Safe Harbor Statement: The information on IBM products is intended to outline IBM's general product direction and it should not be relied on in making a purchasing decision. The information on the new products is for informational purposes only and may not be incorporated into any contract. The information on IBM products is not a commitment, promise, or legal obligation to deliver any material, code, or functionality. The development, release, and timing of any features or functionality described for IBM products remains at IBM's sole discretion.
Tony Pearson is a an active participant in local, regional, and industry-specific interests, and does not receive any special payments to mention them on this blog.
Tony Pearson receives part of the revenue proceeds from sales of books he has authored listed in the side panel.
Tony Pearson is a Master Inventor and Senior IT Specialist for the IBM System Storage product line at the
IBM Executive Briefing Center in Tucson Arizona, and featured contributor
to IBM's developerWorks. In 2011, Tony celebrated his 25th year anniversary with IBM Storage on the same day as the IBM's Centennial. He is
author of the Inside System Storage series of books. This blog is for the open exchange of ideas relating to storage and storage networking hardware, software and services. You can also follow him on Twitter @az990tony.
(Short URL for this blog: ibm.co/Pearson
Well, it's Wednesday! Normally, IBM makes its announcements on Tuesday's, but this week that landed on the 13th, and some people are superstitious, so we pushed it back to today. Fellow IBM master inventory Barry Whyte starts the first in a series of posts with: New SVC v5 CF8 node with native SSD support.
There are really two separate items being announced for the IBM System Storage SAN Volume Controller (SVC):
SVC v5 software
The software moves from a 32-bit kernel to a 64-bit kernel. Fortunately, IBM had the foresight to know that would happen back in 2005, so models 8F2, 8F4 and 8G4 can be upgrade to this new software level and gain new functionality. This is because these models have 64-bit capable processors. Those with six-year-old 4F2 will continue to run on SVC 4.3.1, but should consider it's about time to upgrade soon.
New 2145-CF8 model
The CF8 is based on the IBM System x 3550M2. Each node can have up to 4 Solid-State Drives (SSD) that can be treated as SVC Managed Disk Groups. Virtual disks can be easily migrated from hard disk drives (HDD) over to SSD, processed, and then move them back to HDD. By treating the SSD as managed disks, rather than an extension of the cache, we are able to support all of the features and functions in a seamless manner.
As Barry says, IBM has been working on this for quite a while, and based on initial responses looks to be quite successful in the market!
Well, it's Tuesday again, but this time, today we had our third big storage launch of 2009! A lot got announced today as part of IBM's big "Dynamic Infrastructure" marketing campaign. I will just focus on the
disk-related announcements today:
IBM System Storage DS8700
IBM adds a new model to its DS8000 series with the
[IBM System Storage DS8700]. Earlier this month, fellow blogger and arch-nemesis Barry Burke from EMC posted [R.I.P DS8300] on this mistaken assumption that the new DS8700 meant that DS8300 was going away, or that anyone who bought a DS8300 recently would be out of luck. Obviously, I could not respond until today's announcement, as the last thing I want to do is lose my job disclosing confidential information. BarryB is wrong on both counts:
IBM will continue to sell the DS8100 and DS8300, in addition to the new DS8700.
Clients can upgrade their existing DS8100 or DS8300 systems to DS8700.
BarryB's latest post [What's In a Name - DS8700] is fair game, given all the fun and ridicule everyone had at his expense over EMC's "V-Max" name.
So the DS8700 is new hardware with only 4 percent new software. On the hardware side, it uses faster POWER6 processors instead of POWER5+, has faster PCI-e buses instead of the RIO-G loops, and faster four-port device adapters (DAs) for added bandwidth between cache and drives. The DS8700 can be ordered as a single-frame dual 2-way that supports up to 128 drives and 128GB of cache, or as a dual 4-way, consisting of one primary frame, and up to four expansion frames, with up to 384GB of cache and 1024 drives.
Not mentioned explicitly in the announcements were the things the DS8700 does not support:
ESCON attachment - Now that FICON is well-established for the mainframe market, there is no need to support the slower, bulkier ESCON options. This greatly reduced testing effort. The 2-way DS8700 can support up to 16 four-port FICON/FCP host adapters, and the 4-way can support up to 32 host adapters, for a maximum of 128 ports. The FICON/FCP host adapter ports can auto-negotiate between 4Gbps, 2Gbps and 1Gbps as needed.
LPAR mode - When IBM and HDS introduced LPAR mode back in 2004, it sounded like a great idea the engineers came up with. Most other major vendors followed our lead to offer similar "partitioning". However, it turned out to be what we call in the storage biz a "selling apple" not a "buying apple". In other words, something the salesman can offer as a differentiating feature, but that few clients actually use. It turned out that supporting both LPAR and non-LPAR modes merely doubled the testing effort, so IBM got rid of it for the DS8700.
Update: I have been reminded that both IBM and HDS delivered LPAR mode within a month of each other back in 2004, so it was wrong for me to imply that HDS followed IBM's lead when obviously development happened in both companies for the most part concurrently prior to that. EMC was late to the "partition" party, but who's keeping track?
Initial performance tests show up to 50 percent improvement for random workloads, and up to 150 percent improvement for sequential workloads, and up to 60 percent improvement in background data movement for FlashCopy functions. The results varied slightly between Fixed Block (FB) LUNs and Count-Key-Data (CKD) volumes, and I hope to see some SPC-1 and SPC-2 benchmark numbers published soon.
