, 2003.
TCP/IP broadcast support for the OSA-Express features was announced in April 2002, and was made
available in the following environments: z/VM V4.3 (May 2002), z/OS V1.4
(September 2002), and Linux for zSeries.
For Linux for zSeries support, refer to the following Web site for further information:
HiperSockets Spanned Channels:
As previously mentioned, HiperSockets (IQD CHPID type) can be
configured to multiple Channel SubSystems and transparently shared by any
or all configured LPARs without regard to the LCSS to which the LPAR is
configured. Refer to the
Planned Availability
and
Software Requirements
sections of this announcement.
HiperSockets — Transparently Bridging Traffic in a Linux Environment:
New Linux Layer 2 Switch (Linux L2S) support can simplify network addressing between HiperSockets
and OSA-Express. You can now seamlessly integrate HiperSockets-connected operating systems into
external networks, without requiring intervening network routing overhead, thus increasing
performance and simplifying configuration.
With the new Linux Layer 2 Switch support, you can now configure a special-purpose Linux operating
system instance, which can transparently bridge traffic between a HiperSockets internal Local Area
Network (LAN) and an external OSA-Express network attachment, similar to a real Layer 2 switch which
bridges between different network segments. This support can make the internal HiperSockets network
address connection appear as if it were directly connected to the external network.
This Linux L2S support facilitates consolidation of physically dispersed servers using Linux on
zSeries, without requiring additional network configuration changes. In addition, the Linux L2S
support can increase scalability, allowing a single zSeries system to now host increasing numbers of
operating system images and their associated network connectivity by consolidating the multiple
systems, including their HiperSockets IP address space requirements, into a single IP network
domain.
The Linux L2S support is performed using the next-hop-IP-address in the Queued Direct Input/Output
(QDIO) header, instead of using a Media Access Control (MAC) address. Therefore, Virtual Local Area
Networks (VLANs) in a switched Ethernet fabric are not supported by this Linux Layer 2 Switch
support. Operating system images using only HiperSockets to communicate among each other with no
external network connection see no difference, and the HiperSockets support and the networking
characteristics are unchanged.
Refer to the
Planned Availability
and
Software Requirements
sections for Linux for zSeries support.
Doubling ESCON Connectivity to Facilitate Server Consolidation:
With the introduction of two Logical Channel SubSystems (LCSS's), it is now possible to define
up to a maximum of
512 ESCON channels
on your z990, up to a maximum of 35 features. The maximum number of configurable channels is 256
per CSS and per operating system image.
The high-density, 16-port Enterprise Systems CONnection (ESCON) feature can have 15 active ports.
One is always reserved as a spare in the event of a failure of one of the other ports. When four
ports are ordered, two 16-port ESCON features are installed and two ports are activated on each
feature. After the first pair, ESCON features are installed in increments of one. ESCON channels
continue to be ordered in increments of four.
Fiber Quick Connect, an Integrated Quick Connect for ESCON Channels:
Fiber Quick Connect (FQC), coupled with the IBM Fiber Transport System (FTS) from IBM Network
Integration and Deployment Services for enterprise fiber cabling, delivers a solution designed to
reduce the amount of time required for on-site installation and setup of cabling, to minimize
disruptions, and to isolate the activity away from the active system as much as possible. FQC
facilitates adds, moves, and changes of ESCON multimode fiber optic cables in the data center and
can reduce fiber connection time by up to 80%.
The Fiber Quick Connect feature, for factory installation of IBM FTS direct-attach fiber harnesses,
may be the right answer for your installation, to minimize the number of jumper cables exiting the
server and residing under the floor. FQC supports all installed ESCON features in all installed I/O
cages.
Cabling is a customer responsibility. Refer to the IBM Networking Services section of this
announcement for information on fiber optic cabling available from IBM Global Services. Refer to
the
IBM Sales Manual
for more information on Fiber Quick Connect.
Parallel Channels No Longer Supported:
Parallel channels features are not supported on z990. They are not offered as a new build
option, and are not offered on an upgrade from z900. This satisfies the Statement of General
Direction in the Hardware Announcement dated October 3, 2000.
Customers who are still using parallel-attached devices can obtain a parallel channel converter
product:
New for FICON Express
A 25% Increase in the Number of FICON Express Channels:
z990 now supports up to a maximum of 60 FICON Express features, a total
of
120 FICON channels.
This is a 25% growth over what was available on z900. FIbre CONnection
(FICON) Express continues to be a high-performance I/O channel optimized
for efficient, high-speed communication. The FICON Express SX and LX
features are capable of supporting an average of 3600 start I/Os per
second on a channel that is 50% utilized.
The FICON Express features offer multiple I/O connectivity options:
CHPID
Function Type Connectivity
FICON Bridge FCV ESCON devices
FICON Native FC FICON devices
FICON CTC FC Server-to-Server
or LPAR-to-LPAR
Fibre Channel FCP SCSI devices/Linux
Protocol
The FICON Express features support a 1 gigabit per second (Gbps) or 2
Gbps link data rate, auto-negotiating the speed, point-to-point, without
application or end-user intervention. Two features continue to be
available — FICON Express LX (long wavelength) for use with 9 micron
single mode fiber optic cabling, and FICON Express SX (short wavelength)
for use with 50 or 62.5 micron multimode fiber optic cabling. Refer to
the
Sales Manual
for a more complete description of these features.
The most common use of these features continues to be FICON Native with
large infrastructures often feeding terabytes of storage.
Fiber optic cabling is a customer responsibility. Refer to the
Fiber Optic Cabling Services from IBM Networking Services
section of this announcement for information on fiber optic cabling services.
Cascaded Director Support:
On January 31, 2003, IBM made generally available the FICON Cascaded Director function.
FICON Cascading enables two FICON Directors to be interconnected, or cascaded. FICON Cascading,
used for cross-site connectivity, can minimize the number of channels and cross-site connections,
reducing implementation costs for disaster recovery applications such as Geographically Dispersed
Parallel Sysplex™ (GDPS™) and Remote Copy. This support is applicable to z800, z900, and z990.
Support is offered in z/OS V1.3 with PTFs or later releases. Support is also available in
z/VM V4.4.
For more information on Cascaded Directors, consult the I/O Connectivity Web site at:
Maximum Unrepeated Distance between Cascaded Directors:
For information regarding the maximum unrepeated distance
supported between cascaded directors, refer to the following Web sites:
FCP for Attachment to SCSI Devices in Linux Environments:
The Fibre Channel Protocol (FCP) capability, supporting attachment to
Small Computer System Interface (SCSI) devices in Linux for zSeries
environments, was made available for z800 and z900
February 20, 2003.
Refer to the
Planned Availability
section of this announcement for availability of FCP on z990.
Refer to the connectivity Web site listed at the end of this section for updated information on
supported devices.
For Linux operating as a guest under VM, z/VM V4.3 or later is required. For Linux for zSeries
support, refer to the Web site listed at the end of this section.
FICON/FCP Intermix:
On March 31, 2003, zSeries announced support for intermixing FIbre CONnection (FICON)
and the Fibre Channel Protocol (FCP) within the same physical FICON Director. FICON/FCP intermix is
supported on the McDATA Intrepid 6000 Series Directors and INRANGE FC/9000 Directors. Now a single
director may be shared, on a port-by-port basis, between FICON-capable servers/devices and
FCP-capable servers/devices.
FICON/FCP Intermix facilitates asset utilization. If you have chosen to implement small director
footprints which can scale up over time, this may be of assistance as you pursue that strategy.
The Control Unit Port (CUP) continues to be used to communicate with operating systems, including
Linux.
For additional details, refer to FICON/FCP White Papers which are available at the following Web
sites.
For Linux for zSeries support of FCP, refer to the following Web site for
further information:
For more information, consult IBM's I/O Connectivity Web site:
Update on the OSA-Express Family of LAN Adapters
Now 48 Connections To Your Network:
With the introduction of z990, its increased processing capacity, and
the availability of multiple Channel SubSystems, the Open Systems Adapter
family of Local Area Network (LAN) adapters is also expanding by offering
up to a maximum of 24 features per system, versus the maximum of up to 12
features per system on prior generations. You now have 48 ports of LAN
connectivity. You can choose any combination of OSA-Express features:
the new OSA-Express Gigabit Ethernet LX (#1364), the new OSA-Express
Gigabit Ethernet SX (#1365), the new OSA-Express 1000BASE-T Ethernet
(#1366), and the current OSA-Express Token Ring (#2367). You can also
carry forward on an upgrade from z900, OSA-Express Gigabit Ethernet LX
(#2364), OSA-Express Gigabit Ethernet SX (#2365), OSA-Express Fast
Ethernet (#2366), and OSA-Express Token Ring (#2367).
New OSA-Express Gigabit Ethernet Features:
A new generation of Open Systems Adapter-Express (OSA-Express) Gigabit
Ethernet (GbE) features is being offered with z990, and new feature
numbers are being introduced. These new features are offered on new
builds and replace the current OSA-Express GbE features (#2364, #2365)
which can also be brought forward to z990 on an upgrade from z900.
The new OSA-Express Gigabit Ethernet SX (short wavelength) and
OSA-Express Gigabit Ethernet LX (long wavelength) features support
Checksum Offload,
and have a new connector type, LC Duplex, replacing the current SC Duplex
connector. This conforms to the fiber optic connectors currently in use
for ISC-3 and the FICON Express features which began shipping
October 31, 2001.
The new OSA-Express GbE features continue to be dual-port features
occupying a single I/O slot, and utilize one CHPID per port, two CHPIDs
per feature. The OSA-Express GbE features continue to support Queued
Direct Input/Output (QDIO) mode only, full-duplex operation, and jumbo
frames.
The current OSA-Express GbE features are supported on z990 on an upgrade
from z900, OSA-Express Gigabit Ethernet LX (#2364) and OSA-Express
Gigabit Ethernet SX (#2365). The connector remains the same, SC Duplex.
These new features are exclusive to z990. There are no unique software
dependencies for the new OSA-Express Gigabit Ethernet features, with the
exception of the Checksum Offload support. Refer to the Checksum Offload
text in this announcement for more details. Checksum offload support is
planned to be available in z/OS V1.5. Refer to the Web site for Linux
for zSeries support.
Refer to the standards section of this announcement for conformance
information as well as additional information on the features.
New Ethernet Feature Being Introduced — 1000BASE-T Ethernet:
IBM is introducing a new copper Ethernet feature with z990. This new
feature is offered on new builds and replaces the current OSA-Express
Fast Ethernet (#2366), which can also be brought forward to z990 on an
upgrade from z900.
