StorageRap

Americas Most Wanted: Smart Car

Watch out!  These things are showing up everywhere and they are going very fast.

SWCSA SMACK ATTACK! Vote for Storage Smackup of the Year, 2008

Blogging is like the cold war.  A lot of verbal weapons testing gets done.  We tweak each other to see what the reaction will be, we establish ad-hoc alliances when they are convenient and we try to win the hearts and minds of any customers that are paying attention.  Sometimes it gets a little out of hand and other times civility breaks out. If anybody doubts the intense level of competition in the storage industry, they haven't been reading the smackups in the blogosphere.  There were some real donnybrooks last year - and not surprisingly the majority of them involved EMC and Netapp, the market leaders in our industry.

The SWCSA has decided that the smackups listed below were the ones most deserving of recognition and nominates them for the award of Storage Smackup of the Year, 2008: 

1. Chuck Hollis (EMC)  vs. Kevin Closson (Oracle) 

2. Chuck Hollis (EMC)  vs. Tony Asaro (HDS)

3. Chuck Hollis (EMC)  vs. Alex McDonald (Netapp)

4. Chuck Hollis (EMC)  vs.  Curtis Preston (Glasshouse)

5. Chuck Hollis (EMC)  vs. Val Bercovici (Netapp)

6. Barry Burke (EMC)  vs  nobody (IBM)

7. Marc Farley (3PAR)  vs  John Spiers (Lefthand)

8. Barry Burke (EMC)  vs. Barry Whyte (IBM)

9. Alex McDonald (Netapp)  vs. The SAN Technologist (Dell) 

10. A battle royale with Storagezilla, Val Bercovici, Tony Pearson, Scott Waterhouse, Kostadis Roussos,

There have undoubtedly been omissions which I mightily apologize for - NOT.  Write in votes and all manner of comments and outrage will be published (except profane or abusive).  Of course, there have been instances of competing storage bloggers agreeing and getting along well enough, but don't bother submitting them as candidates because they are not, by definition, smackups.  Cast your vote by comment, email or Facebook message. Votes submitted by Twitter will not be counted.  One vote per communique, but multiple communiques are allowed from an individual.  Vote early and often.

3PAR Reports Financial Results for Third Quarter of Fiscal 2009

The bottom line:

Revenue for the third quarter was $48.2 million, an increase of 57% compared to revenue of $30.8 million for the same period a year ago, and an increase of 7% compared to $45.1 million in the prior quarter, which ended September 30, 2008.

David Scott's (our CEO) said:

"The strong financial results we report for the third quarter of fiscal 2009 result from a continued focus on fiscal discipline and reflect the unique value that 3PAR Utility Storage brings to organizations that need to do more with less in challenging economic times.  As datacenters have continued to look for new ways to economize, they have continued to turn to 3PAR to lower CAPEX and OPEX by reducing disk capacity purchases, energy costs, and storage administration effort by up to 80%."

For more details, go the the CNN Money page.

Is unified fabric breaking through during the downturn?

People who have followed my writing at EqualLogic, Dell and now here at 3PAR know that I haven't been a big believer in unified fabric.  But, over time I'm starting to think there are good reasons for customers to consider moving to unified fabric.

This post by Richard Jones at the Burton Group has some interesting insights into the dynamics of decisions and deployments of unified fabric. 

Here is a taste:

I’ve noticed a few organizations just starting to warm up to the ideas of a unified fabric.  Have they moved there yet?  No. The technologies are not mature enough to complete an entire unified fabric infrastructure.  Cisco knows this and is positioning the Nexus switch capabilities as a future proofing move.

Worth popping over for the short read.  

Putin to Dell: we'd rather just compete

You've probably all seen it by now, Vladimir Putin telling Michael Dell that he doesn't want his (or our industry's) help.

