Storage and Technology Academy

IBM Storage Compression - game changer? or late to the game?

raj.sharma — Thu, 12 Jul 2012 03:36:00 GMT

Since the announcement of Real Time Compression technology in the IBM Storwize v7000 and IBM SVC disk subsystems, I have been flooded with questions asking ..."Is this technology really a game-changer or is IBM playing catch up with the other vendors"?

The answer: "It is a game changer because unlike anyone else, IBM compression is compression without compromise".

Before I explain this further, we have to rewind the clock a bit to understand the latter part of my statement. Unlike other storage efficiency features (ie thin provisioning, tiered infrastructure etc), compression on primary data forces customers to strike a balance between compressibility and performance penalty. Compression adds overhead to the disk subsystem so in the traditional approach to compression customers would have to selectively choose what data they would want to compress. If you pick to high an amount, compression will chew up a significant number of CPU cycles and grind the storage subsystem to a halt - resulting in a degradation of service. Other vendors will be quick to point out that customers use compression as an optional feature to either 1) use it as a post process (this means data will get written to disk and get compressed off hours) or 2) compress test/dev or any other workload that is insignificant.

I call IBM compression a game changer because we announced it on June 4th, and we want customers to actively use it........Use it everywhere - on all data sets! No compromise - just compression.

I recent IDC analyst report summarized the real differentiator or Real Time Compression.

..........The primary differentiator is that in Real-time Compression, the compression engine processes a variable data stream based on the patterns of data that are actually written. Real-time Compression takes advantage of "temporal locality" and not physical location: Data that is accessed together is compressed together independent of its physical location. So when applications make related updates to different parts of the volume, such updates are handled together in a contiguous manner. This is akin to real system operation because it takes advantage of the structure of the data, the size of the data, and the data's relationship with other data. This awareness of workload minimizes the number of compression and/or decompression operations, resulting in fewer disk IOPS and less overhead on the storage controller. This increases compression ratios and efficiency without compromising performance.

Here is a link to the IDC paper entitled "Real-time Compression Advances Storage Optimization" http://public.dhe.ibm.com/common/ssi/ecm/en/tsw03129usen/TSW03129USEN.PDF

The new v7000 and SVC GUI also have a Real Time Compression GUI support that shows separate CPU utilization for Compression and System workloads.

So there you have it. Some workloads such as databases, VM's, and CAD files compress better then others. Contact you local IBM storage representative or send me an email (rajs@ca.ibm.com) if you would like more information.

Raj S

Saying goodbye to the trees - Tivoli Storage Productivity Center

Paul Moretto — Thu, 12 Jul 2012 02:24:00 GMT

For my inaugural post I'll highlight one of the biggest changes that has happened in the last couple years of me speaking with customers as an IBM Storage guy.

Before I do this I'll step back; I am the person for Canada that has been blessed with specializing IBM Tivoli Storage Productivity Center and SRM (that is Storage Resource Management not VMWare's Site Recovery Manager). The topic of Storage Resource Management doesn't generally evoke feelings of excitement or passion in most people, but I live in breath it. Sometimes changes can excite me beyond what any common human would understand. IBM's Tivoli Storage Productivity Center is a great product is ranks among the top of its class according to Gartner. It can manage, automate, and control storage in an highly effective manner, and yet I get the same objections from customers across Canada.

"I can't find anything"

"The interface scares me"

"It isn't easy to use"

Why would anyone say this about this amazing product? It all comes down to the dreaded navigation tree.

I get it. You need a report on how much capacity you've consumed over the last six months, but where do you start? I use this product day in and day out so I know where to find things, but to any common user this is not intuitive and can even evoke anxiety, fear, or heightened blood pressure. With IBM's Storage brand on board with the ipodification of all their GUI's it was only a matter of time before the Tivoli Productivity Center got the Web 2.0 treatment.

There you have it, no more navigation tree. No more headaches or nightmares. If you're one of the rare few that has been using TPC for years and likes the old gui (like myself) then the legacy mode is still available with the tree.

Stay tuned for more updates on the new GUI and enhancements to help you sleep at night.

Announcement - new Blogger

raj.sharma — Thu, 12 Jul 2012 02:11:00 GMT

Announcing a new blogger

Hi Folks. After a short hiatus to take care of some housekeeping, we are back online!! My first order of business is to introduce a new blogger to the site (drumroll) ...... Paul Moretto!!!