This new feature, 1000BASE-T Ethernet, is capable of operating at 10,
100, or 1000 Mbps (1 Gbps) using the same Category 5 Unshielded
Twisted Pair (UTP) cabling infrastructure that is utilized for Fast
Ethernet. The new gigabit over copper adapter allows a migration to
gigabit speeds wherever there is a copper cabling infrastructure instead
of a fiber optic cabling infrastructure. The new 1000BASE-T Ethernet
feature supports:
-
Auto-negotiation
-
A QDIO and a non-QDIO environment allowing you to make the most of your
TCP/IP and SNA/APPN®/HPR environments at up to gigabit speeds
-
Checksum Offload when in QDIO mode
When configured at 1 Gbps, the 1000BASE-T Ethernet feature has the
following attributes:
-
Operates in QDIO mode or non-QDIO mode
-
Carries SNA (non-QDIO mode) and TCP/IP packets (QDIO or non-QDIO mode)
-
Operates in full-duplex mode only
-
Supports jumbo frames in QDIO mode
There are no unique software dependencies for the new 1000BASE-T Ethernet
feature, with the exception of the Checksum Offload support. Checksum
Offload support is planned to be available in z/OS V1.5. Refer to the
Web site for Linux for zSeries support.
Refer to the standards section of this announcement for conformance
information as well as additional information on the feature.
Checksum Offload for IPv4 Packets when in QDIO Mode — Linux and z/OS:
A new function,
Checksum Offload,
offered for the new OSA-Express GbE and 1000BASE-T Ethernet features, is
being introduced for the Linux for zSeries and z/OS environments.
Checksum Offload provides the capability of calculating the Transmission
Control Protocol (TCP), User Datagram Protocol (UDP), and Internet
Protocol (IP) header checksums. Checksum verifies the correctness of
files. By moving the checksum calculations to a Gigabit or 1000BASE-T
Ethernet feature, host CPU cycles are reduced and performance is
improved.
When checksum is offloaded, the OSA-Express feature performs the checksum
calculations for Internet Protocol Version 4 (IPv4) packets. This
function applies to packets which actually go onto the Local Area Network
(LAN) or come in from the LAN. When multiple IP stacks share an
OSA-Express, and an IP stack sends a packet to a next hop address owned
by another IP stack sharing the OSA-Express, OSA-Express sends the IP
packet directly to the other IP stack without placing it out on the LAN.
Checksum Offload does not apply to such IP packets. This function does
not apply to IPv6 packets. TCP/IP will continue to perform all checksum
processing for IPv6 packets. This function also does not apply to ICMP
checksum processing. TCP/IP will continue to perform processing for ICMP
checksum.
Checksum Offload is supported by the new OSA-Express GbE features (#1364,
#1365) and the new 1000BASE-T Ethernet feature (#1366). This is
applicable to the QDIO mode only.
Checksum offload support is planned to be available in z/OS V1.5. Refer
to Software Announcement
203-131
, dated May 13, 2003.
For Linux for zSeries support, refer to the following Web site for
further information:
Full VLAN Support for z/OS to Increase Traffic Flow when in QDIO Mode:
z/OS Communications Server (CS) now supports Virtual Local Area Network
Identifications (VLAN IDs). Support is offered for one global VLAN (ID)
per Internet Protocol (IP) version per TCP/IP stack:
-
One Global VLAN (ID) for IPv4 and
-
One Global VLAN (ID) for IPv6
Previously CS supported VLAN priority tagging. CS will support both
priority tagging and VLAN IDs.
VLAN support conforms to the IEEE 802.1q standard which defines a VLAN as
a subset of the active topology of the LAN. VLANs ease the
administration of logical groups of users so that they can communicate as
if they were on the same LAN. VLANs increase traffic flow and reduce
overhead by allowing the organization of networks by traffic patterns
rather than by physical location.
Full VLAN support is offered in z/OS V1.5 Communications Server. Full
VLAN support is available on the following OSA-Express features:
1000BASE-T Ethernet (#1366), Fast Ethernet (#2366), and Gigabit Ethernet
(#2364, #2365, #1364, #1365) when in QDIO mode.
Refer to Software Announcement
203-131
, dated May 13, 2003.
Virtual Local Area Network (VLAN) Support in z/VM:
z/VM V4.4 now exploits the VLAN technology and conforms to the IEEE
802.1q standard. TCP/IP for z/VM supports Virtual Local Area Network
Identifications (VLAN IDs). Support is offered for one global VLAN ID
for IPv4. The z/VM TCP/IP stack supports one VLAN ID per OSA-Express
port. Each port can be configured with a different VLAN ID.
Support is offered for the following OSA-Express features: 1000BASE-T
Ethernet (#1366), Fast Ethernet (#2366), and Gigabit Ethernet (#2364,
#2365, #1364, #1365) when in QDIO mode.
Refer to Software Announcement
203-128
, dated May 13, 2003.
VLAN support in a Linux for zSeries environment was delivered for the
OSA-Express Fast Ethernet and Gigabit Ethernet features in QDIO mode in
April 2002 for z800 and z900.
Intrusion Detection Services (IDS) Enhancements when in QDIO Mode:
Introduced in z/OS V1.2 Communications Server, Intrusion Detection
Services can help detect attacks on the TCP/IP stack that would
potentially harm its ability to function normally and cause a misuse of
system resources.
In z/OS V1.5 Communications Server, interface flood detection (Denial of
Service — DoS) will be added to the current IDS.
These enhancements are exclusive to z990 and are supported by all of the
OSA-Express features supported on z990 when in QDIO mode (OSD CHPID type)
— one of the IP Assist functions.
Refer to Software Announcement
203-131
, dated May 13, 2003.
Now 160 TCP/IP Stacks per OSA-Express Port when in QDIO Mode:
TCP/IP connectivity is now increased with the capability to allow up to
a maximum of 160 IP stacks per OSA-Express port and 480 devices. There
is up to a maximum of 84 IP stacks and 255 devices per LPAR. There
continue to be four QDIO priorities. Previously, the OSA-Express
features, when configured for QDIO, had been capable of supporting up to
80 IP stacks and 240 devices.
This support is applicable to all of the OSA-Express features available
on z990, either as a new build z990 or on an upgrade from z900 when using
QDIO mode (OSD CHPID type). This support is offered in the z990 Licensed
Internal Code (LIC). There are no prerequisite software requirements.
OSA/SF Java™ GUI for Client Operating System Independence:
A new version of the Open Systems Adapter Support Facility (OSA/SF) is
being introduced and includes a Java-based Graphical User Interface (GUI)
in support of the client application. The Java GUI is independent of any
operating system/server (transparent to operating system), and is
expected to operate wherever the Java 1.4 runtimes are available.
Interoperability testing has been performed for Windows® 2000, Windows
XP, and Linux for zSeries. In the past, workstation support was
downloaded to a client supporting Windows NT®, Windows 95, or OS/2®.
Use of the GUI is optional; a REXX command interface is also included
with OSA/SF. OSA/SF is not required to set up the OSA-Express features
in QDIO mode (OSD CHPID type).
OSA/SF has been, and continues to be, integrated in z/OS, OS/390®,
z/VM, and VSE/ESA and runs as a host application. For OSA/SF, Java GUI
communication is supported via TCP/IP only. In the past, communication
was supported via EHLLAPI (3270), APPC, and TCP/IP.
This new integrated version of OSA/SF is a complete replacement for the
currently integrated versions in z/OS, OS/390, and VSE/ESA. This new
version of OSA/SF is not being offered as a separately orderable program
product.
The Open Systems Adapter Support Facility (OSA/SF) is used primarily to:
-
Manage all OSA ports
-
Configure all OSA non-QDIO ports
-
Configure ATM LANE ports on G3 — G6 Servers, z800, and z900
-
Configure local MAC
This deliverable is a complete replacement for OSA/SF V2.1
(5655-B57). A separately orderable program product is no longer
available. This new integrated version is applicable to all in service
releases of z/OS, OS/390, z/VM, and VSE/ESA.
In the z/OS environment, delivery will be via the z/OS V1.4 z990
Compatibility Support feature (for release z/OS V1.4) and z990
Compatibility for Selected Releases Web deliverable (for releases OS/390
V2.10, z/OS V1.2, and z/OS V1.3). In the z/OS environment, the new
integrated version of OSA/SF can coexist with OSA/SF V2.1 and does not
overlay it.
The new version of OSA/SF is integrated in z/VM V4.4 and replaces V2.1.
In currently supported versions/releases of z/VM, and VSE/ESA, the new
version of will be delivered as a PTF and will overlay OSA/SF V2.1.
This support is applicable to all OSA-Express and OSA-2 features on all
supported servers.
Note:
The OSA-2 features are not available on z990.
Refer to Software Announcement
203-131
, dated May 13, 2003.
Refer to Software Announcement
203-128
, dated May 13, 2003.
OSA-Express Direct SNMP Subagent Support for dot3StatsTable:
Now the OSA-Express Direct Simple Network Management Protocol (SNMP)
subagent support is offered for Ethernet data for the dot3StatsTable from
the SNMP EtherLike-MIB module in RFC 2665, which provides statistics for
Ethernet interfaces. These statistics can assist in the analysis of
network traffic congestion.
This support is applicable to all of the OSA-Express Ethernet features
supported on z990, whether GbE, 1000BASE-T Ethernet, or Fast Ethernet
when in QDIO mode (OSD CHPID type). Support for the dot3StatsTable is
also available for the Ethernet features (GbE, Fast Ethernet) offered on
z800 and z900 when in QDIO mode. OSA/SF is no longer a prerequisite for
this function. This support is offered for z990, z900, and z800. For
z900 and z800 support, refer to the PSP bucket for OSA. This support is
offered for z/OS V1.4 or later. Refer to the appropriate PSP bucket for
PTFs.
In April 2002, IBM announced Direct SNMP subagent support. Prior to
the OSA-Express Direct SNMP subagent, OSA-Express management data was
only available from the z/OS Communications Server TCP/IP subagent which
required the Open Systems Adapter Support Facility (OSA/SF) to obtain the
data. With the release of the OSA-Express Direct SNMP subagent, OSA/SF
was no longer required to manage SNMP data for OSA-Express. At that time
support was offered for the "SNMP get" command in the z/OS V1.4
environment. Support for the "SNMP get" command was also made available
in the Linux for zSeries environment.
OSA-Express Direct SNMP MIB Module:
An updated Simple Network Management Protocol (SNMP) Management
Information Base (MIB), which supports all of the OSA-Express features
that can exist in a zSeries 990, is available on Resource Link. The
updated SNMP MIB includes support for:
-
New Gigabit Ethernet features
-
New 1000BASE-T Ethernet feature
-
Logical Channel SubSystems (LCSS's)
-
Updated performance table with more detailed information
The MIB can be found on Resource Link.
OSA-Express ATM No Longer Supported:
The OSA-Express Asynchronous Transfer Mode (ATM) features are not
supported on z990. They are not offered as a new build option and are
not offered on an upgrade from z900. This satisfies the Statement of
General Direction in the Hardware Announcement dated
April 30, 2002.
If ATM connectivity is still desired, a multiprotocol switch or router
with the appropriate network interface (for example, 1000BASE-T Ethernet,
Gigabit Ethernet) can be used to provide connectivity between the z990
and an ATM network.
OSA-2 FDDI No Longer Supported:
The OSA-2 Fiber Distributed Data Interface (FDDI) feature is not
supported on z990. It is not offered as a new build option and is not
offered on an upgrade from z900. This satisfies the Statement of General
Direction in the Hardware Announcement dated October 4, 2001.