The DS8700 is compatible for Metro Mirror, Global Mirror, and Metro/Global Mirror with the rest of the DS8000 series, as well as the ESS model 750, ESS model 800 and DS6000 series.
New 600GB FC and FDE drives
IBM now offers [600GB drives] for the DS4700 and DS5020 disk systems, as well as the EXP520 and EXP810 expansion drawers. In each case, we are able to pack up to 16 drives into a 3U enclosure.
Personally, I think the DS5020 should have been given a DS4xxx designation, as it resembles the DS4700
more than the other models of the DS5000 series. Back in 2006-2007, I was the marketing strategist for IBM System Storage product line, and part of my job involved all of the meetings to name or rename products. Mostly I gave reasons why products should NOT be renamed, and why it was important to name the products correctly at the beginning.
IBM System Storage SAN Volume Controller hardware and software
Fellow IBM master inventory Barry Whyte has been covering the latest on the [SVC 2145-CF8 hardware]. IBM put out a press release last week on this, and today is the formal announcement with prices and details. Barry's latest post
[SVC CF8 hardware and SSD in depth] covers just part of the entire
The other part of the announcement was the [SVC 5.1 software] which can be loaded
on earlier SVC models 8F2, 8F4, and 8G4 to gain better performance and functionality.
To avoid confusion on what is hardware machine type/model (2145-CF8 or 2145-8A4) and what is software program (5639-VC5 or 5639-VW2), IBM has introduced two new [Solution Offering Identifiers]:
5465-028 Standard SAN Volume Controller
5465-029 Entry Edition SAN Volume Controller
The latter is designed for smaller deployments, supports only a single SVC node-pair managing up to
150 disk drives, available in Raven Black or Flamingo Pink.
EXN3000 and EXP5060 Expansion Drawers
IBM offers the [EXN3000 for the IBM N series]. These expansion drawers can pack 24 drives in a 4U enclosure. The drives can either be all-SAS, or all-SATA, supporting 300GB, 450GB, 500GB and 1TB size capacity drives.
The [EXP5060 for the IBM DS5000 series] is a high-density expansion drawer that can pack up to 60 drives into a 4U enclosure. A DS5100 or DS5300
can handle up to eight of these expansion drawers, for a total of 480 drives.
Pre-installed with Tivoli Storage Productivity Center Basic Edition. Basic Edition can be upgraded with license keys to support Data, Disk and Standard Edition to extend support and functionality to report and manage XIV, N series, and non-IBM disk systems.
Pre-installed with Tivoli Key Lifecycle Manager (TKLM). This can be used to manage the Full Disk Encryption (FDE) encryption-capable disk drives in the DS8000 and DS5000, as well as LTO and TS1100 series tape drives.
IBM Tivoli Storage FlashCopy Manager v2.1
The [IBM Tivoli Storage FlashCopy Manager V2.1] replaces two products in one. IBM used
to offer IBM Tivoli Storage Manager for Copy Services (TSM for CS) that protected Windows application data, and IBM Tivoli Storage Manager for Advanced Copy Services (TSM for ACS) that protected AIX application data.
The new product has some excellent advantages. FlashCopy Manager offers application-aware backup of LUNs containing SAP, Oracle, DB2, SQL server and Microsoft Exchange data. It can support IBM DS8000, SVC and XIV point-in-time copy functions, as well as the Volume Shadow Copy Services (VSS) interfaces of the IBM DS5000, DS4000 and DS3000 series disk systems. It is priced by the amount of TB you copy, not on the speed or number of CPU processors inside the server.
Don't let the name fool you. IBM FlashCopy Manager does not require that you use Tivoli Storage Manager (TSM) as your backup product. You can run IBM FlashCopy Manager on its own, and it will manage your FlashCopy target versions on disk, and these can be backed up to tape or another disk using any backup product. However, if you are lucky enough to also be using TSM, then there is optional integration that allows TSM to manage the target copies, move them to tape, inventory them in its DB2 database, and provide complete reporting.
Yup, that's a lot to announce in one day. And this was just the disk-related portion of the launch!
This week several IBM executives will present at the 28th Annual Data Center Conference here in Las Vegas. Here is a quick recap:
Steve Sams: Data Center Cost Saving Actions Your CFO Will Love
A startling 78 percent of today's data centers were built in the last century, before the "dot com" era and the adoption of high-density blade servers. IBM Vice President of Global Site and Facility Services, Steve Sams, presented actions that can help extend the life of existing data centers, help rationalize the infrastructure across the company, and design a new data center that is flexible and responsive to changing needs.
In one example, an 85,000 square foot datacenter in Lexington had reached 98 percent capacity based on power/cooling requirements. They estimated it would take $53 million US dollars to either upgrade the facility or build a new facility to meet projected growth. Instead, IBM was able to consolidate servers six-to-one, an 85 percent reduction. IBM also was able to make changes to the cooling equipment, redirect airflow and changed out the tiles, re-oriented the servers for more optimal placement, and implement measurement and management tools. The end result? The facility now has eight times the compute capability and enjoys 15 percent headroom for additonal growth. All this for only 1.5 million US dollar investment, instead of 53 million.
IBM builds hundreds of data centers for clients large and small. In addition to the "Portable Modular Data Center"(PMDC) shipping container on display at the Solution Showcase, IBM offers the "Scalable Modular Data Center", a turn-key system with a small 500 to 2500 square foot size for small customers. For larger deployments, the "Enterprise Modular Data Center" offers standardized deployments in 5000 square foot increments. IBM also offers "High Density Zones" which can be perfect way to avoid a full site retrofit.
