Tony Pearson is a Master Inventor and Senior IT Architect for the IBM Storage product line at the
IBM Systems Client Experience Center in Tucson Arizona, and featured contributor
to IBM's developerWorks. In 2016, Tony celebrates his 30th year anniversary with IBM Storage. 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.
(Short URL for this blog: ibm.co/Pearson )
My books are available on Lulu.com! Order your copies today!
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 not a medical doctor, and this blog does not reference any IBM product or service that is intended for use in the diagnosis, treatment, cure, prevention or monitoring of a disease or medical condition, unless otherwise specified on individual posts.
A lot was announced yesterday, so I decided to break it up into several separate posts. This is part 2 in my 3-part series, focusing on: Storwize V7000 Unified, LTO-6 tape, and the SmartCloud Virtual Storage Center.
The Storwize V7000 Unified is a product that consists of a 2U-high Storwize V7000 control enclosure that provides block-based access, combined with two 2U-high File Modules that provide file-based NAS protocols: CIFS, NFS, HTTPS, SCP and FTP. The problem was that when it was introduced, it was based on Storwize V7000 v6.3, so when the Storwize V7000 v6.4 features were announced last June, they did not apply to the Storwize V7000 Unified.
That is all fixed now, so the Storwize V7000 Unified now supports the full v6.4 features, including Real-time Compression for both file and block-based access to primary data, and Fibre Channel over Ethernet (FCoE) for block access.
The two File Modules are no longer limited to a single Storwize V7000 control enclosure, you can now connect to up to four control enclosures clustered together. Combined with up to nine expansion enclosures for additional disk raises the total maximum to 960 drives.
If you don't already have an Active Directory or LDAP server, the Storwize V7000 Unified now offers an embedded LDAP server, for smaller deployments that want to reduce the number of servers they need to purchase for a complete solution.
Like the [IBM XIV Gen3 storage system], both the Storwize V7000 and V7000 Unified now also support the OpenStack Nova-volume interface.
Lastly, if you have a Storwize V7000 v6.4, you can upgrade it to a Storwize V7000 Unified by simply adding the two File Modules. This can be done in the field.
IBM LTO-6 for tape libraries and drives
IBM introduces the sixth generation of Linear Tape Open (LTO-6) drives, which can be used as stand-alone IBM TS1060 drives, or in IBM tape libraries. As with previous models of LTO, the LTO-6 can read two older generations (LTO-4 and LTO-5) tape media, and can write to previous generation (LTO-5) tape media. You can buy the LTO-6 drives now, and use the older media until LTO-6 tape cartridges are available (hopefully later this year!)
My friend, Brad Johns, from Brad Johns Consulting, has a great post on this [LTO-6 Announcement]. While you expect the new drives to be faster with a denser tape media format, the key advantage to the LTO-6 is that it improves the compression algorithm, from the previous 2:1 to the new 2.5:1 compression ratio:
Thus, with the improved compression, the LTO-6 is 40 percent faster, with double the tape cartridge density. This can reduce backup times by 30 percent, increase the amount of data that sits in your automated tape libraries, and reduce the courier costs sending tapes off-site.
IBM SmartCloud Virtual Storage Center v5.1
Last year, IBM coined the phrase "Storage Hypervisor" to refer to the underlying technology in the IBM SAN Volume Controller (SVC) and Storwize V7000 disk systems.
At the IBM Edge conference last June, my colleague Mike Griese presented [SmartCloud Virtual Storage Center]. Back then, it was a pilot program (beta test), and this week, IBM announces that it will be formally available as a product.
The idea was simple: take the basic storage hypervisor, and add the necessary software to make it a complete solution.
If all of your disk is currently virtualized behind IBM SAN Volume Controller (SVC), or you want to put all of your data behind SVC, then SmartCloud Virtual Storage Center is for you. Basically, for one per-TB price, you get all of the following:
The software features of SAN Volume Controller v6.4, including FlashCopy, Metro Mirror and Global Mirror.
The full advanced features of IBM Tivoli Storage Productivity Center v5.1, including the Storage Analytics Engine that does "Right-Tiering", recommending which LUNs should be moved entirely from one disk system to another, based on policies and access patterns.
IBM Tivoli Storage FlashCopy Manager v3.2 which manages FlashCopy with full coordination with applications, including Microsoft Exchange, SQL Server, DB2, Oracle, SAP, and VMware. This ensures that the FlashCopy destination copies are clean, eliminating the need to run backout or redo logs to correct any incomplete units of work.
If this combination sounds familiar, it was based on IBM's previous attempt called [Rapid Application Storage] which combined the Storwize V7000 with Tivoli Storage Productivity Center Midrange Edition and FlashCopy Manager.
The key difference is that SmartCloud VSC does not include the SVC hardware itself, you buy this separately. If you want Real-time Compression, that is charged separately for the subset of TB of the volumes that you select for compression.
Continuing my coverage of the [IBM System Storage Technical University 2011], 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, we had two: one focused on Tivoli Storage software, and the second to cover storage hardware. This year, we had two, one for System x called "Ask the eXperts", and one for System Storage called "Storage Free-for-All". This post covers the latter.
(Disclaimer: Do not shoot the messenger! We had a dozen or more experts on the panel, representing System Storage hardware, Tivoli Storage software, and Storage services. I took notes, trying to capture the essence of the questions, and the answers given by the various IBM experts. I have spelled out acronyms and provided links to relevant materials. The answers from individual IBMers may not reflect the official position of IBM management. Where appropriate, my own commentary will be in italics.)
You are in the wrong session! Go to "Ask the eXperts" session next door!
The TSM GUI sucks! Are there any plans to improve it?
Yes, we are aware that products like IBM XIV have raised the bar for what people expect from graphical user interfaces. We have plans to improve the TSM GUI. IBM's new GUI for the SAN Volume Controller and Storwize V7000 has been well-received, and will be used as a template for the GUIs of other storage hardware and software products. The GUI uses the latest HTML5, Dojo widgets and AJAX technologies, eliminating Java dependencies on the client browser.
Can we run the TSM Admin GUI from a non-Windows host?
IBM has plans to offer this. Most likely, this will be browser-based, so that any OS with a modern browser can be used.
As hard disk drives grow larger in capacity, RAID-5 becomes less viable. What is IBM doing to address this?
IBM is aware of this problem. IBM offers RAID-DP on the IBM N series, RAID-X on the IBM XIV, and RAID-6 on its other disk systems.
TPC licensing is outrageous! What is IBM going to do about it?
About 25 percent of DS8000 disk systems have SSD installed. Now that IBM DS8000 Easy Tier supports "any two" tiers, roughly 50 percent of DS8000 now have Easy Tier activated. No idea on how Easy Tier has been adopted on SVC or Storwize V7000.