If FDDI connectivity is still desired, a multiprotocol switch or router
with the appropriate network interface (for example, 1000BASE-T Ethernet,
Gigabit Ethernet) can be used to provide connectivity between the z990
and a FDDI LAN.
Further Integration of Cryptographic Support with z990
New Cryptographic Function on Every Processor Unit (PU):
To achieve the required throughput and implement new functions while
maintaining balanced usage of system resources, integrated hardware is
key. zSeries 990 introduces the Message Security Assist Architecture
along with the new CP Assist for Cryptographic Function (CPACF),
delivering cryptographic support on every Processor Unit (PU) with DES
and TDES data encryption/decryption and SHA-1 hashing. This offers
balanced use of system resources and is designed to provide unprecedented
scalability — a z990 can have from one to 32 PUs, depending upon model
— and data rates at 2X or more faster than the CMOS Cryptographic
Coprocessor Facility (CCF). Since these cryptographic functions are
implemented in each and every PU, the association of cryptographic
functions to specific PUs, as was done with previous generations of
zSeries, is eliminated.
The DES and TDES functions use clear key values. The SHA-1 hash function
is shipped enabled. However, DES and TDES functions require enablement
of the CPACF function (#3863) for export control.
For IBM and customer written programs the CPACF for DES, TDES, and SHA-1
functions can be invoked by five new problem state instructions as
defined by an extension to the zSeries architecture.
Support is also available via the Integrated Cryptographic Service
Facility (ICSF) via the z/OS V1.4 z990 Compatibility Support feature for
z/OS V1.4, z990 Cryptographic CP Assist Support Web deliverable for z/OS
V1.3, and z990 Cryptographic Support Web deliverable for z/OS V1.2 and
OS/390 V2.10.
Secure Encrypted Transactions with Higher Performance — New PCIXCC:
The PCIX Cryptographic Coprocessor (PCIXCC) (#0868) is a replacement
for the PCI Cryptographic Coprocessor (PCICC) (#0861) and the CMOS
Cryptographic Coprocessor Facility that were offered on z900. All of the
equivalent PCICC functions that are implemented are designed to offer
higher performance. In addition, the functions on the CMOS Cryptographic
Coprocessor Facility used by known applications have also been
implemented in the PCIXCC feature.
The PCIXCC feature supports:
-
Secure cryptographic functions
-
Use of secure encrypted key values
-
User-Defined Extensions
The installation of the CPACF feature (#3863) is required to enable the
use of the PCIXCC feature. Support is available via the Integrated
Cryptographic Service Facility (ICSF) and the z990 Cryptographic Support
Web deliverable for z/OS V1.4, z/OS V1.3, z/OS V1.2, and OS/390 V2.10.
Refer to the
Software Requirements
section for availability dates. Refer to the Supplemental section,
Limitations
section of this announcement for more information.
User-Defined Extensions to Support Unique Requirements:
User-Defined Extensions to the Common Cryptographic Architecture (CCA) support program that
executes within the PCIX Cryptographic Coprocessor will be supported via an IBM Service Offering.
For unique customer applications, the PCIX Cryptographic Coprocessor will support the loading of
customized cryptographic functions on z990. Support is available via ICSF and the z990
Cryptographic Support. Refer to the
Software Requirements
section for availability dates.
Under a special contract with IBM, PCIX Cryptographic Coprocessor customers will gain the
flexibility to define and load custom cryptographic functions themselves. This service offering can
be requested via the IBM "Cryptocards" Web site by selecting the "Custom Programming" option.
The Web site will direct the customer's request to an IBM Global Services
(IGS) location appropriate for the customer's geographic location. A
special contract will be negotiated between IGS and the customer,
covering development of the UDX by IGS per the customer's specifications
as well as an agreed-upon level of the UDX. The contract may also
include e-mail/telephone/on-site support if so negotiated (IGS may
subcontract any piece of this work to a third-party vendor).
PCICA — Continued Cryptographic Support for the e-business Environment:
The PCI Cryptographic Accelerator (PCICA) feature, supported on z900,
is available on z990 and may be carried forward on upgrades from z900 to
z990. This hardware-based cryptographic solution continues to address
the high Secure Sockets Layer (SSL) performance needs of on demand
businesses. The SSL and Transport Layer Security (TLS) protocols are
essential and widely used protocols to help support secure e-business
applications. Compute-intensive public key cryptographic processes, as
used by SSL/TLS, can be offloaded from the host to the PCICA feature to
reduce CP usage and to increase system throughput. Each PCICA feature
can support up to 30 LPARs (up to 15 LPARs per PCI Cryptographic
Accelerator card — there are two cards per PCICA feature).
The new z990 server, with six PCICA features and 16 CPs active, is
designed to provide increased secure Web transaction performance by
supporting greater than 11,000 SSL handshakes per second. To enable the
use of the PCICA feature, the CPACF feature (#3863) must be installed.
With VSE/ESA V2.7 and IBM TCP/IP for VSE/ESA V1.5, support is offered for
the PCICA feature. Refer to the
Limitations
section and
Software Requirements
section of this announcement for further information on the PCICA feature.
Trusted Key Entry (TKE) 4.0 Code Level Workstation:
The TKE 4.0 code level workstation is an optionally priced feature that provides a basic key
management system. The key management system allows an authorized person a method for key
identification, exchange, separation, update, backup, and management. The TKE workstation is a tool
for security administrators to use in setting up and establishing the security policy and placing it
into production. The TKE feature is a combination of workstation hardware and software
networked-connected to z990. The TKE workstation and 4.0 code level are designed to provide a
secure, remote, and flexible method of providing Master Key Entry, and to remotely manage PCIX
Cryptographic Coprocessors.
z990 Cryptographic Migration:
The IBM PCI Cryptographic Accelerator (#0862) is supported on z990, and can be carried forward on
an upgrade from z900. Customers with TKE 3.X installed workstations may carry them forward to
control legacy systems only. An update from TKE 3.0 or TKE 3.1 code level to TKE 4.0 code level is
required to control the z990. To use the TKE function, the PCIX Cryptographic Coprocessor (PCIXCC)
feature, the TKE 4.0 code level, and the CP Assist for Cryptographic Function (CPACF) must be
installed.
Refer to the
Software Requirements
section of this announcement for Web deliverables to support PCICA and TKE and exploit PCIXCC.
Fiber Optic Cabling Services from IBM Networking Services
Fiber optic cabling complexity continues to be a part of the Information Technology (IT)
infrastructure as diverse protocols (ESCON, FICON, FCP, Gigabit Ethernet) and industry-standard
small form factor (SFF) connectors continue to proliferate in the enterprise. If your enterprise
has multiple generations of products, the problem can be exacerbated. Even more critical is the
availability of skills and dedicated personnel tasked to support the short-term as well as the
long-term requirements of the IT infrastructure in addition to supporting the open systems
e-business environment.
To better satisfy the cabling requirements of z800, z900, and z990,
IBM Networking Services
has enhanced their fiber optic cabling services.
IBM Network Integration and Deployment Services for zSeries fiber cabling
provides planning and installation services for individual fiber optic connections.
IBM Network Integration and Deployment Services for enterprise fiber cabling
provides planning for IBM Fiber Transport System trunking components and installation services for
small, medium, and large enterprises.
Refer to Services Announcement
603-012
, dated May 13, 2003.
IBM Networking Services
is continually working to ensure IBM has a comprehensive set of services geared for today and
tomorrow. These services take into consideration the requirements for all of the protocols/media
types supported on zSeries (for example, ESCON, FICON, Coupling Links, OSA), whether the focus is
the data center, Storage Area Network (SAN), Local Area Network (LAN), or end-to-end enterprise.
Fiber optic cabling services from
IBM Networking Services
make the most of your cabling investments. IBM has the skills to facilitate the design and
installation of your cabling with the consulting services so critical when the clock matters.
Knowing that new installations, upgrades, and relocations require flawless execution, comprehensive
solutions are available to reduce the complexity and simplify the process. The
zSeries fiber cabling services
and
enterprise fiber cabling services
have been designed to solve the IT infrastructure dilemmas at the product level and the enterprise
level.
These services can meet the needs of your system configurations be they small, medium, or large.
IBM connectivity experts will help you select the option that is best suited to achieve your
business goals, to simplify the zSeries deployment with effective on-site fiber optic cable planning
and installation support.
Fiber optic cables, cable planning, labeling, and installation are all customer responsibilities for
new z990 installations and upgrades. Fiber optic conversion kits and Mode Conditioning Patch (MCP)
cables are not orderable as features on z990. Installation Planning Representatives (IPRs) and
System Service Representatives (SSRs) will not perform the fiber optic cabling tasks without a
services contract.
Under the
zSeries fiber cabling services
umbrella there are
three
options to provide individual fiber optic cables (jumper cables, conversion kits, MCP cables) for
connecting to z800, z900, or z990.
-
Option 1 — Fiber optic jumper cabling package (available today for z800 and z900). IBM does the
detailed planning. This option includes planning, new cables, installation, and documentation. An
analysis of the zSeries channel configuration, I/O devices, and any existing fiber optic cabling is
required to determine the appropriate fiber optic cables.
-
Option 2 — Fiber optic jumper cable migration and reuse for a zSeries upgrade (new option). This
option includes planning, reuse of existing cables, and documentation. IBM organizes the existing
fiber optic cables based upon the new z990 connection details. Relabeling, rerouting, and
reconnection to the appropriate z990 channels is performed. New cables are not offered as a part of
this option.
-
Option 3 — Fiber optic jumper cables and installation (new option). The customer tells us what
they need. They do the detailed planning. The service includes new cables, installation, and
documentation. Planning and providing the list of required cables are customer responsibilities.
Options 1 and 2 can be combined within one statement of work to provide complete upgrade coverage.
Under the
enterprise fiber cabling services
umbrella there are
two
options to provide fiber optic trunk cables (fiber optic trunk cables, fiber harnesses, panel-mount
boxes) for connecting to the z800, z900, or z990.
-
Option 1 — zSeries fiber optic trunk cabling package (new option). IBM reduces the cable clutter
under the floor. An analysis of the zSeries (z800, z900, z990) channel configuration and any
existing fiber optic cabling is performed to determine the required FTS fiber optic trunking
commodities (trunk cables, harnesses, panel-mount boxes). This option includes zSeries planning,
FTS fiber optic trunking commodities, installation, and documentation. This option does not include
enterprise-level planning.
-
Option 2 — Enterprise fiber cabling services. IBM organizes the entire enterprise. This option
includes enterprise planning, new cables, fiber optic trunking commodities, installation, and
documentation. This is the most comprehensive set of services.
A complete evaluation of the enterprise ensures your long-term planning goals can be achieved.
Enterprise fiber cabling services employ the use of a proven modular cabling system, the Fiber
Transport System (FTS), which includes trunk cables, zone cabinets, and panels for your servers,
directors, and storage devices.