Helene Armitage: IT-wide Virtualization
Helene is IBM General Manager of the newly formed IBM System Software division. A smarter planet will require more dynamic infrastructures, which is IBM's approach to helping clients through the virtualization journey. The virtualization of resources, workloads and business processes will require end-to-end management. To help, IBM offers IBM Systems Director.
Helene indicated that there are four stages of adoption:
Physical consolidation - VMware and Hyper-V are the latest examples of running many applications on fewer physical servers. Of course, IBM has been doing this for decades with mainframes, and has had virtualization on System i and System p POWER systems as well. A quick survey of the audience found that about 20 percent were doing server virtualization on non-x86 platforms (for example, PowerVM or System z mainframe z/VM)
Pools of resources - SAN Volume Controller is an example solution to manage storage as a pool of disparate storage resources. Supercomputers manage pools of servers.
Integrated Service Management - in the past, resources were managed by domain, resulting in islands of management. Now, with IBM Systems Director, you can manage AIX, IBM i, Linux and Windows servers, including non-IBM servers running Linux and Windows.
Service management can provide monitoring, provisioning, service catalog, self-service, and business-aligned processes.
Cloud computing - Helene agreed that not everyone will get to this stage. Some will adopt cloud computing, whether public, private or some kind of hybrid, and others may be fine at stage 3.
For those clients that want assistance, IBM offers three levels of help:
Help me decide what is best for me
Help me implement what I have decided to do
Help me manage and run my operations
With IBM's compelling vision for the future, best of breed solutions, leadership in management software, extensive experience in services, and solid business industry knowledge, it makes sense to tap IBM to help with your next IT transformation.
Continuing my coverage of the Data Center Conference, Dec 1-4, 2009 here in Las Vegas, this post focused on data protection strategies.
Two analysts co-presented this session which provided an overview of various data protection techniques. A quick survey of the audience found that 27 percent have only a single data center, 13 percent have load sharing of their mission critical applications across multiple data centers, and the rest use a failover approach to either development/test resources, standby resources or an outsourced facility.
There are basically five ways to replicate data to secondary locations:
Array-based replication. Many high-end disk arrays offer this feature. IBM's DS8000 and XIV both have synchronous and asynchronous mirroring. Data Deduplication can help in this regard to reduce the amount of data transmitted across locations.
NAS-based replication. I consider this just another variant of the first, but this can be file-based instead of block-based, and can often be done over the public internet rather than dark fiber.
Network-based replication. This is the manner that IBM SAN Volume Controller, EMC RecoverPoint, and others can replicate. The analysts liked this approach as it was storage vendor-independent.
Host-based replication. This is often done by the host's Operating System, such as through a Logical Volume Manager (LVM) component.
Application/Database replication. There are a variety of techniques, including log shipping of transactions, SQL replication, and active/active application-specific implementations.
The analysts felt that "DR Testing" has become a lost art. People are just not doing it as often as they should, or not doing it properly, resulting in surprises when a real disaster strikes.
A question came up about the confusion between "Disaster Recovery Tiers" and Uptime Institute's "Data Center Facilities Tiers". I agree this is confusing. Many clients call their most mission critical applications as Tier 1, less critical as Tier 2, and least critical as Tier 3. In 1983, IBM User Group GUIDE came up with "Business Continuity Tiers" where Tier 1 was the slowest recovery from manual tape, and Tier 7 was the fastest recovery with a completely automated site, network, server and storage failover. However, for Data Center facility tiers, Uptime has the simplest least available (99.3 percent uptime) data center as Tier 1, and the most advanced, redundant, highest available (99.995 percent) data center as Tier 4. This just goes to show that when one person starts using "Tier 1" or "Tier 4" terminology, it can be misinterpreted by others.
"With Cisco Systems, EMC, and VMware teaming up to sell integrated IT stacks, Oracle buying Sun Microsystems to create its own integrated stacks, and IBM having sold integrated legacy system stacks and rolling in profits from them for decades, it was only a matter of time before other big IT players paired off."
Once again we are reminded that IBM, as an IT "supermarket", is able to deliver integrated software/server/storage solutions, and our competitors are scrambling to form their own alliances to be "more like IBM." This week, IBM announced new ordering options for storage software with System x servers, including BladeCenter blade servers and IntelliStation workstations. Here's a quick recap:
IBM Tivoli Storage Manager FastBack v6.1 supports both Windows and Linux! FastBack is a data protection solution for ROBO (Remote Office, Branch Office) locations. It can protect Microsoft Exchange, Lotus Domino, DB2, Oracle applications. FastBack can provide full volume-level recovery, as well as individual file recovery, and in some cases Bare Machine Recovery. FastBack v6.1 can be run stand-alone, or integrated with a full IBM Tivoli Storage Manager (TSM) unified recovery management solution.
FlashCopy Manager v2.1
FlashCopy Manager uses point-in-time copy capabilities, such as SnapShot or FlashCopy, to protect application data using an application-aware approach for Microsoft Exchange, Microsoft SQL server, DB2, Oracle, and SAP. It can be used with IBM SAN Volume Controller (SVC), DS8000 series, DS5000 series, DS4000 series, DS3000 series, and XIV storage systems. When applicable, FlashCopy manager coordinates its work with Microsoft's Volume Shadow Copy Services (VSS) interface. FlashCopy Manager can provide data protection using just point-in-time disk-resident copies, or can be integrated with a full IBM Tivoli Storage Manager (TSM) unified recovery management solution to move backup images to external storage pools, such as low-cost, energy-efficient tape cartridges.