We have an 8-node SVC cluster, should we put 8 SSD drives into a single node-pair, or spread them out?
We recommend putting a separate Solid-State Drive in each SVC node, with RAID-1 between nodes of a node-pair. By separating the SSD across I/O groups, you can reduce node-to-node traffic.
How well has SVC 6.2 been adopted?
The inventory call-home data is not yet available. The only SVC hardware model that does not support this level of software was the 2145-4F2 introduced in 2003. Every other model since then can be updated to this level.
Will IBM offer 600GB FDE drives for the IBM DS8700?
Currently, IBM offers 300GB and 450GB 15K RPM drives with the Full-Disk Encryption (FDE) capability for the DS8700, and 450GB and 600GB 10K RPM drives with FDE for the IBM DS8800. IBM is working with its disk suppliers to offer FDE on other disk capacities, and on SSD and NL-SAS drives as well, so that all can be used with IBM Easy Tier.
Is there a reason for the feature lag between the Easy Tier capabilities of the DS8000, and that of the SVC/Storwize V7000?
We have one team for Easy Tier, so they implement it first on DS8000, then port it over to SVC/Storwize V7000.
Does it even make sense to have separate storage tiers, especially when you factor in the cost of SVC and TPC to make it manageable?
It depends! We understand this is a trade-off between cost and complexity. Most data centers have three or more storage tiers already, so products like SVC can help simplify interoperability.
Are there best practices for combining SVC with DS8000? Can we share one DS8000 system across two or more SVC clusters?
Yes, you can share one DS8000 across multiple SVC clusters. DS8000 has auto-restripe, so consider having two big extent pools. The queue depth is 3 to 60, so aim to have up to 60 managed disks on your DS8000 assigned to SVC. The more managed disks the better.
The IBM System Storage Interopability Center (SSIC) site does not seem to be designed well for SAN Volume Controller.
Yes, we are aware of that. It was designed based on traditional Hardware Compatability Lists (HCL), but storage virtualization presents unique challenges.
How does the 24-hour learning period work for IBM Easy Tier? We have batch processing that runs from 2am to 8am on Sundays.
You can have Easy Tier monitor across this batch job window, and turn Easy Tier management between tiers on and off as needed.
Now that NetApp has acquired LSI, is the DS3000 still viable?
Yes, IBM has a strong OEM relationship with both NetApp and LSI, and this continues after the acquisition.
If have managed disks from a DS8000 multi-rank extent pool assigned to multiple SVC clusters, won't this affect performance?
Yes, possibly. Keep managed disks on seperate extent pools if this is a big concern. A PERL script is available to re-balance SVC striped volumes as needed after these changes.
Is the IBM [TPC Reporter] a replacement for IBM Tivoli Storage Productivity Center?
No, it is software, available at no additional charge, that provides additional reporting to those who have already licensed Tivoli Storage Productivity Center 4.1 and above. It will be updated as needed when new versions of Productivity Center are released.
We are experiencing lots of stability issues with SDD, SDD-PCM and SDD-DSM multipathing drivers. Are these getting the development attention they deserve?
IBM's direction is to shift toward native OS-based multipathing drivers.
Is anyone actually thinking of deploying public cloud storage in the near-term?
A few hands in the audience were raised.
None of the IBM storage devices seem to have [REST API]. Cloud storage providers are demanding this. What are IBM plans?
IBM plans to offer REST on SONAS. IBM uses SONAS internally for its own cloud storage offerings.
If you ask a DB2 specialist, an AIX specialist, and a System Storage specialist, on how to configure System p and System Storage for optimal performance, you get three different answers. Are there any IBMers who are cross-functional that can help?
Yes, for example, Earl Jew is an IBM Field Technical Support Specialist (FTSS) for both System p and Storage, and can help you with that.
Both Oracle and Microsoft recommend RAID-10 for their applications.
Don't listen to them. Feel free to use RAID-5, RAID-6 or RAID-X instead.
Resizing SVC source volumes forces ongoing FlashCopy or Metro Mirror relatiohships to be stopped. Does IBM plan to address this?
Currently, you have to stop, resize both source and target, then start the relationship again. Consider getting IBM Tivoli Storage Productivity Center for Replication (TPC-R).
IBM continues to support this for exising clients. For new deployments, IBM offers SONAS and the Information Archive (IA).
When will I be able to move SVC volumes between I/O groups?
You can today, but it is disruptive to the operating system. IBM is investigating making this less disruptive.
Will XIV ever support the mainframe?
It does already, with support for both Linux and z/VM today. For VSE support, use SVC with XIV. For those with the new zBX extension, XIV storage can be used with all of the POWER and x86-based operating systems supported. IBM has no plans to offer direct FICON attachment for z/OS or z/TPF.
Not a question - Kudos to the TSM and ProtecTIER team in supporting native IP-based replication!
When will IBM offer POWER-based models of the XIV, SVC and other storage devices?
IBM's decision to use industry-standard x86 technology has proven quite successful. However, IBM re-looks at this decision every so many years. Once again, the last iteration determined that it was not worth doing. A POWER-based model might not beat the price/performance of current x86 models, and maintaining two separate code bases would hinder development of new innovations.
We have both System i and System z, what is IBM doing to address the fact that PowerHA and GDPS are different?
IBM TPC-R has a service offering extension to support "IBM i" environments. GDPS plans to support multi-platform environments as well.
This was a great interactive session. I am glad everyone stayed late Thursday evening to participate in this discussion.
I always try to catch a session from Jim Blue, who works in our "SAN Central" center of competency team. This session was a long list of useful hints and tips, based on his many years of experience helping clients.
SAN Zoning works by inclusion, limiting the impact of failing devices. The best approach is to zone by individual initiator port. The default policy for your SAN zoning should be "deny".
Ports should be named to identify who, what, where and how.
While many people know not to mix both disk and tape devices on the same HBA, Jim also recommends not mixing dissimilar disks, test and production, FCP and FICON.
The sweet spot is FOUR paths. Too many paths can impact performance.
When making changes to redundant fabrics, make changes to the first fabric, then allow sufficient time before making the same changes to the other fabric.
Use software tools like Tivoli Storage Productivity Center (Standard Edition) to validate all changes to your SAN fabric.
Do not mix 62.5 and 50.0 micron technology.
Use port caps to disable inactive ports. In one amusing anecdote, he mention that an uncovered port was hit by sunlight every day, sending error messages that took a while to figure out.
Save your SAN configuration to non-SAN storage for backup
Consider firmware about two months old to be stable
Rule of thumb for estimating IOPS: 75-100 IOPS per 7200 RPM drive, 120-150 IOPS per 10K RPM drive, and 150-200 IOPS per 15K RPM drive.