FTS supports Fiber Quick Connect (FQC), a fiber harness integrated in the zSeries frame for "quick"
connect, which is offered as a feature on zSeries for connection to ESCON channels.
Whether you choose a packaged service or a custom service, high-quality components are used to
facilitate moves, adds, and changes in the enterprise to prevent extending your maintenance
"window."
Refer to the
Planning Information
and
Cabling Responsibilities
sections of this announcement for more information or contact IBM Global Services for details.
Parallel Sysplex Marches On
Enhancing the Scalability of Parallel Sysplex:
All z990 servers ship with Coupling Facility Control Code (CFCC) Level 12, which includes
support for:
-
64-bit addressing to support larger structure sizes and to eliminate the 2 gigabyte (GB) "control
store" line in the Coupling Facility (CF). With this support, the distinction between "control
store" and "non-control store" (data storage) in the Coupling Facility is eliminated, and large
central storage can be used for all Coupling Facility control and data objects.
-
Up to 48 internal tasks for improved multiprocessing of CF requests.
-
System-Managed Coupling Facility Structure Duplexing.
-
Message Time Ordering.
As server and Coupling Facility link technologies have improved over the years, the synchronization
tolerance between systems in a Parallel Sysplex has become more rigorous. To help ensure that any
exchanges of time-stamped information between systems in a Parallel Sysplex involving the Coupling
Facility observe the correct time ordering, time stamps are now included in the message-transfer
protocol between the systems and the Coupling Facility. Therefore, when a Coupling Facility is
configured as an ICF on any z990, the Coupling Facility requires connectivity to the same Sysplex
Timer® that the other systems in its Parallel Sysplex are using for time synchronization. If the
ICF is on the same server as a member of its Parallel Sysplex, no additional Sysplex Timer
connectivity is required, since the server already has connectivity to the Sysplex Timer. However,
when an ICF is configured on a z990 which does not host any systems in the same Parallel Sysplex, it
is necessary to attach the server to the Sysplex Timer.
Enhancing the Availability of Parallel Sysplex:
z990 fully supports System-Managed CF Structure Duplexing. This is a set of architectural
extensions to Parallel Sysplex in support of duplexing of Coupling Facility structures for high
availability. All three structure types — cache structures, list structures, and locking
structures — can be duplexed using this architecture.
The IBM technical paper
System Managed CF Structure Duplexing
(GM13-0103) includes information about:
-
The cost/benefit trade-offs in duplexing
-
Determining which structures should be duplexed in a specific Parallel Sysplex
This paper is available at:
Support for these extensions on z990 is on the same schedule as the
entire zSeries family of servers and prerequisites CFCC Level 12
with current maintenance including the CFCC compatibility code as well as
z/OS V1.2 and later with the service as defined in the CFDUPLEXING
PSP bucket.
System-Managed CF Structure Duplexing is now available exclusively
through a Readiness Review process which was announced on
April 8, 2003. Refer to Hardware Announcement
103-089
, dated April 8, 2003.
System-Managed CF Structure Duplexing also requires the appropriate level
for the exploiter support. Installing these releases and enabling this
new function provides the necessary base for highly available Coupling
Facility structure data through the redundancy of duplexing.
Additionally, it enhances Parallel Sysplex ease of use by reducing the
complexity of CF structure recovery, and may also enable some
installations to eliminate the requirement for standalone CFs in their
Parallel Sysplex configuration.
z990 Coupling Connectivity for Parallel Sysplex
ICs, ICBs, and ISC-3 on z990:
Like its predecessor, z990 supports Internal Coupling Channels (ICs),
Integrated Cluster Bus (ICB), and InterSystem Channel-3 (ISC-3) for
passing information back and forth in a Parallel Sysplex environment.
These technologies are all members of the family of coupling connectivity
options available on z990.
ICs are for internal communication between Coupling Facilities defined in
LPARs and z/OS images on the same server. ICBs are used for
server-to-server communication over short distances. The newest ICB
member, ICB-4, supports a link data rate of 2 gigabytes per second
(GBps), compared to ICB-3 which supports a link data rate of 1 GBps.
ICBs carry traffic over 10 meter (33 feet) copper cables, of which 3
meters is reserved for intraserver connection.
The InterSystem Channel-3 (ISC-3) is another member of the family of
Coupling Link options available on z990. ISC-3s support a link data rate
of 2 gigabits per second (Gbps) and carry traffic over 9 micron single
mode fiber optic cables. Refer to the
Sales Manual
for more information on the ISC-3 feature.
z990 supports:
-
Internal Coupling Channels — up to a maximum of 32 ICs
-
Integrated Cluster Bus-2 (ICB-2) to attach G5/G6 Servers to z990 — up
to a maximum of 8 links
-
Integrated Cluster Bus-3 (ICB-3) to attach z800s and z900s to z990 — up
to a maximum of 16 links
-
Integrated Cluster Bus-4 (ICB-4) to attach z990 to z990 — up to a
maximum of 16 links
-
InterSystem Channel-3 (ISC-3) — up to a maximum of 32 links
The maximum number of external Coupling Links combined (ICB-2, ICB-3,
ICB-4, and active ISC-3 links) cannot exceed 32 per system.
There are unique cables for ICB-2, ICB-3, and ICB-4. Refer to the
Sales Manual
for more information on these cables.
Sysplex Timer Model 1 Reaching End of Service:
The Sysplex Timer Model 1, 9037-001, has an announced worldwide
end of service (EOS) date of December 31, 2003. The Sysplex
Timer Model 1, a critical component of any multi-CEC sysplex or
Parallel Sysplex, will no longer be supported by IBM after
December 31, 2003. IBM recommends that you begin the planning
to replace all Sysplex Timer Model 1s with Sysplex Timer
Model 2s.
The Sysplex Timer Model 2 is a full replacement box for the Sysplex
Timer Model 1. There is no upgrade path from the Sysplex Timer
Model 1 to the Sysplex Timer Model 2. An International
Technical Support Organization (ITSO) Redpaper, Migration Planning for
the 9037 Model 2 Sysplex Timer, REDP3666, is available. Refer to
the following Web site for a link to the Redpaper.
Contact your local IBM support representative for a detailed migration
procedures document.
Refer to Services Announcement
602-015
, Declaration of Plan to Discontinue, dated June 18, 2002.
Geographically Dispersed Parallel Sysplex (GDPS) Enhancements:
GDPS, an industry leading e-business continuity solution, is a
multisite solution that is designed to provide the capability to manage
the remote copy configuration and storage subsystems, automate Parallel
Sysplex operational tasks, and perform failure recovery from a single
point of control, thereby helping to improve application availability.
GDPS supports both the synchronous Peer-to-Peer Remote Copy (PPRC), as
well as the asynchronous Extended Remote Copy (XRC) forms of remote copy.
Depending on the form of remote copy, the solution is referred to as
GDPS/PPRC or GDPS/XRC.
GDPS/PPRC and GDPS/XRC have been enhanced to include new functions.
GDPS/PPRC HyperSwap™ Function:
The GDPS/PPRC HyperSwap function is designed to broaden the continuous
availability attributes of GDPS/PPRC.
Stage 1 of the GDPS/PPRC HyperSwap
function provided the ability to:
-
Transparently switch all primary PPRC disk subsystems with the secondary
PPRC disk subsystems for a planned switch reconfiguration
-
Perform disk configuration maintenance and planned site maintenance
without requiring any applications to be quiesced.
Stage 1 of GDPS/PPRC HyperSwap function became generally available in
December 2002.
Stage 2 of GDPS/PPRC HyperSwap
function will contain additional function to transparently switch to use
secondary PPRC disk subsystems in the event of unplanned outages of the
primary PPRC disk subsystems or a failure of the site containing the
primary PPRC disk subsystems.
Stage 2 of GDPS/PPRC HyperSwap support allows:
-
Production systems to remain active during a disk subsystem failure.
Disk subsystem failures will no longer constitute a single point of
failure for an entire Parallel Sysplex.
-
Production systems to remain active during a failure of the site
containing the primary PPRC disk subsystems if applications are cloned
and exploiting data sharing across the two sites. Even though the
workload in the second site will need to be restarted, an improvement in
the Recovery Time Objective (RTO) will be accomplished.
Stage 2 of GDPS/PPRC HyperSwap is planned to be generally available
second quarter 2003.
GDPS/PPRC Management for Open Systems Logical Unit Numbers (LUNs):
GDPS/PPRC technology has been extended to manage a heterogeneous
environment of z/OS and open systems data. If installations share their
disk subsystems between the z/OS and open systems platforms, GDPS/PPRC,
running in a z/OS system, can manage the PPRC status of devices that
belong to the other platforms and are not even defined to the z/OS
platform. GDPS/PPRC will also provide data consistency across both z/OS
and open systems data.
GDPS/PPRC management of open systems LUNs is planned to be generally
available second quarter 2003.
Performance Enhancements for GDPS/PPRC and GDPS/XRC Configurations
-
Concurrent activation of Capacity Backup Upgrade (CBU) can now be
performed in parallel across multiple servers, which results in an
improved RTO. This improvement applies to both the GDPS/PPRC and
GDPS/XRC configurations.
-
In a GDPS/XRC configuration, it is often necessary to have multiple
System Data Movers (SDMs) — the number of SDMs is based on many
factors, such as the number of volumes being copied and the I/O rate.
Functions are now capable of being executed in parallel across multiple
SDMs, thus providing improved scalability for a coupled SDM
configuration.
Performance enhancements for GDPS/PPRC and GDPS/XRC became generally
available March 24, 2003.
These GDPS enhancements are applicable to z800, z900, and z990.
For a complete list of other supported hardware platforms and software
prerequisites, refer to the GDPS executive summary white paper, available
at:
SNA Operations Management
With the industry move to TCP/IP networks, Systems Network Architecture (SNA) Operations Management
commands will no longer be supported on z990 servers. These commands were previously used by the
System Automation for OS/390 product as well as NetView®. It is recommended that you now use the
Simple Network Management Protocol (SNMP) Application Programming Interfaces (APIs) for your
automation needs.
-
If you previously used the System Automation for OS/390 product, you must now use Version 2.2
or later. This will allow you to define an automation policy for SNMP APIs rather than a policy for
SNA Operations Management commands.
-
If you directly used the SNA Operations Management commands on NetView, you must now use an SNMP
agent and the SNMP APIs for systems automation management.
For detailed information on the SNMP APIs commands and environment requirements, refer to
zSeries Application Programming Interfaces
(SB10-7030). For more information on the SNA Operations Management command support that is not
offered on z990, refer to
Managing Your Processors
(GC38-0452). Both publications are available on IBM Resource Link.
Reliability, Availability, and Serviceability (RAS)
Continuing to provide high availability and continuous reliable operation, zSeries 990 delivers
significant new features for Reliability, Availability, and Serviceability (RAS).
Fault Tolerant Interconnect Design:
The memory design of the z990 Models B16, C24, and D32 provides a fault tolerant dual
interconnect fabric. The z990 is designed to prevent a single failure within the fabric from
disrupting full memory access.