General Parallel File System (GPFS) v3.3 Multiplatform
GPFS can support AIX, Linux, and Windows! Version 3.3 adds support for Windows 2008 Server on 64-bit chipset architectures from AMD and Intel. Now you can have a common GPFS cluster with AIX, Linux and Windows servers all sharing and accessing the same files. A GPFS cluster can have up to 256 file systems. Each of these file systems can be up to 1 billion files, up to 1PB of data, and can have up to 256 snapshots. GPFS can be used stand-alone, or integrated with a full IBM Tivoli Storage Manager (TSM) unified recovery management solution with parallel backup streams.
For full details on these new ordering options, see the IBM [Press Release].
Continuing my drawn out coverage of IBM's big storage launch of February 9, today I'll cover the IBM System Storage TS7680 ProtecTIER data deduplication gateway for System z.
On the host side, TS7680 connects to mainframe systems running z/OS or z/VM over FICON attachment, emulating an automated tape library with 3592-J1A devices. The TS7680 includes two controllers that emulate the 3592 C06 model, with 4 FICON ports each. Each controller emulates up to 128 virtual 3592 tape drives, for a total of 256 virtual drives per TS7680 system. The mainframe sees up to 1 million virtual tape cartridges, up to 100GB raw capacity each, before compression. For z/OS, the automated library has full SMS Tape and Integrated Library Management capability that you would expect.
Inside, the two control units are both connected to a redundant pair cluster of ProtecTIER engines running the HyperFactor deduplication algorithm that is able to process the deduplication inline, as data is ingested, rather than post-process that other deduplication solutions use. These engines are similar to the TS7650 gateway machines for distributed systems.
On the back end, these ProtecTIER deduplication engines are then connected to external disk, up to 1PB. If you get 25x data deduplication ratio on your data, that would be 25PB of mainframe data stored on only 1PB of physical disk. The disk can be any disk supported by ProtecTIER over FCP protocol, not just the IBM System Storage DS8000, but also the IBM DS4000, DS5000 or IBM XIV storage system, various models of EMC and HDS, and of course the IBM SAN Volume Controller (SVC) with all of its supported disk systems.
Here I am, day 11 of a 17-day business trip, on my last leg of the trip this week, in Kuala Lumpur in Malaysia. I have been flooded with requests to give my take on EMC's latest re-interpretation of storage virtualization, VPLEX.
I'll leave it to my fellow IBM master inventor Barry Whyte to cover the detailed technical side-by-side comparison. Instead, I will focus on the business side of things, using Simon Sinek's Why-How-What sequence. Here is a [TED video] from Garr Reynold's post
[The importance of starting from Why].
Let's start with the problem we are trying to solve.
Problem: migration from old gear to new gear, old technology to new technology, from one vendor to another vendor, is disruptive, time-consuming and painful.
Given that IT storage is typically replaced every 3-5 years, then pretty much every company with an internal IT department has this problem, the exception being those companies that don't last that long, and those that use public cloud solutions. IT storage can be expensive, so companies would like their new purchases to be fully utilized on day 1, and be completely empty on day 1500 when the lease expires. I have spoken to clients who have spent 6-9 months planning for the replacement or removal of a storage array.
A solution to make the data migration non-disruptive would benefit the clients (make it easier for their IT staff to keep their data center modern and current) as well as the vendors (reduce the obstacle of selling and deploying new features and functions). Storage virtualization can be employed to help solve this problem. I define virtualization as "technology that makes one set of resources look and feel like a different set of resources, preferably with more desirable characteristics.". By making different storage resources, old and new, look and feel like a single type of resource, migration can be performed without disrupting applications.
Before VPLEX, here is a breakdown of each solution:
Non-disruptive tech refresh, and a unified platform to provide management and functionality across heterogeneous storage.
Non-disruptive tech refresh, and a unified platform to provide management and functionality between internal tier-1 HDS storage, and external tier-2 heterogeneous storage.
Non-disruptive tech refresh, with unified multi-pathing driver that allows host attachment of heterogeneous storage.
New in-band storage virtualization device
Add in-band storage virtualization to existing storage array
New out-of-band storage virtualization device with new "smart" SAN switches
SAN Volume Controller
HDS USP-V and USP-VM
For IBM, the motivation was clear: Protect customers existing investment in older storage arrays and introduce new IBM storage with a solution that allows both to be managed with a single set of interfaces and provide a common set of functionality, improving capacity utilization and availability. IBM SAN Volume Controller eliminated vendor lock-in, providing clients choice in multi-pathing driver, and allowing any-to-any migration and copy services. For example, IBM SVC can be used to help migrate data from an old HDS USP-V to a new HDS USP-V.
With EMC, however, the motivation appeared to protect software revenues from their PowerPath multi-pathing driver, TimeFinder and SRDF copy services. Back in 2005, when EMC Invista was first announced, these three software represented 60 percent of EMC's bottom-line profit. (Ok, I made that last part up, but you get my point! EMC charges a lot for these.)
Back in 2006, fellow blogger Chuck Hollis (EMC) suggested that SVC was just a [bump in the wire] which could not possibly improve performance of existing disk arrays. IBM showed clients that putting cache(SVC) in front of other cache(back end devices) does indeed improve performance, in the same way that multi-core processors successfully use L1/L2/L3 cache. Now, EMC is claiming their cache-based VPLEX improves performance of back-end disk. My how EMC's story has changed!