This whole thing is pretty screwed up and I feel a bit sorry for Michael Dell for not realizing what he was walking into.  He asked a question that I'm sure was meant to be a polite conversation starter and Putin, the world leader, took Dell's overture and crushed it. 

Michael is no newcomer to international business, but he should have known better than to speak first during the Q&A.  Putin was clearly on a mission to promote his country's capabilities, industries and products.  He was not interested in buying anything - he was selling and selling hard - confident in what he had to offer. 

Dell's question was not very well formed and I can only imagine the interpreter had some difficulty with it.  The word that seems to have caught Putin's ire was "help", which may have been interpreted to mean something closer to "aid" than "work together".  If that was the case, Putin's response,  "We don't need help. We are not invalids. We don't have limited mental capacity", makes sense. Why wouldn't he be insulted by an offer from Dell for technology aid?

If you watch the video, you'll hear Putin (through a translator) discuss working as equals in the world economy.  US companies dominate the technology sector and Putin clearly wants to get Russia more heavily involved.  I don't think Dell's question was all that awful, but he probably could have used a better choice of words. A question exploring opportunities for working together would certainly have drawn less hostility. 

By the same token, it's not clear that Putin's response will serve him well in the global IT industry.  Russia's technology industry will need to find partnership opportunities to grow, and since many of those partnership opportunities will be with US companies, it might be good to demonstrate the spirit of partnering, as opposed to competing.

Dave Graham, SWCSA member on tapeless backup

SWCSA member Dave Graham just posted on tapeless backup technology.  Its a good quick read, not as much fun as the Hardy Boys, but more likely to stir debates in the storage mosh pit.

My father in law is 87, still downhill skis and remembers how back in the 60's when he was working on disk drive technology, they thought tape was going to die and shrivel up before the end of the decade.  (Maybe Nixon helped restore everybody's faith in the technology. )

Whatever, the stuff is hard to kill, like cockroaches and storage analysts.    Disk R&D spending has probably exceeded tape a thousand times since then and tape hasn't bitten it yet.  It surpasses all human (and storage analyst) understanding.

About this thin provisioning stuff

This article on thin provisioning ran in Network World yesterday.  One of our customers, Matthew Yotko from IAC in New York is quoted in it, explaining some of the issues he had with thin provisioning and giving some numbers for how they are using it.  It's a quick read - and Yotko shares some good advice - and good results about his 3PAR experience.

The main takeaways are:

First - definitely know how to manage the capacity growth of a thin provisioning system, including setting thresholds for capacity growth so you don't have to rush buying additional capacity.

Second, you can get amazing utilization rates.  Mr. Yotko told Network World that he took his disk utilization from 20% to 80% using 3PAR thin provisioning.

Some of the other points in the article are relevant, but its important to realize that there are significant differences between different vendors implementations.  Like many thin provisioning articles, it's a bit lopsided in presenting warnings and caveats and light on covering the benefits and advantages. But that's OK, most storage administrators have already figured out how important the technology is to their line of work.  

Five ways thin storage can help you get through lean times

In this economic climate, where business costs are under such close scrutiny, storage customers need to know how to get the largest return on their storage investments.  Here are five ways 3PAR InServ storage arrays can help customers get through these tough times.

1.  3PAR customers typically buy only the disk capacity they need when they first install their InServ arrays and then add capacity later when it is needed. This "pay as you grow" approach to buying storage is what thin provisioning was designed for.  Customers save money with their initial purchase and also when they later buy additional capacity due to the declining cost of storage capacity over time.  

2. 3PAR customers run their InServ arrays at higher disk utilization levels, which means they get more productivity from their storage investment.  Higher utilization is one of several benefits derived from 3PAR’s wide striping disk layout scheme that spreads data across a very large number of drives (hundreds of drives are common).  This randomizes the workload, effectively applying all drive resources to all applications and removing drive performance bottlenecks – even at high utilization levels.