Enhanced PU Sparing:
Transparent sparing for failed Processor Units (PUs) is enhanced for z990. In addition to PU
sparing on an MCM, the z990 is designed to support sparing to other MCMs in Models B16, C24,
and D32 in the unlikely event that an MCM has no spares.
Concurrent Model Upgrade:
In October 2003, a z990 can be upgraded to a larger z990 model via the concurrent
installation of additional processor units, memory, and I/O links with no disruption to current
operations.
Sparing for Storage Protect Keys:
The robust configuration of the Storage Protect Keys has been enhanced with chip sparing.
Already a triple-array with parity protection and voting, the Key structure for z990 has added
sparing, similar to Main Memory chip sparing, to further enhance the availability of this critical
function.
Failure Containment for the Memory Bus Adapter (MBA):
Memory Bus Adapters are designed to provide the critical link between the z990 Processor Units
and the I/O subsystem. The Model A08 has three MBAs, the B16 has six MBAs, the C24 has nine
MBAs, and the Model D32 has 12 MBAs. In the unlikely event of a catastrophic failure of an MBA
chip, the z990 is designed to isolate the failure of that chip such that the remaining MBA chips
continue to operate. This helps minimize the impact of a failure and allows for scheduling
maintenance.
Enhanced Firmware Simulation:
The z990 process for design, development, and test of Licensed Internal Code (LIC) has been
significantly enhanced with the use of simulation to improve quality and early availability.
Virtually every action/reaction of the code can be tested with the simulated hardware/code of the
rest of the server. The result is to discover and correct design errors much earlier in the
process.
These new features, together with legacy features such as high levels of recovery, concurrent
processor upgrade, concurrent memory upgrade, concurrent I/O upgrade, and concurrent maintenance for
hardware and LIC give the z990 a very impressive RAS structure.
Software Pricing
For updated software pricing, refer to Software Announcement
203-130
, Software Pricing for IBM
zSeries 990 and Enhancements to Variable Workload Charges (VWLC) dated May 13, 2003.
Resource Link
Resource Link is a customized Web-based solution, helping to provide customers everything needed to
plan for, install, and maintain IBM
zSeries and IBM S/390® servers and associated software. Simply register for a
Resource Link user ID and password to access product Planning, Education, and Library services, all
without charge.
-
Subscribe to Resource Link content for e-mail notification any time content changes.
-
Create Personal Folders to organize site information according to your personal interests.
The Resource Link Web address is:
Other Resource Link functions include the use of Forums to collaborate
with fellow product owners, links to hardware and software Fixes,
hardware and software Problem solving databases, Services for servers,
software, and solutions, and Tools designed for specific server and
software planning, installation, and configuration. Resource Link
content includes:
-
Customized Planning Aids — Allow for the planning, arrival, and
installation of
hardware on order. All that is needed is an order number
and the Configuration Control Number (CCN) to access the planning aids
that were customized for a specific order. Included in these planning
aids are the feature numbers that affect the weight, height, power
requirements, cooling, and cables necessary for system installation. The
aids also provide customized system information for processor codes and
descriptions, system power information, frame requirements, cooling
requirements, and hardware management console and support element
information.
-
Customer Initiated Upgrade (CIU) — An IBM Resource Link Web-based
solution that enables customers to set up, order, and download microcode
for vertical upgrades of process and/or memory. CIU simplifies planning
and installation because no IBM service visit is required to perform the
upgrade.
-
CHPID Mapping Tool — Downloadable from Resource Link, this tool allows
you to map hardware I/O ports on order to your IOCP CHPID definitions.
An availability option automatically maps hardware I/O ports to CHPIDs
minimizing single points of failure. Use of this tool is not mandatory,
but recommended for all new z990 hardware builds or for upgrades from a
z900 processor to a z990, as well as for making changes to an already
installed machine after hardware upgrades that change or increase the
number of channels.
-
Fiber Cabling Service — Provides the cabling connectivity necessary for
installing your new z990 processor. Because IBM offers a wide range of
cables and cabling solutions to meet customer requirements for connecting
I/O and networks, this cabling service is available through your IBM
service representative from Resource Link.
-
Education — Use this free, online education to train and refresh skills
at the convenience of your enterprise. Product education includes
planning, introduction, getting started, and "how to" courses. Learn how
to use your IBM server products before your system arrives.
-
Library — View, print, or download documents on IBM hardware and
software products. These documents include product overviews, users
guides, planning information, parts catalogues, and safety documentation.
Also available are product and description documents, which contain
high-level descriptions of specific microcode driver and version levels
for each machine family including EC changes for the associated Hardware
Management Console and Support Element.
-
Machine Information — View or search for machine information from
reports for machines with IBM maintenance agreements. Simply register
for machine information on Resource Link and receive authorization to
view reports based on machine serial number as well as links to MES
content information. The report data is collected from IBM servers on a
weekly basis using the Call Home feature, which transmits critical
machine data and also detects failures or pending failures.
Technical Information
Specified Operating Environment
Physical Specifications
2084 Servers:
The mechanical package for the IBM
zSeries 990 conforms to EIA™ guidelines for frames. Packaging for new build
systems is two frames. Upgrades will be contained within the two frames initially shipped. Note
that height reduction for transport is available. Keep in mind that the height of the z990 is equal
to the height of a reduced z900. When considering placement of a z990, review clearances for doors
to be opened. A new build system at the minimum level will consist of two frames with one I/O cage.
The maximum will consist of two frames with three I/O cages. 60 Amp service is required.
Environmental Specifications — Server Models
New Build New Build
Minimum Maximum
System System
Model A08 with Model D32 with
One I/O Cage Three I/O Cages
Power Requirements
50/60 Hz, kVA 4.5 kW 15.8 kW
Heat Output kBtu/hr 18.36 51.0
Air Flow CFM (at 16(degs)C)
Air Flow m*3/min 580 CFM 1432 CFM
Floor Space
-- Sq meters 2.49 2.49
-- Sq feet 26.78 26.78
Including service clearance
-- Sq meters 5.45 5.45
-- Sq feet 58.69 58.69
Approximate weight
System with IBF Feature
-- Kg 1263 2007
-- Lb 2779 4415
System without IBF Feature
-- Kg 1174 1738
-- Lb 2582 3824
Dimensions
Depth Width Height
System with All Covers
-- Inches 62.1 62.1 76.4
-- Centimeters 157.7 157.7 194.1
System with Covers and Reduction
-- Inches 62.1 62.1 70.3
-- Centimeters 157.7 157.7 178.5
Each Frame with One Side
Cover and without Packaging
-- Inches 46.2 30.7 76.4
-- Centimeters 117.3 78.0 194.1
Each Frame on Casters with One
Side Cover and with Packaging
-- Inches 47.6 32.4 76.9
-- Centimeters 120.9 82.2 195.3
Each Frame with One Side Cover
and with Packaging (ARBO)
-- Inches 51.5 36.5 87.6
-- Centimeters 130.8 92.7 222.5
See your IBM representative for the available width and height reduction transport option for return
of z900 or 9672 systems. There is no feature option available for width. There is height reduction
on z990. However, keep in mind that the z990 is equal in height to a height reduced z900.
Therefore, one may not automatically assume that a reduction is necessary due to their experience
with z900. Check the height and only order as necessary.
Standards
-
PCIXCC feature designed for FIPS 140-2 Level 4 Certification — October 2003.
The OSA-Express family of LAN adapters conforms to the following standards.
-
Ethernet (10BASE-T)
IEEE 802.2 Logical Link Control
IEEE 802.3 CSMA/CD
ISO/IEC 8802-3
DIX Version 2
-
Ethernet (100BASE-TX)
-
Ethernet (1000BASE-T)
IEEE 802.1p
IEEE 802.1q
IEEE 802.3ab
IEEE 802.3ac
IEEE 802.3ad
IEEE 802.3u
IEEE 802.3x
PCI v2.2
-
Ethernet (1000BASE-SX, 1000BASE-LX)
DIX Version 2
IEEE 802.1p
IEEE 802.1q
IEEE 802.3ac
IEEE 802.3ad
IEEE 802.3x
IEEE 802.3z
PCI v2.2
-
Token Ring
IEEE 802.2 Logical Link Control
IEEE 802.5 MAC
ISO/IEC 8802-5
The OSA-Express family of LAN adapters have the following characteristics:
OSA-Express Gigabit Ethernet SX (#1365)
-
Data rate: 1000 Mbps (1 Gbps)
-
Operating mode: Full duplex
-
Channel type: OSD (QDIO)
-
Traffic type: TCP/IP only
-
Frame size: IEEE 802.3: 1492 bytes
-
Frame size: DIX V2: 1500 bytes; for jumbo frame 8992 bytes
-
Connector type: LC Duplex
-
Port count: Two
-
Cable types: Multimode fiber (50 or 62.5 micron)
OSA-Express Gigabit Ethernet LX (#1364)
-
Data rate: 1000 Mbps (1 Gbps)
-
Operating mode: Full duplex
-
Channel type: OSD (QDIO)
-
Traffic type: TCP/IP only
-
Frame size: IEEE 802.3: 1492 bytes
-
Frame size: DIX V2: 1500 bytes; for jumbo frame 8992 bytes
-
Connector type: LC Duplex
-
Port count: Two
-
Cable types: Single mode fiber (9 micron); accommodates reuse of existing multimode fiber (50 or
62.5 micron) when used with a pair of mode conditioning patch (MCP) cables
OSA-Express 1000BASE-T Ethernet (#1366)
-
Data rate: 10 Mbps, 100 Mbps, or 1000 Mbps
-
Operating modes: Auto-negotiate, half duplex, full duplex
-
Channel types: OSD (QDIO) or OSE (non-QDIO)
-
Traffic types: TCP/IP and/or SNA/APPN®/HPR
-
Frame size: IEEE 802.3: 1492 bytes
-
Frame size: DIX V2: 1500 bytes; for jumbo frame 8992 bytes (when operating at 1 Gbps)
-
Connector type: RJ-45
-
Port count: Two
-
Cable type: EIA/TIA Category 5 Unshielded Twisted Pair (UTP) cable with a maximum length of
100 meters (328 feet)
OSA-Express Token Ring (#2367)
-
Data rate: 4 Mbps, 16 Mbps, or 100 Mbps
-
Operating modes: Autosense, 4 Mbps half or full duplex, 16 Mbps half or full duplex, 100 Mbps full
duplex
-
Channel types: OSD (QDIO) or OSE (non-QDIO)
-
Traffic types: TCP/IP and/or SNA/APPN/HPR
-
Frame size: 4 Mbps, up to 4550 bytes; 16/100 Mbps, up to 18200 bytes
-
Connector types: RJ-45 or DB-9 D shell
-
Port count: Two
-
Cable types: EIA/TIA Category 5 Unshielded Twisted Pair (UTP) cable with a maximum length of
100 meters (328 feet) and an RJ-45 connector or a Shielded Twisted Pair (STP) cable with a DB-9 D
Shell connector
Operating Environment:
Refer to the
zSeries Installation Manual — Physical Planning
(IMPP) that is available via IBM's Resource Link.