Paul, like myself, works on the IBM Storage pre-sales technical team. He brings with him a wealth of knowledge on IBM SAN Hardware as well as Software features (eg, Tivoli Storage Productivity Center, SAN/NAS Encryption etc).

Paul, welcome aboard.....and "may the bloggin' be with you ....."

VAAI and why Storage Admins should know about it (Part 2 of 2)

raj.sharma — Wed, 14 Dec 2011 02:00:00 GMT

Hi Folks.

Finally - I have been able to upload part 2 of 2 of the series. Sorry for the delay (and thanks to those who sent emails requesting it. Means people are reading).

THe second part of the series concentrates on BLock Zeroing and Hardware assisted locking. Enjoy -

Raj (rajs@ca.ibm.com)

--------------------------------------------------

Block Zeroing

Block zeroing is designed reduce the amount of processor and storage IO required to write zeroes across an entire Eager Zeroed Thick (EZT) VMDK when it is created. With the block zeroing primitive, zeroing operation for EZT VMDK files are offloaded to the XiV storage system without the host having to issue several commands

Benefit: Reduce the amount of CPU effort and IOPs required to write zeroes across an entire Eager Zeroed Thick (EZT) VMDK

Hardware Assisted Locking

It is important to understand why locking occurs in the first place. For block storage environments, VMware datastores are formatted with the VMFS filesystem. VMFS is a clustered filesystem which uses SCSI reservations to handle distributive lock management Although normal IO operations do not require this mechanism, these boundary conditions have become more common as features such as Vmotion with DRS are used more frequently. With hardware assisted locking, the LUN locking processing is transferred to the storage system. This reduces the number of commands required to access a lock, provides locks to be more granular than the entire LUN, and leads to better scalability of the virtual infrastructure.

Benefit: Hardware assisted locking will increase VMs per datastore, ESX servers per datastore, and overall performance

IBM Support for VAAI

VAAI is a key enabler to truly gaining the operational benefits of VMware. Although offered by more and more storage vendors, different storage hardware architectures will lend themselves far better to VAAI integration than others (mileage will vary ). Enterprise array architectures will leverage all of the VAAI capabilities while lower level arrays may not be able to handle the additional tasks assigned to them by VAAI. This is why many vendors are only offering support for parts of the VAAI API’s.

IBM is offering VAAI support on a variety of its disk platforms. The IBM XIV, Storwize v7000, and San Volume Controller (SVC), will support all 3 of the VAAI primitives. Each enterprise class system has specific features that can best leverage VAAI. Distributed Distribution Grid (XIV) , pre-fetch algorithms (XIV/v700/SVC), , storage pooling for more disk spindles (XIV/v700/SVC), load balancing (XIV/v700/SVC), thin provisioning (XIV/v700/SVC), and other high-end features have made them excellent for VAAI functionality.

IBM has been working with the VMware Reference Architecture Lab for many years and will continue to partner with VMWare to create the next set of API’s. Future integration like the VMware vStorage APIs for Storage Awareness (VASA) are expected to improve visibility into the physical storage as well as the virtual data store through vCenter and Capacity IQ. IBM Storage continues to be an excellent choice for VMware environments with support for VAAI, vCentre Storage Provisioning Plugin’s, SRM, Storage vMotion and future VMware storage enhancements.

VAAI and why Storage Admins should know about it (Part 1 of 2)

raj.sharma — Tue, 25 Oct 2011 04:13:00 GMT

As a Storage technical specialist, I get the opportunity to talk to customers about new technologies that are intended to make life much simpler. Recently, I have been introducing folks to a technology that was originally introduced in 2008 called Vstorage API for Array Integration or VAAI for short. Often, many people describe it as VMware API's for Array Integration. I dont know (nor do I care) what the real answer is - all I know is that it is a breakthrough technolgy that should be adopted.

So what is it? In a nutshell - VAAI allows one to offload array specific operations from VMware servers to the external storage device. Brilliant!

This blog is a two part series that I extracted from a document I wrote up . The first part discusses the first primitive - Full Copy Offload. The 2nd part of the blog (next week) will discuss the primitives Block Zeroing and Hardware Assisted Locking.