So now, EMC announces VPLEX, which sports a blend of SVC-like and Invista-like characteristics. Based on blogs, tweets and publicly available materials I found on EMC's website, I have been able to determine the following comparison table. (Of course, VPLEX is not yet generally available, so what is eventually delivered may differ.)
Scalable, 1 to 4 node-pairs
One size fits all, single pair of CPCs
SVC-like, 1 to 4 director-pairs
Works with any SAN switches or directors
Required special "smart" switches (vendor lock-in)
SVC-like, works with any SAN switches or directors
Broad selection of IBM Subsystem Device Driver (SDD) offered at no additional charge, as well as OS-native drivers Windows MPIO, AIX MPIO, Solaris MPxIO, HP-UX PV-Links, VMware MPP, Linux DM-MP, and comercial third-party driver Symantec DMP.
Limited selection, with focus on priced PowerPath driver
Invista-like, PowerPath and Windows MPIO
Read cache, and choice of fast-write or write-through cache, offering the ability to improve performance.
No cache, Split-Path architecture cracked open Fibre Channel packets in flight, delayed every IO by 20 nanoseconds, and redirected modified packets to the appropriate physical device.
SVC-like, Read and write-through cache, offering the ability to improve performance.
Space-Efficient Point-in-Time copies
SVC FlashCopy supports up to 256 space-efficient targets, copies of copies, read-only or writeable, and incremental persistent pairs.
Like Invista, No
Remote distance mirror
Choice of SVC Metro Mirror (synchronous up to 300km) and Global Mirror (asynchronous), or use the functionality of the back-end storage arrays
No native support, use functionality of back-end storage arrays, or purchase separate product called EMC RecoverPoint to cover this lack of functionality
Limited synchronous remote-distance mirror within VPLEX (up to 100km only), no native asynchronous support, use functionality of back-end storage arrays
Provides thin provisioning to devices that don't offer this natively
Like Invista, No
SVC Split-Cluster allows concurrent read/write access of data to be accessed from hosts at two different locations several miles apart
I don't think so
PLEX-Metro, similar in concept but implemented differently
Non-disruptive tech refresh
Can upgrade or replace storage arrays, SAN switches, and even the SVC nodes software AND hardware themselves, non-disruptively
Tech refresh for storage arrays, but not for Invista CPCs
Tech refresh of back end devices, and upgrade of VPLEX software, non-disruptively. Not clear if VPLEX engines themselves can be upgraded non-disruptively like the SVC.
Heterogeneous Storage Support
Broad support of over 140 different storage models from all major vendors, including all CLARiiON, Symmetrix and VMAX from EMC, and storage from many smaller startups you may not have heard of
Invista-like. VPLEX claims to support a variety of arrays from a variety of vendors, but as far as I can find, only DS8000 supported from the list of IBM devices. Fellow blogger Barry Burke (EMC) suggests [putting SVC between VPLEX and third party storage devices] to get the heterogeneous coverage most companies demand.
Back-end storage requirement
Must define quorum disks on any IBM or non-IBM back end storage array. SVC can run entirely on non-IBM storage arrays
HP SVSP-like, requires at least one EMC storage array to hold metadata
SVC 2145-CF8 model supports up to four solid-state drives (SSD) per node that can treated as managed disk to store end-user data
Invista-like. VPLEX has an internal 30GB SSD, but this is used only for operating system and logs, not for end-user data.
In-band virtualization solutions from IBM and HDS dominate the market. Being able to migrate data from old devices to new ones non-disruptively turned out to be only the [tip of the iceberg] of benefits from storage virtualization. In today's highly virtualized server environment, being able to non-disruptively migrate data comes in handy all the time. SVC is one of the best storage solutions for VMware, Hyper-V, XEN and PowerVM environments. EMC watched and learned in the shadows, taking notes of what people like about the SVC, and decided to follow IBM's time-tested leadership to provide a similar offering.
EMC re-invented the wheel, and it is round. On a scale from Invista (zero) to SVC (ten), I give EMC's new VPLEX a six.
Well, I'm back safely from my tour of Asia. I am glad to report that Tokyo, Beijing and Kuala Lumpur are pretty much how I remember them from the last time I was there in each city. I have since been fighting jet lag by watching the last thirteen episodes of LOST season 6 and the series finale.
Recently, I have started seeing a lot of buzz on the term "Storage Federation". The concept is not new, but rather based on the work in database federation, first introduced in 1985 by [A federated architecture for information management] by Heimbigner and McLeod. For those not familiar with database federation, you can take several independent autonomous databases, and treat them as one big federated system. For example, this would allow you to issue a single query and get results across all the databases in the federated system. The advantage is that it is often easier to federate several disparate heterogeneous databases than to merge them into a single database. [IBM Infosphere Federation Server] is a market leader in this space, with the capability to federate DB2, Oracle and SQL Server databases.
Storage expansion: You want to increase the storage capacity of an existing storage system that cannot accommodate the total amount of capacity desired. Storage Federation allows you to add additional storage capacity by adding a whole new system.
Storage migration: You want to migrate from an aging storage system to a new one. Storage Federation allows the joining of the two systems and the evacuation from storage resources on the first onto the second and then the first system is removed.
Safe system upgrades: System upgrades can be problematic for a number of reasons. Storage Federation allows a system to be removed from the federation and be re-inserted again after the successful completion of the upgrade.