Decide whether your shop is just-in-time or just-in-case provisioning. Just-in-time gets additional capacity on demand as needed, and just-in-case over-provisions to avoid scrambling last minute.
Avoid oversubscribing your inter-switch links (ISL). Aim for around 7:1 to 10:1 ratio.
Don't go cheap on bandwidth between sites for long-distance replication
Next Generation Network Fabrics - Strategy and Innovations
Mike Easterly, IBM Director of Global Field Marketing, presented IBM System Networking strategy, in light of IBM's recent acquisition of Blade Network Technologies (BNT). BNT is used in 350 of the Fortune 500 companies, and is ranked #2 behind Cisco in sales of non-core Ethernet switches (based on number of units sold).
Based on a recent survey, companies are upgrading their Ethernet networks for a variety of reasons:
56 percent for Live Partition Mobility and VMware Vmotion
45 percent for integrated compute stacks, like IBM CloudBurst
43 percent for private, public and hybrid cloud computing deployments
40 percent for network convergences
Many companies adopt a three-level approach, with core directors, distribution switches, and then access switches at the edge that connect servers and storage devices. IBM's BNT allows you to flatten the network to lower latency by collapsing the access and distribution levels into one.
IBM's strategy is to focus on BNT for the access/distribution level, and to continue its strategic partnerships for the core level.
IBM BNT provides better price/performance and lower energy consumption. To help with hot-aisle/cold-aisle rack deployments, IBM BNT provides both F and R models. F models have ports on the front, and R models have ports in the rear.
IBM BNT supports virtual fabric and HW-offload iSCSI traffic, and future-enabled for FCoE. Support for TRILL (transparent interconnect of lots of links) and OpenFlow will be implemented through software updates to the switches.
While Cisco Nexus 1000v is focused on VMware Enterprise Plus, IBM BNT's VMready works with VMware, Hyper-V, Linux KVM, XEN, OracleVM, and PowerVM. This allows single pane of management of VMready and ESX vSwitches.
In preparation for Converged Enhanced Ethernet (CEE), IBM BNT will provide full 40GbE support sometime next year, and offer switches that support 100GbE uplinks. IBM offers extended length cables, including passive SFP+ DAC at 8.5 meters, and 10Gbase-T Cat7 cables up to 100 meters.
Inter-datacenter Workload Mobility with VMware vSphere and SAN Volume Controller (SVC)
This session was co-presented between Bill Wiegand, IBM Advanced Technical Services, and Rawley Burbridge, IBM VMware and midrange storage consultant. IBM is the leader in storage virtualization product (SVC), and is the leading reseller of VMware.
Like MetroCluster on IBM N series, or EMC's VPLEX Metro, the IBM SAN Volume Controller can support a stretched cluster across distance that allows virtual machines to move seamlessly from one datacenter to another. This is a feature IBM introduced with SVC 5.1 back in 2009. This can be used for PowerVM Live Partition Mobility, VMware vMotion, and Hyper-V Quick Migration.
SVC stretched cluster can help with both Disaster Avoidance and Disaster Recovery. For Disaster Avoidance, in anticipation of an outage, VMs can be moved to the secondary datacenter. For Disaster Recover, additional automation, such as VMware High Availability (HA) is needed to restart the VMs at the secondary datacenter.
IBM stretched cluster is further improved with a feature called Volume Mirroring (formerly vDisk Mirroring) which creates two physical copies of one logical volume. To the VMware ESX hosts, there is only one volume, regardless of which datacenter it is in. The two physical copies can be on any kind of managed disk, as there is no requirement or dependency of copy services on the back-end storage arrays.
Another recent improvement is the idea of spreading the three quorum disks to three different locations or "failure domains". One in each data center, and a third one in a separate building, somewhere in between the other two, perhaps.
Of course, there are regional disasters that could affect both datacenters. For this reason, SVC stretched cluster volumes can be replicated to a third location up to 8000 km away. This can be done with any back-end disk arrays, as again there is not requirement for copy services from the managed devices. SVC takes care of it all.
Networking is going to be very important for a variety of transformational projects going forward in the next five years.
Every January, we look back into the past as well as look into the future for trends to watch for the upcoming year. Ray Lucchesi of Silverton Consulting has a great post looking back at the [Top 10 storage technologies over the last decade]. I am glad to see that IBM has been involved with and instrumental in all ten technologies.
Looking into the future, Mark Cox of eChannel has an article [Storage Trends to Watch in 2011], based on his interviews with two fellow IBM executives: Steve Wojtowecz, VP of storage software development, and Clod Barrera, distinguished engineer and CTO for storage. Let's review the four key trends:
Cloud Storage and Cloud Computing
No question: Cloud Computing will be the battleground of the IT industry this decade. I am amused by the latest spate of Microsoft commercials where problems are solved with someone saying "...to the cloud". Riding on the coat tails of this is "Cloud Storage", the ability to store data across an Internet Protocol (IP) network, such as 10GbE Ethernet, in support of Cloud Computing applications. Cloud Storage protocols in the running include NFS, CIFS, iSCSI and FCoE.
Mark writes "..vendors who aren't investing in cloud storage solutions will fall behind the curve."
Economic Downturn forces Innovation
The old British adage applies: "Necessity is the mother of invention." The status quo won't do. In these difficult economic times, IT departments are running on constrained budgets and staff. This forces people to evaluate innovative technologies for storage efficiency like real-time compression and data deduplication to make better use of what they currently have. It also is forcing people to take a "good enough" attitude, instead of paying premium prices for best-of-breed they don't really need and can't really afford.
IT Service Management
Companies are getting away from managing individual pieces of IT kit, and are focusing instead on the delivery of information, from the magnetic surface of disk and tape media, to the eyes and ears of the end users. The deployment mix of private, hybrid and public clouds makes this even more important to measure and manage IT as a set of services that are delivered to the business. IT Service Management software can be the glue, helping companies implement ITIL v3 best practices and management disciplines.
Smarter Data Placement
A recent survey by "The Info Pro" analysts indicates that "managing storage growth" is considered more critical than "managing storage costs" or "managing storage complexity".
This tells me that companies are willing to spend a bit extra to deploy a tiered information infrastructure if it will help them manage storage growth, which typically ranges around 40 to 60 percent per year. While I have discussed the concept of "Information Lifecycle Management" (ILM), for the past four years on this blog, I am glad to see it has gone mainstream, helped in part with automated storage tiering features like IBM System Storage Easy Tier feature on the IBM DS8000, SAN Volume Controller and Storwize V7000 disk systems. Not all data is created equal, so the smart placement of data, based on the business value of the information contained, makes a lot of sense.
These trends are influencing what solutions the various different vendors will offer, and will influence what companies purchase and deploy.
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).
"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].