Enjoy - Raj

-------------------------

Virtualization administrators are always looking for ways to improve scalability, performance and efficiency of their vSphere infrastructure. One way to do this is utilizing storage integration with VMware vStorage APIs for Array Integration (VAAI). VAAI is a set of APIs, or primitives, that allow vSphere infrastructures to offload processing of data related tasks which can burden an ESX server. Utilizing a storage platform with VAAI enabled, can provide significant improvements in vSphere performance, scalability and availability.

In ESX 4.1, that vStorage API for Array Integration includes three basic capabilities or primitives: :

1) Full Copy (aka Hardware Copy offload):

Benefit: considerable boost in system performance and fast completion of copy operations; minimized host processing and network traffic

2) Block Zeroing (aka Write Same)

3)Hardware Assisted Locking (aka Atomic Test & Set): replacement of the SCSI-2 lock/reservation in VMFS

Benefit: Significantly improved scalability and performance

Full Copy (Copy Offload)

Tasks such as VM provisioning and VM migration are part of the everyday activities of most VMware administrators. As the virtual environment continues to scale, it is important to keep a watchful eye on the overall impact these activities have on the VMware infrastructure.

Here, the ESX server is removed from the data path of the data copy when HW copy is enabled. This greatly increases the speed of these copy functions while reducing impact to the ESX server.

The Benefit: considerable boost in system performance and fast completion of copy operations; minimized host processing and network traffic. Thousands of commands per second and IOPs on the ESX server can be freed up to process other tasks and promote greater scalability.

Linear Tape File System (LTFS) - What is it and why is it important?

raj.sharma — Thu, 13 Oct 2011 22:02:00 GMT

Tape-based storage is more relevant today than ever before. Depending on the application and overall environment, tape can be an optimal choice for Backup & Disaster Recovery. Tape especially shines for large-scale, long-term archiving. When implementing an active archive strategy, tape has significant advantages over disk, both in cost and functionality. Now that software exists to make tape a first-class online storage target, the cost savings—in particular, the greatly reduced power and cooling expenses—and the benefits of being able to create a portable off-line copy of invaluable data are too great to ignore.

Linear Tape File System (LTFS) is a self describing tape file system that enables tape media to be mounted and read by the operating system when the cartridge is inserted into the tape drive. The LTO-5 specification includes the ability to partition a tape into two segments. A small partition is used to store index information and the of the other, larger partition, is used to store user data. no particular application is needed to figure out what’s on the tape because the contents of the second partition are described by the index and metadata in the first partition.

A tape formatted with LTFS looks just like an ordinary file system; it can be mounted like any hard drive or thumb-drive. (Drivers are available for OS X, Windows, and Linux.) Users can even drag-and-drop files to and from a mounted LTFS formatted tape. See Diagram

Customer Benefits:

Requirements for IBM LTFS are IBM LTO-5 drives and LTO5 media. LTFS will provide the following customer benefits:

· Reduce Costs: An IBM LTFS archive solution is considerably less expensive than a comparable disk-only archive solution

· Ease of Use: –Allows users to read / write, catalog, search data on LTO-5 tape cartridges and libraries using a familiar browser directory tree graphical interface

· Single Standard of Media: Tapes can be moved easily (from one site to another) while files on tape can be accessed using straightforward drag-and drop.

For archival data - which does not need to be retrieved frequently or in a hurry—LTO-5 storage will continue to be a natural choice because it remains a lower cost and more secure alternative than keeping the same data online. The Linear Tape File System feature makes it possible for users or applications to easily store data to tape from their desktop. The two in conjunction casts an entirely new light on the future of tape.

For more information, contact your IBM Storage Representative or click on the following link:

http://www-03.ibm.com/systems/storage/tape/ltfs/

IBM and the History of Random Access File

raj.sharma — Mon, 26 Sep 2011 15:31:00 GMT

IBM in 1953 recognized the immediate application for what it termed a "Random Access File" having high capacity, rapid random access at a relatively low cost. After considering several alternative technologies such as wire matrices, rod arrays, drums, drum arrays, etc., the engineers at IBM San Jose invented the disk drive. The disk drive created a new level in the computer data hierarchy, then termed Random Access Storage but today known as secondary storage, less expensive and slower than main memory (then typically drums) but faster and more expensive than tape drives.The commercial usage of hard disk drives began in 1956 with the shipment of an IBM 305 RAMAC system including IBM Model 350 disk storage.