Load balancing: Similar to storage expansion, but on the performance axis, you might want to add additional storage systems to a Storage Federation in order to spread the workload across multiple systems.
Storage tiering: In a similar light, storage systems in a Storage Federation could have different capacity/performance ratios that you could use for tiering data. This is similar to the idea of dynamically re-striping data across the disk drives within a single storage system, such as with 3PAR's Dynamic Optimization software, but extends the concept to cross storage system boundaries.
To some extent, IBM SAN Volume Controller (SVC), XIV, Scale-Out NAS (SONAS), and Information Archive (IA) offer most, if not all, of these capabilities. EMC claims its VPLEX will be able to offer storage federation, but only with other VPLEX clusters, which brings up a good question. What about heterogenous storage federation? Before anyone accuses me of throwing stones at glass houses, let's take a look at each IBM solution:
IBM SAN Volume Controller
The IBM SAN Volume Controller has been doing storage federation since 2003. Not only can IBM SAN Volume Controller bring together storage from a variety of heterogenous storage, the SVC cluster itself can be a mix of different hardware models. You can have a 2145-8A4 node pair, 2145-8G4 node pair, and the new 2145-CF8 node pair, all combined together into a single SVC cluster. Upgrading SVC hardware nodes in an SVC cluster is always non-disruptive.
IBM XIV storage system
The IBM XIV has two kinds of independent modules. Data modules have processor, cache and 12 disks. Interface modules are data modules with additional processor, FC and Ethernet (iSCSI) adapters. Because these two modules play different roles in an XIV "colony", that number of each type is predetermined. Entry-level six-module systems have 2 interface and 4 data modules. Full 15-module systems have 6 interface and 9 data modules. Individual modules can be added or removed non-disruptively in an XIV.
IBM Scale-Out NAS
The SONAS is comprised of three kinds of nodes that work together in concert. A management node, one or more interface nodes, and two or more storage nodes. The storage nodes are paired to manage up to 240 nodes in a storage pod. Individual interface or data nodes can be added or removed non-disruptively in the SONAS. The underlying technology, the General Parallel File System, has been doing storage federation since 1996 for some of the largest top 500 supercomputers in the world.
IBM Information Archive (IA)
For the IA, there are 1, 2 or 3 nodes, which manages a set of collections. A collection can either be file-based using industry-standard NAS protocols, or object-based using the popular System Storage™ Archive Manager (SSAM) interface. Normally, you have as many collections as you have nodes, but nodes are powerful enough to manage two collections to provide N-1 availability. This allows a node to be removed, and a new node added into the IA "colony", in a non-disruptive manner.
Even in an ant colony, there are only a few types of ants, with typically one queen, several males, and lots of workers. But all the ants are red. You don't see colonies that mix between different species of ants. For databases, federation was a way to avoid the much harder task of merging databases from different platforms. For storage, I am surprised people have latched on to the term "federation", given our mixed results in the other "federations" we have formed, which I have conveniently (IMHO) ranked from least effective to most effective:
The Union of Soviet Socialist Republics (USSR)
My father used to say, "If the Soviet Union were in charge of the Sahara desert, they would run out of sand in 50 years." The [Soviet Union] actually lasted 68 years, from 1922 to 1991.
The United Nations (UN)
After the previous League of Nations failed, the UN was formed in 1945 to facilitate cooperation in international law, international security, economic development, social progress, human rights, and the achieving of world peace by stopping wars between countries, and to provide a platform for dialogue.
The European Union (EU)
With the collapse of the Greek economy, and the [rapid growth of debt] in the UK, Spain and France, there are concerns that the EU might not last past 2020.
The United States of America (USA)
My own country is a federation of states, each with its own government. California's financial crisis was compared to the one in Greece. My own state of Arizona is under boycott from other states because of its recent [immigration law]. However, I think the US has managed better than the EU because it has evolved over the past 200 years.
The Organization of the Petroleum Exporting Countries [OPEC]
Technically, OPEC is not a federation of cooperating countries, but rather a cartel of competing countries that have agreed on total industry output of oil to increase individual members' profits. Note that it was a non-OPEC company, BP, that could not "control their output" in what has now become the worst oil spill in US history. OPEC was formed in 1960, and is expected to collapse sometime around 2030 when the world's oil reserves run out. Matt Savinar has a nice article on [Life After the Oil Crash].
United Federation of Planets
The [Federation] fictitiously described in the Star Trek series appears to work well, an optimistic view of what federations could become if you let them evolve long enough.
Given the mixed results with "federation", I think I will avoid using the term for storage, and stick to the original term "scale-out architecture".
Continuing my coverage of the annual [2010 System Storage Technical University], I participated in the storage free-for-all, which is a long-time tradition, started at SHARE User Group conference, and carried forward to other IT conferences. The free-for-all is a Q&A Panel of experts to allow anyone to ask any question. These are sometimes called "Birds of a Feather" (BOF). Last year, they were called "Meet the Experts", one for mainframe storage, and the other for storage attached to distributed systems. This year, we had two: one focused on Tivoli Storage software, and the second to cover storage hardware. This post provides a recap of the Storage Hardware free-for-all.
The emcee for the event was Scott Drummond. The other experts on the panel included Dan Thompson, Carlos Pratt, Jack Arnold, Jim Blue, Scott Schroder, Ed Baker, Mike Wood, Steve Branch, Randy Arseneau, Tony Abete, Jim Fisher, Scott Wein, Rob Wilson, Jason Auvenshine, Dave Canan, Al Watson, and myself, yours truly, Tony Pearson.