Over on the Tivoli Storage Blog, there is an exchange over the concept of a "Storage Hypervisor". This started with fellow IBMer Ron Riffe's blog post [Enabling Private IT for Storage Cloud -- Part I], with a promise to provide parts 2 and 3 in the next few weeks. Here's an excerpt:
"Storage resources are virtualized. Do you remember back when applications ran on machines that really were physical servers (all that “physical” stuff that kept everything in one place and slowed all your processes down)? Most folks are rapidly putting those days behind them.
In August, Gartner published a paper [Use Heterogeneous Storage Virtualization as a Bridge to the Cloud] that observed “Heterogeneous storage virtualization devices can consolidate a diverse storage infrastructure around a common access, management and provisioning point, and offer a bridge from traditional storage infrastructures to a private cloud storage environment” (there’s that “cloud” language). So, if I’m going to use a storage hypervisor as a first step toward cloud enabling my private storage environment, what differences should I expect? (good question, we get that one all the time!)
The basic idea behind hypervisors (server or storage) is that they allow you to gather up physical resources into a pool, and then consume virtual slices of that pool until it’s all gone (this is how you get the really high utilization). The kicker comes from being able to non-disruptively move those slices around. In the case of a storage hypervisor, you can move a slice (or virtual volume) from tier to tier, from vendor to vendor, and now, from site to site all while the applications are online and accessing the data. This opens up all kinds of use cases that have been described as “cloud”. One of the coolest is inter-site application migration.
A good storage hypervisor helps you be smart.
Application owners come to you for storage capacity because you’re responsible for the storage at your company. In the old days, if they requested 500GB of capacity, you allocated 500GB off of some tier-1 physical array – and there it sat. But then you discovered storage hypervisors! Now you tell that application owner he has 500GB of capacity… What he really has is a 500GB virtual volume that is thin provisioned, compressed, and backed by lower-tier disks. When he has a few data blocks that get really hot, the storage hypervisor dynamically moves just those blocks to higher tier storage like SSD’s. His virtual disk can be accessed anywhere across vendors, tiers and even datacenters. And in the background you have changed the vendor storage he is actually sitting on twice because you found a better supplier. But he doesn’t know any of this because he only sees the 500GB virtual volume you gave him. It’s 'in the cloud'."
"Let’s start with a quick walk down memory lane. Do you remember what your data protection environment looked like before virtualization? There was a server with an operating system and an application… and that thing had a backup agent on it to capture backup copies and send them someplace (most likely over an IP network) for safe keeping. It worked, but it took a lot of time to deploy and maintain all the agents, a lot of bandwidth to transmit the data, and a lot of disk or tapes to store it all. The topic of data protection has modernized quite a bit since then.
Fast forward to today. Modernization has come from three different sources – the server hypervisor, the storage hypervisor and the unified recovery manager. The end result is a data protection environment that captures all the data it needs in one coordinated snapshot action, efficiently stores those snapshots, and provides for recovery of just about any slice of data you could want. It’s quite the beautiful thing."
At this point, you might scratch your head and ask "Does this Storage Hypervisor exist, or is this just a theoretical exercise?" The answer of course is "Yes, it does exist!" Just like VMware offers vSphere and vCenter, IBM offers block-level disk virtualization through the SAN Volume Controller(SVC) and Storwize V7000 products, with a full management support from Tivoli Storage Productivity Center Standard Edition.
SVC has supported every release of VMware since the 2.5 version. IBM is the leading reseller of VMware, so it makes sense for IBM and VMware development to collaborate and make sure all the products run smoothly together. SVC presents volumes that can be formatted for VMFS file system to hold your VMDK files, accessible via FCP protocol. IBM and VMware have some key synergies:
Management integration with Tivoli Storage Productivity Center and VMware vCenter plug-in
VAAI support: Hardware-assisted locking, hardware-assisted zeroing, and hardware-assisted copying. Some of the competitors, like EMC VPLEX, don't have this!
Space-efficient FlashCopy. Let's say you need 250 VM images, all running a particular level of Windows. A boot volume of 20GB each would consume 5000GB (5 TB) of capacity. Instead, create a Golden Master volume. Then, take 249 copies with space-efficient FlashCopy, which only consumes space for the modified portions of the new volumes. For each copy, make the necessary changes like unique hostname and IP address, changing only a few blocks of data each. The end result? 250 unique VM boot volumes in less than 25GB of space, a 200:1 reduction!
Support for VMware's Site Recovery Manager using SVC's Metro Mirror or Global Mirror features for remote-distance replication.
Data center federation. SVC allows you to seamlessly do vMotion from one datacenter to another using its "stretched cluster" capability. Basically, SVC makes a single image of the volume available to both locations, and stores two physical copies, one in each location. You can lose either datacenter and still have uninterrupted access to your data. VMware's HA or Fault Tolerance features can kick in, same as usual.
But unlike tools that work only with VMware, IBM's storage hypervisor works with a variety of server virtualization technologies, including Microsoft Hyper-V, Xen, OracleVM, Linux KVM, PowerVM, z/VM and PR/SM. This is important, as a recent poll on the Hot Aisle blog indicates that [44 percent run 2 or more server hypervisors]!
Join the conversation! The virtual dialogue on this topic will continue in a [live group chat] this Friday, September 23, 2011 from 12 noon to 1pm EDT. Join me and about 20 other top storage bloggers, key industry analysts and IBM Storage subject matter experts to discuss storage hypervisors and get questions answered about improving your private storage environment.
Continuing my coverage of the [Data Center 2010 conference], Tuesday afternoon I presented "Choosing the Right Storage for your Server Virtualization". In 2008 and 2009, I attended this conference as a blogger only, but this time I was also a presenter.
The conference asked vendors to condense their presentations down to 20 minutes. I am sure this was inspired by the popular 18-minute lectures from the [TED conference] or perhaps the [Pecha Kucha] night gatherings in Japan where each presenter speaks while showing 20 slides for 20 seconds each, This forces the presenters to focus on their key points and not fill the time slot with unnecessary marketing fluff. This also allows more vendors to have a chance to pitch their point of view.
Jim is an IBM Fellow for IBM Systems and Technology Group. There are only 73 IBM Fellows currently working for IBM, and this is the highest honor IBM can bestow on an employee. He has been working with IBM since 1968.
He is tasked with predicting the future of IT, and help drive strategic direction for IBM. Cost pressures, requirements for growth, accelerating innovation and changing business needs help influence this direction.
IBM's approach is to integrate four different "IT building blocks":
Scale-up Systems, like the IBM System Storage DS8000 and TS3500 Tape Library
Resource Pools, such as IBM Storage Pools formed from managed disks by IBM SAN Volume Controller (SVC)
Integrated stacks and appliances, integrated software and hardware stacks, from Storwize V7000 to full rack systems like IBM Smart Analytics Server or CloudBurst.