3. 3PAR customers use automated tools to provision and expand their InServ arrays, which means they can avoid spending money on professional services or spending lots of time planning and performing provisioning tasks. The ease with which InServ arrays are provisioned and expanded comes from the combination of wide striping, fine-grained volume management (RAID) and autonomic disk layout functionality.  

4. 3PAR customers mix drive types, workloads and platforms in their InServ arrays.  Mixing drive types reduces the cost per GB when lower cost nearline (7200 rpm) drives are used for primary storage. Mixing workloads and platforms means customers can achieve greater levels of storage and server consolidation, which reduces both capital and administrative costs.

 

5. 3PAR customers typically spend less on power, cooling and floor space for a given workload.  This is due primarily to thin provisioning and higher utilization levels on InServ arrays, but it also comes from the use of nearline drives for primary storage, which combine greater densities with reduced power and cooling requirements.

The Obama administration, government cloud computing and 3PAR

Last week, President Obama's transition team made quite a stir in technology circles when the TIGR (Technology, Innovation and Government Reform) Team released the video shown below  which discusses ways they plan innovate with technology to leverage information much more effectively.  Cloud computing is highlighted as a key for getting this work done.

CSC and Terremark, two 3PAR utility computing customers are planning to be heavily involved with these important projects, providing utility data processing from Terremark's data centers built out with products from HP, IBM, Cisco, VMware and 3PAR and covering Windows, Linux and Solaris platforms.

3PAR is extremely excited to be a key part of this re-engineering of government technology.

If the video does not show, you can read about it on the change.gov website here. 

Practical limitations to storage energy standards

 

Geoff Hough,  Sr. Director, Business Strategy

Kudos to the SNIA Green Initiative for continuing to drive awareness and standards around storage energy consumption.  Note last week’s announcement around a taxonomy and power measurement spec  (at idle… but hey, it’s a start and getting a number of opinionated storage folks to agree on a spec is no easy task).

Playing this out to its hoped-for conclusion:  what if all storage devices brought to market were measured for energy consumption using the same set of metrics and would it matter?   The answer is most certainly “yes.”  As part of any storage evaluation process, you would compare the projected consumption of kilowatt hours between alternatives, multiplying any difference by your local utility’s charge per kilowatt hour.

But for those who run their storage devices in co-location environments, the calculus of energy savings can get quite distorted.  Since many co-lo’s charge for energy consumption based on the number of line drops, co-lo occupants lack the economic motivation to select energy efficient storage technology.   How many Californians would select low-flow shower heads if they paid for their water based on the number of pipes connected to their homes?

So, modern storage technologies like MAID, thin provisioning, and SSDs may be attractive for many reasons, but any effective energy savings they provide to their owners is disproportionately small in a co-lo environment.  More to the point, if technologies like these are chosen for their other merits, any residual energy benefit goes largely into the pocket of the co-lo.

One wonders if there are any SNIA-like industry organizations for the co-lo community that contemplate their role in global warming.  If so, we can only speculate if their agenda includes pricing initiatives that give their customers an incentive to implement energy saving storage products.

The green efforts underway in the storage industry will be more effective if we can get alignment on energy efficiency goals across all constituents of the IT industry. 

Ray Lucchesi on snapshots for VDI

Ray Lucchesi, president of Silverton Consulting, is a long time storage guy from Broomfield Colorado.  He published a piece in InfoStor today about the importance of using array-based snapshots for boot images in VDI environments.  Defintely worth the time to read the whole enchilada. 

Below are two paragraphs that talk about the tedium involved and what 3PAR has done to make storage admins lives easier:

Given the compelling benefits and resolvable concerns of snapshots for VDI boot images, administrators using VDI boot image snapshots still have a difficult task.  Once they have created a boot image ".vmdk" file, they need to locate the VMFS datastore it resides on, determine the LUN holding the datastore, locate the physical storage this LUN resides on, issue the requisite subsystem-specific requests to snapshot the datastore, and then export the new LUNs to ESX.  Following the export, the administrator must signal VMware to re-scan for newly snapped LUNs, and to re-signature the snapshot volumes for VDI virtual machines use.  Finally, the administrator must clone the virtual desktop configuration and attach the virtual machine to use the newly created boot image.