What can I do to improve performance on my DS8100 disk system? It is running a mix of sequential batch processing and my medical application (EPIC). I have 16GB of cache and everything is formatted as RAID-5.
We are familiar with EPIC. It does not "play well with others", so IBM recommends you consider dedicating resources for just the EPIC data. Also consider RAID-10 instead for the EPIC data.
How do I evaluate IBM storage solutions in regards to [PCI-DSS] requirements.
Well, we are not lawyers, and some aspects of the PCI-DSS requirements are outside the storage realm. In March 2010, IBM was named ["Best Security Company"] by SC Magazine, and we have secure storage solutions for both disk and tape systems. IBM DS8000 and DS5000 series offer Full Disk Encryption (FDE) disk drives. IBM LTO-4/LTO-5 and TS1120/TS1130 tape drives meet FIPS requirements for encryption. We will provide you contact information on an encryption expert to address the other parts of your PCI-DSS specific concerns.
My telco will only offer FCIP routing for long-distance disk replication, but my CIO wants to use Fibre Channel routing over CWDM, what do I do?
IBM XIV, DS8000 and DS5000 all support FC-based long distance replication across CWDM. However, if you don't have dark fiber, and your telco won't provide this option, you may need to re-negotiate your options.
My DS4800 sometimes reboots repeatedly, what should I do.
This was a known problem with microcode level 760.28, it was detecting a failed drive. You need to replace the drive, and upgrade to the latest microcode.
Should I use VMware snapshots or DS5000 FlashCopy?
VMware snapshots are not free, you need to upgrade to the appropriate level of VMware to get this function, and it would be limited to your VMware data only. The advantage of DS5000 FlashCopy is that it applies to all of your operating systems and hypervisors in use, and eliminates the consumption of VMware overhead. It provides crash-consistent copies of your data. If your DS5000 disk system is dedicated to VMware, then you may want to compare costs versus trade-offs.
Any truth to the rumor that Fibre Channel protocol will be replaced by SAS?
SAS has some definite cost advantages, but is limited to 8 meters in length. Therefore, you will see more and more usage of SAS within storage devices, but outside the box, there will continue to be Fibre Channel, including FCP, FICON and FCoE. The Fibre Channel Industry Alliance [FCIA] has a healthy roadmap for 16 Gbps support and 20 Gbps interswitch link (ISL) connections.
What about Fibre Channel drives, are these going away?
We need to differentiate the connector from the drive itself. Manufacturers are able to produce 10K and 15K RPM drives with SAS instead of FC connectors. While many have suggested that a "Flash-and-Stash" approach of SSD+SATA would eliminate the need for high-speed drives, IBM predicts that there just won't be enough SSD produced to meet the performance needs of our clients over the next five years, so 15K RPM drives, more likely with SAS instead of FC connectors, will continue to be deployed for the next five years.
We'd like more advanced hands-on labs, and to have the certification exams be more product-specific rather than exams for midrange disk or enterprise disk that are too wide-ranging.
Ok, we will take that feedback to the conference organizers.
IBM Tivoli Storage Manager is focused on disaster recovery from tape, how do I incorporate remote disk replication.
This is IBM's Unified Recovery Management, based on the seven tiers of disaster recovery established in 1983 at GUIDE conference. You can combine local recovery with FastBack, data center server recovery with TSM and FlashCopy manager, and combine that with IBM Tivoli Storage Productivity Center for Replication (TPC-R), GDOC and GDPS to manage disk replication across business continuity/disaster recovery (BC/DR) locations.
IBM Tivoli Storage Productivity Center for Replication only manages the LUNs, what about server failover and mapping the new servers to the replicated LUNs?
There are seven tiers of disaster recovery. The sixth tier is to manage the storage replication only, as TPC-R does. The seventh tier adds full server and network failover. For that you need something like IBM GDPS or GDOC that adds this capability.
All of my other vendor kit has bold advertising, prominent lettering, neon lights, bright colors, but our IBM kit is just black, often not even identifying the specific make or model, just "IBM" or "IBM System Storage".
IBM has opted for simplified packaging and our sleek, signature "raven black" color, and pass these savings on to you.
Bring back the SHARK fins!
We will bring that feedback to our development team. ("Shark" was the codename for IBM's ESS 800 disk model. Fiberglass "fins" were made as promotional items and placed on top of ESS 800 disk systems to help "identify them" on the data center floor. Unfortunately, professional golfer [<a href="http://www.shark.com/">Greg Norman</a>] complained, so IBM discontinued the use of the codename back in 2005.)
Where is Infiniband?
Like SAS, Infiniband had limited distance, about 10 to 15 meters, which proved unusable for server-to-storage network connections across data center floorspace. However, there are now 150 meter optical cables available, and you will find Infiniband used in server-to-server communications and inside storage systems. IBM SONAS uses Infiniband today internally. IBM DCS9900 offers Infiniband host-attachment for HPC customers.
We need midrange storage for our mainframe please?
In addition to the IBM System Storage DS8000 series, the IBM SAN Volume Controller and IBM XIV are able to connect to Linux on System z mainframes.
We need "Do's and Don'ts" on which software to run with which hardware.
IBM [Redbooks] are a good source for that, and we prioritize our efforts based on all those cards and letters you send the IBM Redbooks team.
The new TPC v4 reporting tool requires a bit of a learning curve.
The new reporting tool, based on Eclipse's Business Intelligence Reporting Tool [BIRT], is now standardized across the most of the Tivoli portfolio. Check out the [Tivoli Common Reporting] community page for assistance.