Mobility of workloads and resources requires unified end-to-end service management. Fortunately, IBM is the #1 leader in IT Service Management solutions.
Jim addressed three myths:
Myth 1: IT Infrastructures will be homogenous.
Jim feels that innovations are happening too rapidly for this to ever happen, and is not a desirable end-goal. Instead, a focus to find the right balance of the IT building blocks might be a better approach.
Myth 2: All of your problems can be solved by replacing everything with product X.
Jim feels that the days of "rip-and-replace" are fading away. As IBM Executive Steve Mills said, "It isn't about the next new thing, but how well new things integrate with established applications and processes."
Myth 3: All IT will move to the Cloud model.
Jim feels a substantial portion of IT will move to the Cloud, but not all of it. There will always be exceptions where the old traditional ways of doing things might be appropriate. Clouds are just one of the many building blocks to choose from.
Jim's focus lately has been finding new ways to take advantage of virtualization concepts. Server, storage and network virtualization are helping address these challenges through four key methods:
Sharing - virtualization that allows a single resource to be used by multiple users. For example, hypervisors allow several guest VM operating systems share common hardware on a single physical server.
Aggregation - virtualization that allows multiple resources to be managed as a single pool. For example, SAN Volume Controller can virtualize the storage of multiple disk arrays and create a single storage pool.
Emulation - virtualization that allows one set of resources to look and feel like a different set of resources. Some hypervisors can emulate different kinds of CPU processors, for example.
Insulation - virtualization that hides the complexity from the end-user application or other higher levels of infrastructure, making it easier to make changes of the underlying managed resources. For example, both SONAS and SAN Volume Controller allow disk capacity to be removed and replaced without disruption to the application.
In today's economy, IT transformation costs must be low enough to yield near-term benefits. The long-term benefits are real, but near-term benefits are needed for projects to get started.
What set's IBM ahead of the pack? Here was Jim's list:
100 Years of Innovation, including being the U.S. Patent leader for the last 18 years in a row
IBM's huge investment in IBM Research, with labs all over the globe
Leadership products in a broad portfolio
Workload-optimized designs with integration from middleware all the way down to underlying hardware
Comprehensive management software for IBM and non-IBM equipment
Clod is an IBM Distinguished Engineer and Chief Technical Strategist for IBM System Storage. His presentation focused on trends and directions in the IT storage industry. Clod started with five workload categories:
To address these unique workload categories, IBM will offer workload-optimized systems. The four drivers on the design for these are performance, efficiency, scalability, and integration. For example, to address performance, companies can adopt Solid-State Drives (SSD). Unfortunately, these are 20 times more expensive dollar-per-GB than spinning disk, and the complexity involved in deciding what data to place on SSD was daunting. IBM solved this with an elegant solution called IBM System Storage Easy Tier, which provides automated data tiering for IBM DS8000, SAN Volume Controller (SVC) and Storwize V7000.
For scalability, IBM has adopted Scale-Out architectures, as seen in the XIV, SVC, and SONAS. SONAS is based on the highly scalable IBM General Parallel File System (GPFS). File systems are like wine, they get better with age. GPFS was introduced 15 years ago, and is more mature than many of the other "scalable file systems" from our competition.
Areal Density advancements on Hard Disk Drives (HDD) are slowing down. During the 1990s, the IT industry enjoyed 60 to 100 percent annual improvement in areal density (bits per square inch). In the 2000s, this dropped to 25 to 40 percent, as engineers are starting to hit various physical limitations.
Storage Efficiency features like compression have been around for a while, but are being deployed in new ways. For example, IBM invented WAN compression needed for Mainframe HASP. WAN compression became industry standard. Then IBM introduced compression on tape, and now compression on tape is an industry standard. ProtecTIER and Information Archive are able to combine compression with data deduplication to store backups and archive copies. Lastly, IBM now offers compression on primary data, through the IBM Real-Time Compression appliance.
For the rest of this decade, IBM predicts that tape will continue to enjoy (at least) 10 times lower dollar-per-GB than the least expensive spinning disk. Disk and Tape share common technologies, so all of the R&D investment for these products apply to both types of storage media.
For integration, IBM is leading the effort to help companies converge their SAN and LAN networks. By 2015, Clod predicts that there will be more FCoE purchased than FCP. IBM is also driving integration between hypervisors and storage virtualization. For example, IBM already supports VMware API for Array Integration (VAAI) in various storage products, including XIV, SVC and Storwize V7000.
Lastly, Clod could not finish a presentation without mentioning Cloud Computing. Cloud storage is expected to grow 32 percent CAGR from year 2010 to 2015. Roughly 10 percent of all servers and storage will be in some type of cloud by 2015.
As is often the case, I am torn between getting short posts out in a timely manner versus spending some more time to improve the length and quality of information, but posted much later. I will spread out the blog posts in consumable amounts throughout the next week or two, to achieve this balance.
Earlier this year, I wrote a Web article titled [Data Footprint Reduction] which covered data deduplication and compression, and was asked to present this at IBM Edge. I have expanded it to include:
Space-Efficient Point-in-Time copies
After I presented the basic concepts, Sanjay Bhikot, a Unix and Storage admin at RICOH, presented his real-world experiences with data deduplication using the IBM ProtecTIER and real-time compression Beta experience using the SAN Volume Controller (SVC).
IBM Active Cloud Engine Implementation on IBM SONAS 1.3 and IBM Storwize V7000 Unified
John Sing (IBM) presented the latest enhancements in the v1.3.2 release of SONAS and Storwize V7000 Unified.
Introducing VMware vSphere Storage Features
Fellow blogger Stephen Foskett presented this session on VMware's storage features. This included VMware APIs for Array Integration (VAAI), VMware Array Storage Awareness (VASA), vCenter plug-ins, and a new concept he called "vVol" which de-multiplexes the "I/O Blender" that server hypervisors do by tagging individual requests to individual OS guests to provide added benefit. IBM is a leading reseller of VMware, so it makes sense that most of our storage meets all of Steve's requirements for recommendation.
IBM's Storage Strategy in the Smarter Computing Era
Last year, I presented this on the fourth day of the conference, and feedback we received from attendees was that this should have been presented sooner in the week, as it provides great context for the more detailed product presentations.
To address this concern, the IBM executives presented IBM strategy on Monday's keynote session, but allowed me to present this on Wednesday for several reasons:
You may have missed the keynote session. For example, you may not have arrived in time to hear the executives speak due to weather or mechanical problems causing travel delays.
You may have attended the keynote session, but want to hear it again. Maybe you were a bit hung-over, or just may have been overwhelmed with the size and scope of this event. I have read for strategic topics, audiences may have to hear the message five to seven times before they truly appreciate and understand it.