Hardware Requirements
Specified Operating Environment:
The functions delivered on June 16, 2003, are included with MCM EC# J12312, HMC EC#
J11855, and CFCC Product Release Level 12 Service Level 12.06 EC# J12240.
z990 supports LPAR mode only.
In a Parallel Sysplex environment with z990 servers, all z900- or z800-based Coupling Facilities
must be running CFCC Level 12 with current maintenance including CFCC compatibility code.
G5/G6-based Coupling Facilities must be running CFCC Level 11.
Peripheral Hardware and Device Attachments:
For the appropriate peripheral hardware and device attachments, contact your IBM
representative.
IBM devices previously attached to IBM S/370™ or S/390 systems are supported for attachment to
z990 I/O channels, unless noted. The subject I/O devices must meet ESCON or FICON architecture
requirements to be supported. I/O devices that meet OEMI architecture requirements are supported
only via a parallel channel converter. Prerequisite Engineering Change Levels may be required. For
further details, contact IBM service personnel.
Note:
IBM cannot confirm the accuracy of performance, compatibility, or any other claims related to
non-IBM products. Questions regarding the capabilities of non-IBM products should be addressed to
the suppliers of those products.
TKE Workstation:
Customers with TKE 3.X installed workstations may carry forward the features listed below to
control legacy systems only. An update from TKE 3.0 or TKE 3.1 code level to TKE 4.0 code level is
required to control the z990. Workstations updated to the TKE 4.0 code level may be used to control
9672 Generation 5 and 6, z800, z900, and z990 servers. To use the TKE function on a z990, the PCIX
Cryptographic Coprocessor (PCIXCC) feature, the TKE 4.0 code level, and the CP Assist for
Cryptographic Function (CPACF) must be installed.
-
TKE with Token Ring (#0866)
-
TKE with Ethernet (#0869)
-
TKE with Token Ring (#0876)
-
TKE with Ethernet (#0879)
-
TKE with Token Ring (#0886)
-
TKE with Ethernet (#0889)
Software Requirements
Mode of Operation:
The z990 supports LPAR mode only.
Linux for zSeries Support:
The currently available distributions SuSE SLES 7, SuSE SLES 8, Red Hat 7.1, and Red Hat 7.2
support compatibility and exploitation of 30 LPARs and 2 LCSS's. Support to further exploit z990
functions will be delivered as an Open Source Contribution via:
An initial version is planned to be delivered on June 30, 2003,
to exploit the z990 for users performing their own Linux build. An
updated version is planned to follow on July 25, 2003, and is
intended for our Linux Distribution Partners to package for z990
exploitation support. IBM is working with its distribution partners to
provide these functions in future distribution releases.
z/OS™, z/VM™, VSE/ESA™, and TPF Support and Availability by Date:
The new models of IBM
zSeries 990 are planned to be supported on the
following schedule:
-
May 13, 2003
-
TPF V4R1 is compatible with z990.
-
z/VM compatibility support for V3.1, z/VM V4.2, and z/VM V4.3 and limited
exploitation for two LCSS's when running on a Z990 server. Dynamic I/O
configuration is supported when only LCSS0 is defined. Refer to the z/VM
subset of the 2084DEVICE Preventive Service Planning (PSP) bucket prior
to installing a z990.
June 13, 2003
-
z/OS V1R4 z990 Compatibility Support: This orderable, unpriced, and
optional feature is required to allow z/OS V1R4 to run on a z990 server.
It is required on all systems in a Parallel Sysplex when a z/OS or CF
image in that same Parallel Sysplex is running on a z990 and the LPAR ID
of the operating system or CF image is greater than 15 (x"F"). With
regards to cryptography, it provides support for the PCICA feature and
the CP Assist for Cryptographic Function for z/OS V1R4.
-
z990 Compatibility for Selected Releases: This Web deliverable provides
support to allow OS/390® V2.10, z/OS V1.2, and z/OS 1.3 to run on a
z990. It is required on all OS/390 V2.10, z/OS V1.2, z/OS 1.3, and
z/OS.e V1.3 systems in a Parallel Sysplex when a z/OS or CF image in that
same Parallel Sysplex is running on a z990 and the LPAR ID of the
operating system or CF is greater than 15 (x"F").
-
z990 Cryptographic CP Assist Support for z/OS V1R3: This Web deliverable
provides support in z/OS V1.3 for the PCICA adapter and CP Assist for
Cryptographic Function (#3863).
-
z/OS.e V1.4 z990 Coexistence: This orderable, unpriced, and optional
feature is required on all z/OS.e V1.4 systems in a Parallel Sysplex when
a z/OS or CF image in that same Parallel Sysplex is running on a z990 and
the LPAR ID of the operating system or CF is greater than 15 (x"F").
z/OS.e cannot execute on a z990 server.
June 16, 2003
-
VSE/ESA compatibility is provided on VSE/ESA V2 R5, VSE/ESA V2 R6, and
VSE/ESA V2 R7. For the latest information on compatibility support
requirements, refer to the VSE subset of the 2084DEVICE Preventative
Service Planning (PSP) bucket prior to installing the z990 server.
August 15, 2003
z/VM V4.4 and later releases will provide additional exploitation support
for the following functions:
-
Two Logical Channel SubSystems.
-
Dynamic I/O configuration in only LCSS0.
-
Adapter interruptions for OSA-Express and FCP Channels.
-
Performance Assist for V=V guests in z/VM environment.
October 31, 2003
-
z/OS V1.4 z990 Exploitation Support: This orderable, unpriced, and
optional feature provides exploitation support for two Logical Channel
SubSystems and 30 LPARs.
-
z990 Cryptographic Support: This Web deliverable will provide
exploitation support for z/OS 1.4 for PCIXCC (#0868).
-
z/OS.e V1.4 Coexistence Update: This orderable, unpriced, and optional
feature allows customers to maintain a consistent code base for z/OS and
z/OS.e. It contains the z/OS.e Coexistence feature. Note that z/OS.e
cannot execute on a z990 server.
-
z/VM V4.4 and later releases will provide further exploitation support
for the following functions:
-
30 LPARs.
-
Dynamic I/O Support for LCSS0 and LCSS1.
-
Internal Spanned Channels (ICs and HiperSockets).
November 28, 2003
-
z990 Cryptographic Support: This Web deliverable extends exploitation
support for the CP Assist for Cryptographic Function (#3863), PCICA, and
PCIXCC features to z/OS V1.3 and z/OS V1.2.
December 31, 2003
-
z990 Cryptographic Support: This Web deliverable extends exploitation
support for the CP Assist for Cryptographic Function (#3863), PCICA, and
PCIXCC features to OS/390 V2.10.
Note that z/OS.e cannot execute on a z990 server. z/OS.e releases are
provided to permit coexistence within a Parallel Sysplex and maintain
consistent code levels with z/OS.
Software Support Availability by Operating System and Level:
The new models of IBM
zSeries 990 are planned to be supported by
operating system release on the schedules indicated below for each
release.
z/OS V1.4
-
June 13, 2003
-
z/OS V1.4 z990 Compatibility Support Feature: This orderable, unpriced,
and optional feature is required to allow z/OS V1.4 to run on a z990
server. It is also required on all systems in a Parallel Sysplex when a
z/OS or CF image in that same Parallel Sysplex is running on a z990 and
the LPAR ID of the operating system or CF image is greater than 15
(x"F"). With regards to cryptography, it provides support for the PCICA
feature and the CP Assist for Cryptographic Function for z/OS V1R4.
-
October 31, 2003
-
z/OS V1.4 z990 Exploitation Support: This orderable, unpriced, and
optional feature provides exploitation support for two Logical Channel
SubSystems and 30 LPARs.
-
z990 Cryptographic Support: This Web deliverable will provide
exploitation support for z/OS V1R4 for PCIXCC (#0868).
z/OS V1.3
-
June 13, 2003
-
z990 Compatibility for Selected Releases: This Web deliverable provides
support to allow OS/390 V2.10, z/OS V1.2, and z/OS 1.3 to run on a z990.
It is required on all OS/390 V2.10, z/OS V1.2, z/OS 1.3, and z/OS.e V1.3
systems in a Parallel Sysplex when a z/OS or CF image in that same
Parallel Sysplex is running on a z990 and the LPAR ID of the operating
system or CF is greater than 15 (x"F").
-
z990 Cryptographic Assist Support for z/OS V1R3: This Web deliverable
provides support on z/OS V1.3 for the PCICA adapter and CP Assist for
Cryptographic Function (#3863).
November 28, 2003
-
z990 Cryptographic Support: This Web deliverable will provide
exploitation support for z/OS V1R3 for PCIXCC (#0868).
z/OS V1.2
-
June 13, 2003
-
z990 Compatibility for Selected Releases: This Web deliverable provides
support to allow OS/390 V2.10, z/OS V1.2, and z/OS 1.3 to run on a z990.
It is required on all OS/390 V2.10, z/OS V1.2, z/OS 1.3, and z/OS.e V1.3
systems in a Parallel Sysplex when a z/OS or CF image in that same
Parallel Sysplex is running on a z990 and the LPAR ID of the operating
system or CF is greater than 15 (x"F").
-
November 28, 2003
-
z990 Cryptographic Support: This Web deliverable will extend
exploitation support for the CP Assist for Cryptographic Function
(#3863), PCICA, and PCIXCC features to z/OS V1.2.
OS/390 V2.10
-
June 13, 2003
-
z990 Compatibility for Selected Releases: This Web deliverable provides
support to allow OS/390 V2.10, z/OS V1.2, and z/OS 1.3 to run on a z990.
It is required on all OS/390 V2.10, z/OS V1.2, z/OS 1.3, and z/OS.e V1.3
systems in a Parallel Sysplex when a z/OS or CF image in that same
Parallel Sysplex is running on a z990 and the LPAR ID of the operating
system or CF is greater than 15 (x"F").
-
December 31, 2003
-
z990 Cryptographic Support: This Web deliverable will provide
exploitation support for the CP Assist for Cryptographic Function
(#3863), PCICA, and PCIXCC features on OS/390 V2.10.
z/OS.e V1.3
-
June 13, 2003
-
z990 Compatibility for Selected Releases: This Web deliverable provides
support to allow OS/390 V2.10, z/OS V1.2, and z/OS 1.3 to run on a z990.
It is required on all OS/390 V2.10, z/OS V1.2, z/OS 1.3, and z/OS.e V1.3
systems in a Parallel Sysplex when a z/OS or CF image in that same
Parallel Sysplex is running on a z990 and the LPAR ID of the operating
system or CF is greater than 15 (x"F"). z/OS.e cannot execute on a z990
server.
z/OS.e V1.4
-
June 13, 2003
-
z/OS.e V1.4 z990 Coexistence Feature. This orderable, unpriced, and
optional feature is required on all z/OS.e V1.4 systems in a Parallel
Sysplex when a z/OS or CF image in that same Parallel Sysplex is running
on a z990 and the LPAR ID of the operating system or CF is greater than
15 (x"F").