An unfortunate side-effect of using server virtualization like VMware is that it worsens management and backup issues. We now have many guests on each blade server.
IBM is the leading reseller of VMware, and understands that VMware adds an added layer of complexity. Thankfully, IBM Tivoli Storage Manager backups uses a lightweight agent. IBM [System Director VMcontrol] can help you manage a variety of hypervisor environments.
This was a great interactive session. I am glad everyone stayed late Thursday evening to participate in this discussion.
Last week, in Computer Technology Review's article [Tiering: Scale Up? Scale Out? Do Both], Mark Ferelli interviews fellow blogger Hu Yoshida, CTO of Hitachi Data Systems (HDS). Here's an excerpt:
"MF/CTR: A global cache should be required to implement that common pool that you’re talking about going across all tiers.
Hu/HDS: Right. So that is needed to get to all the resources. Now with our system, we can also attach external storage behind it for capacity so that as the storage ages out or becomes less active we can move it to the external storage. They would certainly have less performance capability, but you don’t need it for the stale data that we’re aging down. Right now we’re the only vendor that can provide this type of tiering.
If you look at other people who do virtualization like IBM’s SVC, the SVC has no storage within it because it’s sitting so if you attach any storage behind it, there is some performance degradation because you have this appliance sitting in front. That appliance is also very limited in cache and very limited in the number of storage boards on it. It cannot really provide you additional performance than what is attached behind it. And in fact, it will always degrade what is attached behind it because it’s not storage, where as our USP is storage and it has a global cache and it has thousands of port connections, load balancing and all that. So our front end can enhance existing storage that sits behind it."
This is not the first time I have had to correct Hu and others of misperceptions of IBM's SAN Volume Controller (SVC). This month marks my four year "blogoversary", and I seem to spend a large portion of my blogging time setting the record straight. Here are just a few of my favorite posts setting the record straight on SVC back in 2007:
Since day 1, SAN Volume Controllers has focused primarily on external storage. Initially, the early models had just battery-protected DRAM cache memory, but the most recent model of the SVC, the 2145-CF8, adds support for internal SLC NAND flash solid state drives. To fully appreciate how SVC can help improve the performance of the disks that are managed, I need to use some visual aids.
In this first chart, we look at a 70/30/50 workload. This indicates that 70 percent of the IOPS are reads, 30 percent writes, and 50 percent can be satisfied as cache hits directly from the SVC. For the reads, this means that 50 percent are read-hits satisfied from SVC DRAM cache, and 50 percent are read-miss that have to get the data from the managed disk, either from the managed disk's own cache, or from the actual spinning drives inside that managed disk array.
For writes, all writes are cache-hits, but some of them will be destaged to the managed disk. Typically, we find that a third of writes are over-written before this happens, so only two-thirds are written down to managed disk.
In this example, the SVC reduced the burden of the managed disk from 100,000 IOPS down to 55,000, which is 35,000 reads and 20,000 writes. Some have argued against putting one level of cache (SVC) in front of another level of cache (managed disk arrays). However, CPU processor designers have long recognized the value of hierarchical cache with L1, L2, L3 and sometimes even L4 caches. The cache-hits on SVC are faster than most disk system's cache-hits.
This is a Ponder curve, mapping millisecond response (MSR) times for different levels of I/O per second, named after the IBM scientist John Ponder that created them. Most disk array vendors will publish similar curves for each of their products. In this case, we see that 100,000 IOPS would cause a 25 millisecond response (MSR) time, but when the load is reduced to 55,000 IOPS, the average response time drops to only 7 msec.
To be fair, the SVC does introduce 0.06 msec of additional latency on read-misses, so let's call this 7.06 msec. This tiny amount of latency could be what Hu Yoshida was referring to when he said there was "some performance degradation". There are other storage virtualization products in the market that do not provide caching to boost performance, but rather just map incoming requests to outgoing requests, and these can indeed slow down every I/O they process. Perhaps Hu was thinking of those instead of IBM's SVC when he made his comments.
Of course, not all workloads are 70/30/50, and not every disk array is driven to its maximum capability, so your mileage may vary. As we slide down the left of the curve where things are flatter, the improvement in performance lowers.
IOPS before SVC
IOPS after SVC
MSR before SVC
MSR after SVC
Hitachi's offerings, including the HDS USP-V, USP-VM and their recently announced Virtual Storage Platform (VSP) sold also by HP under the name P9500, have similar architecture to the SVC and can offer similar benefits, but oddly the Hitachi engineers have decided to treat externally attached storage as second-class citizens instead. Hu mentions data that "ages out or becomes less active we can move it to the external storage." IBM has chosen not to impose this "caste" system onto its design of the SAN Volume Controller.
The SVC has been around since 2003, before the USP-V came to market, and has sold over 20,000 SVC nodes over the past seven years. The SVC can indeed improve performance of managed disk systems, in some cases by a substantial amount. The 0.06 msec latency on read-miss requests represents less than 1 percent of total performance in production workloads. SVC nearly always improves performance, and in the worst case, provides same performance but with added functionality and flexibility. For the most part, the performance boost comes as a delightful surprise to most people who start using the SVC.
To learn more about IBM's upcoming products and how IBM will lead in storage this decade, register for next week's webcast "Taming the Information Explosion with IBM Storage" featuring Dan Galvan, IBM Vice President, and Steve Duplessie, Senior Analyst and Founder of Enterprise Storage Group (ESG).