You may want to ask questions, and explore the implications in more detail. While keynote sessions can reach a broader audience, the communication is very much uni-directional. With break-out sessions with a few hundred people, the venue is more intimate and can afford opportunties for information exchange.
The title of this session rolls off the tongue nicely, much like "James and the Giant Peach", "Harold and the Purple Crayon", or "Charlie and the Chocolate Factory".
When people say they are interested in "Cloud Storage", what exactly do they mean. After discussions with hundreds of clients, IBM has worked out a "taxonomy" that identifies four distinct types of storage:
In this session, I presented how IBM SONAS addresses all four of these categories, as well as other IBM storage products that can address specific categories in the taxonomy.
In the evening, the attendees at IBM Edge joined the attendees from Innovate2012 (focused on IBM Rational products) at SeaWorld, with BBQ dinner, rides, Shamu the whale show, and a concert featuring Foreigner!
This week I am in Orlando, Florida for the IBM Edge conference. This is the last day, so it ends early for people who want to get home to their datacenters (er.. families) for the weekend.
How Real-Time Compression Can Maximize Storage Efficiency for Production Applications
This was a split session with two speakers. First, Ian Rimmer, Senior IT Engineer and Architect at iBurst, presented their experience with the IBM Real-Time Compression Appliance in front of NetApp NAS storage arrays. Second, Jerry Haigh, IBM offering manager for IBM System Storage, presented the new Real-Time compression feature announced this week on IBM SAN Volume Controller (SVC) and Storwize V7000.
iBurst is the #1 Wireless Telecom for South Africa. The also offer cable broadband and VOIP. They have 200 employees servicing 120,000 subscriber/households. They need to keep five years' worth of text files, and have chosen real-time compression of their NAS storage. This was before IBM acquired the Storwize company, as they have been using it for the past six years.
The monetary savings from compression was used to purchase Performance Accelerator Modules (PAM) cards for their NetApp NAS gear, which benefit from the compression (more data stored in SSD to improve performance).
For backup, they use NDMP with Symantec NetBackup that keeps data in its compressed form as it is written to tape. They have an IBM TS3100 library with LTO tape as the backup repository.
Jerry Haigh presented Real-Time compression for primary disk data. Unlike the competition, this is designed to be used with primary data, including databases, and does this real-time, not post-process. In some performance tests, DB2 compressed on 48 drives out-performed the same data uncompressed on 96 drives. In another test focused on VMware Vmark benchmark, the compressed data was able to be same or better performance as uncompressed. In a third test with SVC virtualizing XIV running Oracle ORION test, the Oracle databases compressed 50 to 64 percent, and had better performance.
For those who already have SVC or Storwize V7000, consider a 45-day trial to check out compression for yourself.
NAS File Systems: Access and Authentication
Mark Taylor, IBM Technical Specialist for SONAS, N series and Storwize V7000 Unified, presented the nuances of authentication and authorization for NAS file systems. The differences between these two are:
Authentication - Yes, you are who you are.
Authorization - Yes, you are permitted to do what you are trying to do
(Prior to working with SONAS, my only experience with access and authentication in NAS was setting up my LAN at home, which I have connecting my Mac, Linux and Windows machines. I have both N series and SONAS at the IBM Executive Briefing Center in Tucson, Arizona, so I know first-hand how complicated NAS access and authentication systems can be.
A few months ago, I taught "Intro to NAS" as one of my topics at the Top Gun class in Argentina and Brazil. Several of the students had mentioned they thought they knew NAS solutions but had not realized all the technical issues with access and authentication that I discussed in my presentation.)
Mark explained the differences between Windows NTFS-style System identifiers (SID), versus UNIX-style user and group identifiers (UID, GID). For NAS solutions that support both CIFS and NFS, there are four options:
Microsoft Active Director (AD) extended with Identity Management for UNIX, formerly known as Services for UNIX (SFU). AD servers normally store SID information, but the extensions add extra columns to hold UID/GID mappings.
AD with Network Information Service (NIS) server. The problem with this approach is that AD and NIS are separate databases, and you need to coordinate updates to them, and their backups.
Lightweight Directory Access Protocol (LDAP) with SAMBA extensions. LDAP holds UID/GID information, and the SAMBA extensions adds extra columns to hold SID mapping.
Local mapping. The dangerous part of local mapping is that the storage admin is also the security admin, and you may want different people doing these roles.
Of these four methods, Mark recommends the first and third as best practices for multi-protocol authentication.
SID-to-UID mapping, UID-to-SID mapping
SONAS and Storwize V7000
SID-to-UID/GID mapping, NFS v4 ACLs
NFS v4 ACLs
Mark then explained how NFS v4 ACLs work, basically an ordered collection of "Access Control Elements" or ACEs. Each ACE on the ACL may "allow" or "deny" the request. You want to avoid "Inheritance" as that can cause problems and unxpected results.
That's it folks. Next week, I am spending time with my research buddies at the Almaden Research Center near San Jose, California, and then it is off to Moscow, Russia to kick off a series of IBM events called "Edge Comes to You" (ECTY).
The ECTY conferences will be a smaller subset of the Edge conference here in Orlando, but offered in other countries for those who were unable to travel to the United States.
This week I am in Orlando, Florida for the IBM Edge conference. Thursday evening after all the other sessions, we had a Free-for-All, a Q&A panel across all storage topics, moderated by Scott Drummond. The conference officially ends at noon tomorrow, but for many, this is the last session, as people fly out Friday morning. Here are the questions and the panel responses during the session.
When will IBM unify their storage management between Mainframe z/OS and the distributed systems platforms?
IBM offers a Change and Configuration Management Data Base (CCMDB) for this purpose with appropriate collectors from z/OS and distributed systems, but hasn't sold well.
When will IBM devices have RESTful interfaces?
Both IBM Systems Director and IBM Tivoli Storage Productivity Center (TPC) offer RESTful APIs. IBM Systems Director can manage z/VM and Linux on System z, as well as Power Systems and x86 based distributed systems. Since October 2008, IBM's Project Zero introduced RESTful interfaces to PHP and Groovy software running on WebSphere sMash environments. We have not heard much about this since 2008.
Will IBM TPC support NPIV on Power Systems?
TPC 5.1 has toleration support for this, showing the first port connection discovered, but not all connections, and we expect to retrofit this toleration to TPC 4.2.2 Fixpack 2. Hopefully, we will have full support in a future release.
We would like TPC for Replication to run on Linux for System z. We do not run z/OS at the disaster recovery site location.
Submit an IBM Request for Enhancement [RFE] for this. We have TPC for Replication on z/OS, as well as the distributed systems version that runs on Windows, Linux and AIX.
We have enhancements we would like to see for XIV and SONAS also, can we use the RFE process for this also?