-
October 31, 2003
-
z/OS.e V1.4 Coexistence Update: This Web deliverable allows customers to
maintain a consistent code base for z/OS and z/OS.e. It contains the
z/OS.e Coexistence feature. Note that z/OS.e cannot execute on a z990
processor.
z/VM
-
May 13, 2003
-
z/VM compatibility support for V3.1, z/VM V4.2, and z/VM V4.3 and limited
exploitation for two LCSS's when running on a z990 server. Dynamic I/O
configuration is supported when only LCSS0 is defined. Refer to the z/VM
subset of the 2084DEVICE Preventive Service Planning (PSP) bucket prior
to installing a z990.
-
August 15, 2003
-
z/VM V4.4 and later releases will provide additional exploitation support
for the following functions:
-
Two Logical Channel SubSystems.
-
Dynamic I/O configuration in only LCSS0.
-
Adapter interruptions for OSA-Express and FCP Channels.
-
Performance Assist for V=V guests in z/VM environment.
October 31, 2003
-
z/VM V4.4 and later releases will provide further exploitation support
for the following functions:
-
30 LPARs.
-
Dynamic I/O Support for LCSS0 and LCSS1.
-
Internal Spanned Channels (ICs and HiperSockets).
VSE/ESA
-
June 16, 2003
-
VSE/ESA compatibility is provided on VSE/ESA V2 R5, VSE/ESA V2 R6, and
VSE/ESA V2 R7. For the latest information on compatibility support
requirements, refer to the VSE subset of the 2084DEVICE PSP bucket prior
to installing the z990 server.
Software Requirements for the FICON Express Features:
The minimum software requirements for the FICON Express features
installed in a z990 are identified.
FICON Express (FCV CHPID type)
attachment to the ESCON Director Model 5 FICON Bridge feature
requires at a minimum:
-
OS/390 V2.10
-
z/VM V3.1, z/VM V4.2
-
VSE/ESA V2.5
-
TPF V4.1 at PUT 16
FICON Express (FC CHPID type)
, including Channel-To-Channel (CTC), requires at a minimum:
-
OS/390 V2.10
-
z/VM V3.1, z/VM V4.2
-
VSE/ESA V2.5
-
Linux Kernel V2.4
-
TPF V4.1 at PUT 16
FICON Express (FCP CHPID type)
, for support of Linux environments, requires at a minimum:
-
z/VM V4.3 (for Linux as a guest under z/VM)
-
z/VM V4.4 for
-
Performance Assist for Adapter Interruptions
-
Performance Assist for V=V Guests
SuSE Linux Enterprise Server 8
For current SuSE security fixes and patches, refer to the SuSE Linux
Maintenance Web site at:
Cascaded FICON Directors (FC CHPID type)
, including CTC with Cascading, requires at a minimum:
-
OS/390 V2.10 (see note below)
-
z/OS V1.3 with PTFs
-
z/OS V1.4 with PTFs
-
z/VM V4.4
-
VSE/ESA V2.5
-
Linux Kernel V2.4
-
TPF V4.1 at PUT 16
Note:
OS/390 V2.10 can support a cascaded configuration. z/OS V1.3 is
required in an LPAR to dynamically define a cascaded director, for
dynamic I/O changes, and to use the enhanced display functions. Refer to
the 9032/9042 PSP buckets.
Software Requirements for Cryptographic Functions:
The following are required as indicated to support cryptographic
function on z990:
-
z990 Cryptographic Assist Support for z/OS V1.3: This Web deliverable
will provide support for the CP Assist for Cryptographic Function (#3863)
and the PCICA (#0862) feature for z/OS V1R3.
-
z/OS V1.4 z990 Compatibility Support: This orderable, unpriced, and
optional feature provides the CP Assist for Cryptographic Function
(#3863) and the PCICA (#0862) feature on z/OS V1R4.
-
z990 Cryptographic Support: This Web deliverable provides support for
the CP Assist for Cryptographic Function (#3863), PCICA (#0862), and
PCIXCC (#0868) features for z/OS V1R4, z/OS V1R3, z/OS V1R2, and OS/390
V2R10. A staged rollout is planned. Review the Software Support
Availability by Operating System and Level of the Software Requirements
for specific plans for each software level.
The following is required for User-Defined Extensions (UDX) support:
-
One or more PCIXCC features
-
An HMC
-
A TKE workstation at 4.0 code level, if the UDX requires access control
point
-
z/OS V1.4
-
z990 Cryptographic Support a Web deliverable feature
The minimum software requirements for the CP Assist for Cryptographic
Function (CPACF), PCI Cryptographic Accelerator (PCICA), and PCIX
Cryptographic Coprocessor (PCIXCC) are as follows:
For CPACF:
-
z/OS V1.3
-
z/OS V1.2, OS/390 V2.10 (available by year end 2003)
-
z/VM V3.1, V4.2, and later
For PCICA:
-
z/OS V1.3
-
z/OS V1.2, OS/390 V2.10 (available by year end 2003)
-
z/VM V4.2 with PTFs (for Linux Guests only)
-
VSE/ESA V2.7 and IBM TCP/IP for VSE/ESA V1.5
-
Linux 2.4.7 for zSeries
For PCIXCC:
-
z/OS V1.4
-
z/OS V1.2, V1.3, OS/390 V2.10 (available by year end 2003)
Software Requirements for the OSA-Express Features:
The minimum software requirements for the OSA-Express features
available on a z990 are identified. Other functions, not identified in
this announcement, may also have unique software prerequisites. For
further details, including the QDIO and non-QDIO modes of operation,
consult the
Open Systems Adapter — Express Customer's Guide and Reference
(SA22-7935) or refer to the Preventive Service Planning (PSP) bucket for
OSA.
-
OS/390 V2.10, all features and modes
-
z/VM V3.1, all features and modes with exception noted below
-
z/VM V4.2 (required for OSA-Express Token Ring in QDIO mode)
-
VSE/ESA V2.6, all features and modes
-
Linux Kernel V2.4 (supports the QDIO mode only)
-
TPF 4.1 at PUT 13 (supports Gigabit Ethernet only)
Checksum Offload for IPv4 Packets
applies to OSA-Express Gigabit Ethernet (#1364, #1365) and 1000BASE-T
Ethernet (#1366) when in QDIO mode (OSD CHPID type) and is supported by
the following operating systems.
-
Linux for zSeries. Refer to the Web site at the end of this section.
-
z/OS V1.5.
VLAN (802.1q) support
applies to OSA-Express 1000BASE-T Ethernet (#1366), Fast Ethernet
(#2366), and Gigabit Ethernet (#2364, #2365, #1364, #1365) when in QDIO
mode (OSD CHPID type) and is supported by the following operating
systems.
-
z/OS V1.5 Communications Server
-
z/VM V4.4
Intrusion Detection Services
applies to all of the OSA-Express features supported on z990 when in
QDIO mode (OSD CHPID type) and is supported by the following operating
system.
-
z/OS V1.5 Communications Server
160 TCP/IP stacks per OSA-Express port
applies to all of the OSA-Express features supported on z990 when in
QDIO mode (OSD CHPID type).
There are no prerequisite software requirements.
New Version of OSA/SF and Java™ GUI:
A new version of the Open Systems Adapter Support Facility (OSA/SF) is
available. This deliverable is a complete replacement of OSA/SF
Version 2 Release 1 (5655-B57), is integrated in the operating
systems, and will be shipped as a PTF where applicable. Refer to the
appropriate PSP bucket. This support applies to all of the OSA-Express
and OSA-2 features and to all servers that support them. The new version
of OSA/SF with the Java GUI is integrated in the operating systems and is
available as follows:
-
z/OS V1.4 z990 Compatibility Support feature.
-
z990 Compatibility for Selected Releases Web deliverables.
-
OS/390 V2.10, z/OS V1.2, z/OS V1.3.
-
New version can coexist with OSA/SF V2.1.
-
Integrated in z/VM V4.4.
-
z/VM V3.1, z/VM V4.2, z/VM V4.3 delivered as a PTF.
-
VSE/ESA V2.6 delivered as a PTF.
OSA-Express Direct SNMP subagent support for dot3StatsTable
applies to all of the OSA-Express Ethernet features supported on z990,
whether GbE, 1000BASE-T Ethernet, or Fast Ethernet when in QDIO mode (OSD
CHPID type).
-
z/OS V1.4 or later with PTFs. Refer to the PSP bucket.
Software Requirements for HiperSockets Enhancements
HiperSockets Support in a VSE/ESA Environment
HiperSockets VLAN (802.1q) support
is supported by the following operating system.
-
Linux for zSeries. Refer to the Web site at the end of this section.
HiperSockets support — 16 Internal LANs and 4096 TCP/IP stacks
is new for HiperSockets (IQD CHPID type), and is exclusive to z990.
There are no prerequisite software requirements.
HiperSockets support of Broadcast for IPv4 packets
is supported by the following operating systems.
-
Linux for zSeries. Refer to the Web site.
-
z/OS V1.5.
-
z/VM V4.4.
HiperSockets (IQD CHPID type) Spanned Channels
is supported by the following operating systems.
-
z/OS support is available via the following PTFs.
-
OA02212/UA01784 for z/OS V1.2 and z/OS V1.3
-
UA01785 for z/OS V1.4
z/VM V3.1, z/VM V4.2, z/VM V4.3, and z/VM V4.4
HiperSockets — Transparently Bridging Traffic (L2S Support)
-
Linux for zSeries. Support is expected to be aligned with delivery of
the hardware support which is planned for October 31, 2003.
Software Requirements for Performance Assists
Adapter Interruptions
are applicable to the FICON Express features (FCP CHPID type) and
OSA-Express features (OSD CHPID type) and are supported by the following
operating systems.
-
z/VM V4.4.
-
Linux for zSeries. Refer to the Web site listed previously.
The Performance Assist for V=V Guests
is applicable to the FICON Express features (FCP CHPID type),
HiperSockets (IQD CHPID type), and OSA-Express features (OSD CHPID type)
and is supported by the following operating system.
Compatibility:
zSeries 800 may not be upgraded to a zSeries 990.
Limitations
Mode of Operation:
The z990 supports LPAR mode only. Basic mode will not be
supported.
OS/390 2.10 and z/OS Limited to LCCS0:
OS/390 V2.10, z/OS V1.2, z/OS V1.3, and z/OS V1.4, with z990
compatibility code installed, can only be IPLed using LCSS0 when running
in LPAR mode. However, these releases may run as a guest under z/VM 4.4,
where z/VM 4.4 is using any channel subsystem. Therefore, you may run
z/OS (or OS/390 V2.10) in LCSS0, and z/VM in LCSS1 where one or more z/OS
images are running as guests. Note the z/OS V1.4 with the z/OS V1.4
z990 Exploitation feature installed can be IPLed using any LCSS.
Dynamic I/O:
Support for dynamic I/O configuration, with z/OS V1.4 z990
Exploitation Support and z/VM V4.4, is planned to be extended to allow
channel paths, control units, and devices to be dynamically added,
changed, and deleted in multiple LCSS's. However, dynamic I/O
configuration is only supported in LCSS0 until October 2003.