Yes, submit the requirements for our review.
We heard the Statement of Direction that there would be storage integrated into the PureSystems. What exactly does that mean?
The PureSystems family of expert-integrated systems is based on a new chassis that has a front part, a midplane, and a back-part. All IBM System Storage products that support x86 and Power Systems can work with PureSystems. However, IBM does not yet offer storage that fits in the front part of the PureFlex chassis, but the Statement of Direction indicates that we intend to offer that option. Until then, the IBM Storwize V7000 is the storage of choice that can be put into the PureSystems rack, but outside the individual chasses.
We see some features like Real-Time Compression being put into the SAN Volume Controller (SVC), and other features put into the back-end devices. How are we supposed to make sense of this?
IBM's new pilot program, the SmartCloud Virtual Storage Center, to bring these all together. In general, we have design teams of system architects that determine which features go in which products, and prioritize accordingly.
We heard the IBM Executives during the opening session indicate that IBM's strategy involves supporting Big Data, but I haven't seen any storage that supports native Hadoop interfaces. Did I miss something?
First, I want to emphasize that Big Data is more than just MapReduce workloads. IBM offers Streams and BigInsights software to handle text, as well as Business Intelligence and Data Warehouse solutions for structured data. IBM's General Parallel File System (GPFS) has a Shared-Nothing-Cluster (SNC) mode with Hadoop interfaces that runs twice as fast as Hadoop's native HDFS file system. The storage products we recommend for Big Data are the SONAS and the DCS3700 disk systems, as both are optimized for the sequential workloads Big Data represents.
Everytime we upgrade our SVC, we review the list for SDDPCM multi-pathing and see that we need to upgrade our back-end DS8000 microcode up to recommended levels. Can we get a list of combinations that work from other customers?
The advantage of storage hypervisors like SVC is that we can separate the multi-pathing driver from the back-end managed disk systems. You only need the SDDPCM to support the SVC, not the back-end devices. For the most part, SVC has not dropped support for any level of previously supported OS or multi-pathing software.
On SVC, when we migrate volumes (vDisks) from one storage pool to another, we would like to throttle this process during FlashCopy.
Yes, we had several requests like this, which is why we now recommend using Volume Mirorring to perform migrations. In fact the GUI wizard uses Volume Mirroring by default when migrations are performed. As for throttling, IBM has implemented "I/O Priority Manager" that offers Quality of Service classes for DS8000 and XIV Gen3, and might consider porting this to other products in our portfolio.
Sizing systems is an art. I just need to know if the DS8000 is running hot. Can we have the equivalent of "red lines" for our disk systems similar to automobile engines?
Storage Optimizer was added to TPC 4.2 to help in this area, identifying heat-maps for IBM DS8000, DS6000, DS5000, DS4000, SVC and Storwize V7000. We recommend you look at the performance violation reports.
How can we evaluate the characteristics of our workloads?
Yes, TPC can do this.
When we are replacing non-IBM storage with IBM, we don't have good tools to evaluate the non-IBM equipment. What is IBM doing for this?
IBM's Disk Magic modeling tool can take inputs from a variety of sources, including iostat from the servers themselves. You can also install a 90-day trial of TPC to help with this.
We really like EMC's "Grab" program, does IBM have one also?
Updating the Host Attachment Kit (HAK) for AIX is quite painful for the SVC. We prefer the method employed for the XIV.
Thanks for the feedback.
For SVC, we need to correlate disk with VMware and VIOS. Can we get vSCSI information on VIOS?
TPC 5.1 has this support, and we believe it has been retrofitted to TPC 4.2.2 Fixpack 2, coming out this month.
Currently, with SVC, when volumes are part of a Global Mirror (GM) session, we need to cancel GM, expand the source volume, expand the target volume, then restart GM. We would like this to be fully automated and non-disruptive.
Sounds like a great requirement to submit for the RFE process.
Can we get an RSS Feed for the RFE community.
Yes, you can subscribe to it. You can also set up "Watch Lists".
Thanks to all of the IBM experts on the panel for their participation at this event!
Maria Boonie is the IBM Director for IBM Worldwide Training and Technical Conferences. She indicated that there were 1500 attendees this week crossing both the System Storage and System x conferences at this hotel. There are 35 vendors that have sponsored this event, and they will be at the "Solutions Center" being held Monday through Wednesday this week.
She took this opportunity to plug IBM's latest education offerings, including Guaranteed-to-Run implementation classes, and Instructor-Led Online (ILO) technical classes.
Brian Truskowski is IBM General Manager for System Storage and Networking. I used to directly report to him in a previous role, and a few years ago he used to be the IBM CIO that helped with IBM's internal IT transformation.
Brian indicates that the previous approach to growth was to "Just Buy More", but this has some unintended consequences. He argued that companies need to adopt one or more of the following approaches to growth:
Stop storing so much - reduce data footprint using storage efficiency capabilities like data deduplication and compression
Store more with what is already on the floor - improve storage utilization with technologies like storage virtualization and thin provisioning
Move data to the right place - implement automated tiering, such as "Flash & Stash" between Solid-state drives and spinning disk, and/or Information Lifecycle Management (ILM) between disk and tape. Studies at some clients have found over 70 percent of data has not beed touched in the last 90 days
This time of dramatic change is the result of a "perfect storm" of influences, including the rising costs and risks associated with losing data, the increased need to index and search data, the desire for "Business Analytics", and the expectation for 100 percent up-time. This is driving IBM to offer hyper-efficient backup, Continuous Data Availability, and Smart Archive solutions.
The case study of SPRINT is a good example. SPRINT is a Telecommunications provider for cell phone users. They were challenged with 35 percent utilization, 165 storage arrays from six different vendors, and an expected 100 percent increase in their IT maintenance costs. After implementing IBM SAN Volume Controller (SVC) and Tivoli Storage Productivity Center (TPC) to manager 2.9 PB of data, SPRINT increased their utilization to 82 percent, reduced down to 70 storage arrays from only three vendors, and reduced their maintenance costs by 57 percent. Today, SPRINT now manages over 5 PB of data with SVC and TPC, have reduced their power and cooling by 3.5 million KWh, representing $320,000 USD in savings.
Roland Hagan is the IBM Vice President for the System x server platform. He talked about the "IT Conundrum" that represents a vicious cycle of "IT Sprawl", "Untrusted Data" and "Inflexible IT" that seem to feed each other. IBM is trying to change behavior, from thinking and dealing with physical boxes representing servers, storage and network gear, to a more holistic view focused on workloads, shared resource pools, independent scaling, and automated management.