Until z/OS V1.4 z990 Exploitation Support is available, z/OS can run only
in a single LCSS. Therefore, z/OS may only execute in LCSS0. z/VM V4.4
may execute in LCSS0 and LCSS1.
z/VM is providing compatibility support for V3.1, V4.2, and V4.3 and
limited exploitation of two LCSS's when running on a z990 server.
Dynamic I/O configuration is supported when only LCSS0 is defined. z/VM
V4.4 may execute in LCSS0 and LCSS1. However, until October 2003,
dynamic I/O is supported in only LCSS0.
Refer to the
Software Requirements
section for more information on availability dates for z/OS and z/VM.
When z/VM is the controlling LPAR, z/VM's dynamic I/O support is designed to handle all of the new
elements of the multiple Channel SubSystem facility for dynamic I/O configuration changes. To
dynamically change the I/O configuration, one of two techniques can be employed:
-
z/VM Control Program (CP) suite of interactive dynamic I/O commands
-
HCM/HCD — New configuration management tools
TKE Limitations:
Customers with TKE 3.X installed workstations may carry forward the features listed below
to control legacy systems only. An update from TKE 3.0 or TKE 3.1 code level to TKE 4.0 code level
is required to control the z990. Workstations updated to the TKE 4.0 code level may be used to
control 9672 Generation 5 and 6, z800, z900, and z990 servers. To use the TKE function on a z990,
the PCIX Cryptographic Coprocessor (PCIXCC) feature, the TKE 4.0 code level, and the CP Assist for
Cryptographic Function (CPACF) must be installed. Refer to the
Software Requirements
section for availability dates.
-
TKE with Token Ring (#0866)
-
TKE with Ethernet (#0869)
-
TKE with Token Ring (#0876)
-
TKE with Ethernet (#0879)
-
TKE with Token Ring (#0886)
-
TKE with Ethernet (#0889)
Cryptographic Limitations on the z990:
The maximum number of PCIXCC features (#0868) is limited to four per system (one processor
per feature). The total number of PCICAs (#0862) and PCIXCCs (#0868) cannot exceed eight per
system. The number of PCICA features cannot exceed six (12 accelerator features). To enable any
cryptographic function on the z990, the CP Assist for Cryptographic Function (#3863) must be
installed and enabled to use PCIXCC, TKE 4.0 code level hardware, and/or PCICA. Review the
Hardware Requirements,
Software Requirements,
and
Planned Availability
sections.
SNA Operations Management:
With the industry move to TCP/IP networks, Systems Network Architecture (SNA) Operations
Management commands will no longer be supported on z990 servers. These commands were previously
used by the System Automation for OS/390 product as well as NetView®. It is recommended that you
now use the Simple Network Management Protocol (SNMP) Application Programming Interfaces (APIs) for
your automation needs.
-
If you previously used the System Automation for OS/390 product, you must now use Version 2.2
or later. This will allow you to define an automation policy for SNMP APIs rather than a policy for
SNA Operations Management commands.
-
If you directly used the SNA Operations Management commands on NetView, you must now use an SNMP
agent and the SNMP APIs for systems automation management.
For detailed information on the SNMP APIs commands and environment requirements, refer to
zSeries Application Programming Interfaces
(SB10-7030). For more information on the SNA Operations Management command support that is not
offered on z990, refer to
Managing Your Processors
(GC38-0452). Both publications are available on Resource Link.
Planning Information
Customer Responsibilities
Customer Responsibilities for Site Preparation:
Information on customer responsibilities for site preparation can be found in the following
publications which are available in the Library section of Resource Link at:
-
System Overview (SA22-1032)
-
Systems Assurance Product Review (SAPR) Guide (SA02-018)
-
Planning for Fiber Optic Links (GA23-0367)
-
zSeries 990 Installation Manual — Physical Planning (GC28-6824)
This document contains information necessary for planning the physical
installation of a zSeries 990 system. It includes physical
specifications, power and cooling data, cabling, and additional details
required for physical installation planning. The intended audience for
this document includes all customer personnel who are responsible for
providing the necessary and appropriate environment for the successful
installation and operation of a zSeries 990 system.
Cable Orders
Fiber Optic Cable Orders:
Fiber optic cables for z800, z900, and z990 are available via
IBM Networking Services.
IBM Networking Services
is enhancing its fiber cabling services to ensure that IBM has a
comprehensive set of scalable solutions to address all cabling
requirements, from product level to enterprise level. IBM Network
Integration and Deployment Services for zSeries fiber cabling provides
planning and installation services for individual fiber optic
connections. IBM Network Integration and Deployment Services for
enterprise fiber cabling provides planning, IBM Fiber Transport System
trunking components, and installation services for small, medium, and
large enterprises.
IGS has the expertise and personnel available to effectively plan and
deploy the appropriate cabling with the future in mind. These services
may include planning, consultation, cable selection, installation, and
documentation, depending upon the services selected.
These services are designed to be right-sized for your products or the
end-to-end enterprise, and to take into consideration the requirements
for all of the protocols/media types supported on zSeries (for example,
ESCON, FICON, Coupling Links, OSA) whether the focus is the data center,
Storage Area Network (SAN), Local Area Network (LAN), or end-to-end
enterprise.
IBM Networking Services
is designed to deliver convenient, packaged services to help reduce the
complexity of planning, ordering, and installing fiber optic cables. The
appropriate fiber cabling is selected based upon the product requirements
and the installed fiber plant.
The services are packaged as follows:
Under the
zSeries fiber cabling services
umbrella there are
three
options to provide individual fiber optic cables (jumper cables,
conversion kits, MCP cables) for connecting to z800, z900, or z990, each
option incorporating a specific set of services.
-
Option 1 — Fiber optic jumper cabling package (available today for z800
and z900). IBM does the detailed planning. This option includes
planning, new cables, installation, and documentation. An analysis of
the zSeries channel configuration, I/O devices, and any existing fiber
optic cabling is required to determine the appropriate fiber optic
cables.
-
Option 2 — Fiber optic jumper cable migration and reuse for a zSeries
upgrade (new option). IBM organizes the existing fiber optic cables
based upon the new z990 connection details. Relabeling, rerouting, and
reconnection to the appropriate z990 channels is performed. New cables
are not offered as a part of this option.
-
Option 3 — Fiber optic jumper cables and installation (new option).
The customer tells us what they need. They do the detailed planning.
The service includes new cables, installation, and documentation.
Planning and providing the list of required cables are customer
responsibilities.
Options 1 and 2 can be combined within one statement of work to provide
complete upgrade coverage.
Under the
enterprise fiber cabling services
umbrella there are
two
options to provide IBM Fiber Transport System (FTS) trunking commodities
(fiber optic trunk cables, fiber harnesses, panel-mount boxes) for
connecting to the z800, z900, and z990.
-
Option 1 — zSeries fiber optic trunk cabling package (new option). IBM
reduces the cable clutter under the floor. An analysis of the zSeries
(z800, z900, z990) channel configuration and any existing fiber optic
cabling is performed to determine the required FTS fiber optic trunking
commodities (trunk cables, harnesses, panel-mount boxes). This option
includes zSeries planning, FTS fiber optic trunking commodities,
installation, and documentation. This option does not include
enterprise-level planning.
-
Option 2 — Enterprise fiber cabling services. IBM organizes the entire
enterprise. This option includes enterprise planning, new cables, fiber
optic trunking commodities, installation, and documentation. This is the
most comprehensive set of services.
Refer to the services section of Resource Link for further details.
Access Resource Link at:
Cabling Responsibilities:
Fiber optic cables, cable planning, labeling, and placement are all
customer responsibilities for new installations and upgrades. Fiber
optic conversion kits and Mode Conditioning Patch (MCP) cables are not
orderable as features on z990. Installation Planning Representatives
(IPRs) and System Service Representatives (SSRs) will not perform the
fiber optic cabling tasks without a services contract.
The following tasks are required to be performed by the customer prior to
machine installation.
-
All fiber optic cable planning.
-
All purchasing of correct, qualified, fiber optic cables.
-
All installation of any required MCP cables.
-
All installation of any required Conversion Kits.
-
All routing of fiber optic cables to correct floor cutouts for proper
installation to server.
-
Use the Physical CHannel IDentifier (PCHID) report or the report from the
CHannel Path IDentifier (CHPID) Mapping Tool to accurately route all
cables.
-
All labeling of fiber optic cables with PCHID numbers for proper
installation to server.
-
Use the PCHID report or the report from the CHPID Mapping Tool to
accurately label all cables.
Additional service charges may be incurred during the server installation
if the above cabling tasks are not accomplished as required.
The table lists the connectors and cabling supported for each of the
features available for ordering on the z990. For convenience, those
features brought forward on an upgrade from z900 are also listed.
Feature Connector
Number Feature Name Type Cable Type
0219 ISC-3 link LC Duplex 9 micron SM
6154 ETR MTRJ 62.5 micron MM
2323 16-port ESCON MTRJ 62.5 micron MM(1)
2319 FICON Express LX LC Duplex 9 micron SM(2)
2320 FICON Express SX LC Duplex 50, 62.5 micron MM
1364 OSA-E GbE LX(3) LC Duplex 9 micron SM
2364(4) OSA-E GbE LX SC Duplex(5) 9 micron SM
1365 OSA-E GbE SX LC Duplex 50, 62.5 micron MM
2365(4) OSA-E GbE SX SC Duplex(5) 50, 62.5 micron MM
1366 OSA-E 1000BASE-T(6) RJ-45 Category 5 UTP(7)
2366(4) OSA-E Fast Ethernet RJ-45 Category 5 UTP
2367 OSA-E Token Ring RJ-45 UTP or STP(8)
-
MM is multimode fiber.
-
SM is single mode fiber.
-
OSA-E refers to OSA-Express.
-
Brought forward to z990 on an upgrade from z900.
-
The OSA-Express GbE features brought forward from z900 have a different
connector (SC Duplex) than the new OSA-E GbE features (LC Duplex).
-
1000BASE-T is the new Ethernet feature.
-
UTP is Unshielded Twisted Pair.
-
STP is Shielded Twisted Pair.
Fiber Quick Connect (FQC), a fiber harness integrated in the zSeries
frame for "quick" connect, is offered as a feature on zSeries for
connection to ESCON channels.
Cables for Integrated Cluster Bus (ICB) links continue to be available as
features. Refer to the
Sales Manual
for a list of these cables and their feature numbers.
For further details regarding the features, refer to the
Sales Manual,
the
Physical Planning Manual,
or the
General Information Manual (GIM)
available on Resource Link.
Note:
IBM Networking Services can satisfy your fiber optic as well as your
copper cabling requirements.
Installability:
Average Installation time for a zSeries 990 is approximately 16
hours. This does not include planning hours. This assumes the
Pre-Installation Configuration Service, a full System Assurance Product
Review, and implementation of the cable services have been performed.
See your IBM representative for details on these services.
Security, Auditability, and Control
The customer is responsible for evaluation, selection, and implementation of security features,
administrative procedures, and appropriate controls in application systems and communications
facilities.