IBM is leading the server marketplace, in part because of clever things IBM is doing, especially in developing the eX5 chipset that surrounds x86 commondity processors, and in part because of actions or decisions the competition have taken:
It doesn't break IBM's heart that Oracle decided to drop software support of their database on Itanium, which focued entirley against HP. Oracle runs on IBM servers better than Oracle/Sun or HP servers today, so it does not impact us, other than IBM has had a lot of people leaving HP to switch over to IBM.
HP has taken on a new CEO and reduced their R&D budget, causing them to be late-to-market on some of their offerings.
Dell continues to focus on the small and medium sized customer, and have not really broken into the "Enterprise".
Newcomer Cisco has some great technology that only seems to be adoptable in "Green Field" situations, as it does not integrate well with existing data center infrastructures.
The combination of ex5 chip-set architecture, Max5 memory expansion capabilities and Virtual Network Interface Cards (NICs), provide for a very VM-aware platform. For those who are not ready to fully adopt an integrated stack like IBM CloudBurst, IBM offers the Tivoli Service Automation software on its own, and a new [IBM BladeCenter Foundation for Cloud] as stepping stones to get there.
There are certainly more attendees here than last year, which reflects either the change in location (Orlando, Florida rather than Washington DC) as well as the economic recovery. I'm looking forward to an excellent week!
IBM Tivoli Storage Productivity Center v4.2.2 Overview and Update
This was an updated version of the presentation I gave last July in Orlando, Florida (see my post [IBM Storage University - Day 1]). Since it might have been awhile since the Australian audience had heard about the latest and greatest for Tivoli Storage Productivity Center, I decided to cover the enhancements of 4.2.0, 4.2.1 and 4.2.1 combined.
IBM Tivoli Storage Productivity Center is an important part of IBM's "Storage Hypervisor" solution, combining a single pane of glass for management with non-disruptive storage virtualization with SVC and Storwize V7000.
IBM Storwize V7000 and SVC integration with VMware
Alexi Giral from IBM Sydney presented this session on how Storwize V7000 and SVC serve as the "Storage Hypervisor" for VMware server virtualization environments. The focus was on the FCP and iSCSI block-only access modes of these devices, although one could use IBM Storwize V7000 Unified to provide NFS file-level access to VMware. Alexi covered both VMware Vsphere v4 and v5, as there are a few differences.
IBM Storwize V7000 and SVC supports thin provisioning, VMware's VAAI interface, VMware's Site Recovery Manager, and provides a storage management plug-in to Vmware's vCenter. The SVC has extended the distance for split-cluster configurations that support VMware's vMotion live partition mobility and High Availiability (HA) up to 300km using active DWDM.
Tape Storage Reinvented: What's New and Exciting in the Tape World?
Special thanks to Jim Fisher and Jim Karp for providing me this presentation, videos and supporting materials for me to present this session. I gave this as the first break-out session on Tuesday, and then repeated as the last break-out session on Thursday. Several of the attendees in the audience mocked my title, with taunts like "What could be NEW or EXCITING about tape?" I covered four key areas:
The new TS1140 tape drive, including the corresponding model-JC tape that holds 4TB native (12 TB compressed!).
The enhanced TS3500 with the Tape Library Connector Shuttle. I had a video that shows how tapes can be sent from one TS3500 tape library string to another.
The new Linear Tape File System (LTFS), both the single drive edition and the library edition
The new 3592-C07 FICON controller for our mainframe clients
By the end of the session, the folks that taunted me were honestly impressed that they learned a few things, and had not realized so much has been developed recently in the world of tape.
Down the street, in Times Square, IBM made it on the big board.
Continuous Data Availability
Jeanine Cotter, IBM VP for Data Center Services, started out with a video about Sabre. IBM developed this revolutionary airline reservation system to handle the huge volume of transactions. Today, 18 percent of organizations consider downtime unacceptable for their tier-1 applications, and 53 percent would be seriously impacted by an outage lasting an hour or more.
Eventually, companies cross the "Continuous Availability" threshold, the point where they discover that the possibility of downtime is too costly to ignore. IBM has clients using 3-site Metro/Global Mirror that can fail-over an entire data center in just five mouse clicks.
Jeanine also mentioned Euronics, which is using SAN Volume Controller's Stretched Cluster capability, which allows them to easily vMotion virtual guest images from one data center to another. SVC has had this capability for a while, but now, with full VMcenter plug-in and VAAI support, the capability is fully integrated with VMware.
A final example was a mid-sized University, they are using IBM Storwize V7000 with Metro Mirror. The primary location's Storwize V7000 manages Solid-state drives with Easy Tier. The secondary location's Storwize V7000 has high-capacity SATA drives and FlashCopy.
Customer Testimonial - University of Rochester Medical Center
Rick Haverty, Director of IT infrastructure at University of Rochester Medical Center [URMC] provided the next client testimonial. The mission of the URMC is to use science, education and technology to improve health. URMC gets over $400 million USD in NIH grants, which puts them around 23rd largest University-based academic medical centers in the country. They have over 900 doctors, general practice and specialists.
URMC has an IBM BlueGene supercomputer, a Cisco network over 45,000 ports, and over 7.5 million square feet of Wi-Fi wireless internet coverage. They have three datacenters. The first is 7500 square feet, the second is 6000 square feet, and the third is just 800 square feet to hold their "off-site tapes".
URMC has digitized all of their records, including Electronic Medical Records (EMR) system, medical dosage history, imaging "priors", calibration of infusion pumps, RFID monitoring, and even provide IT support while the patient is on the operating table. RFID monitoring ensures all of the refrigerators are keeping medications at the right temperature. A single failed refrigerator can lose $20,000 dollars worth of medication.
When is a good time for downtime? At URMC, they handle 90,000 Emergency Room vists per year, so the answer is never. When is the ER busiest? Monday morning. (not what I expected!)
URMC's EMR software (Epic) runs on clustered POWER7 servers, with DS8700 disk systems using Metro Mirror to secondary location. They also keep a third "shadow" POWER7 for read-only purposes, and a separate system that provides web-based read-only access. Finally, they have 90 stand-alone Personal Computers (PCs) that contain information for all the patients that have reservations this week, just in case all the other systems fail.
The exploding volume of data comes from medical imaging. For radiology (X-rays), each image is called a "study" takes 20-30 MB each, and they have 650,000 studies per year. This represents about 16TB storage per year, with 3 second response time access. These must be kept for 7 years since last view, or until the patient reaches the age of 18 years old, which ever is later.
But radiology is just one discipline. Healthcare has a whole bunch of "ologies". Another is "Pathology" which looks at cells between glass slides in a microscope. Each study consumes 10-20GB, and URMC does about 100,000 pathology studies per year, representing 150TB per year.
URMC has identified that they have 42 mission-critical applications. The data for these are stored on DS8000, XIV, Storwize V7000 and DS5000, all managed behind SAN Volume Controller.