Recognizing the tediousness of many of these configuration tasks, some vendors provide scripts to automate much of the process. For example, 3PAR's Thin Copy Desktop for VMware VDI script, supplied as a customizable script, uses the VMware "perl API" to map the boot image from a VMFS datastore to a  LUN, directs the subsystem to snapshot the source LUN multiple times, exports the new LUNs to ESX, and signals VMware to re-scan and re-signature the resultant LUN(s), which are made visible within the VCenter inventory. While not all configuration tasks have been scripted, much of the drudgery of traditional storage provisioning has been eliminated.

Thanks Ray!

Psst, hey pal, want a blood transfusion?

This video is rated WN (white and nerdy). Serious storage people, like Stephen Foskett, who don't want to see videos like this, should not watch.

Chuck Hollis Applies for a job at Netapp!

Something startling happened today.  Netapp was named the #1 company in Fortune's Best Places to Work.

But that's not what was surprising, THAT reaction was reserved for Mr. Congeniality's post.

Maybe the storage blogosphere will catch the spirit of whatever is blowing around these days.   Or maybe Chuck is trying to soften up the HR folks at Netapp so he can get a job as Alex's boss.  If there ever was an unlikely mutal admiration society, this has to be it.

The heartbreak of more puzzles, or wide striping?

George Crump just posted an brief article on the risks of IT running at capacity.    Its certainly true that when we push things in the directions of their limits, strange and less-than-wonderful things tend to happen.  Often, its bad expectations in the programming of the protein robot that malfunctions most spectacularly, but that's also when bugs are exposed (another case of bad expectations, but in code) and knee curves that were never disclosed make their appearances, delivering blows to the sensitive areas of our operations.

And so what happens next is the realization that there is bone crushing work to do.  Not the panacea of future automation the likes that StorageBod has been gathering from tweets, but  the drudgery that brings on palpable headaches from thinking through the mess that the Storage Architect scratched at this week.  (migrating data and provisioning storage as part of the process)

A lot of storage arrays are going to end up being re-provisioned in the months to come because the fact is, they are probably not out of capacity but they are probably provisioned in such a way as to make too much of the available capacity unusable.  They are gut wrenching puzzles.  Puzzling to understand what the current status is, puzzling to remember why they were provisioned the way they were, puzzling to figure out how round circles are going to fit into square holes, and puzzling to figure out what to tell others their expectations should be while you figure out all the aspects of the puzzle. Hard emotional puzzles - the kind that drain you much more than energize you.

Why does this happen?  Because there were performance goals and growth projections and politics and all kinds of pressures stemming from expectations when the work was done before.  There always are.  But some of those previous expectations don't matter as much any more.  Tough times tend to re-establish priorities.

Storagebod threw me a curving softball this week when - in reaction to Storage Architects blog - he asked for "a plug" from me about how thick to thin provisioning might work.  I know he's spoken to some of our folk, but the disappointing thing is that I can't really go into that in a public forum just yet.

Instead, I'll say that there are alternatives to arcane disk layouts that make the puzzles much easier to deal with.  The main one is wide striping.  Wide striping that:

• Spreads the data from all applications and systems across all the spindles in your system
• Uses capacity evenly on all disks, not by unevenly sized RAID containers
• Applies all spindles all the time to getting work done and where there are no reserve spindles
• Expands LUN capacity by adding incremental amounts from all disks in the system

 I'm not talking about wide striping that uses something like 16 or 32 disks, but hundreds of them. 

Its why I think in the end, wide striping is a more important technology for 3PAR than thin provisioning.  Yes, I'm a 3PAR heretic.  Don't get me wrong, thin provisioning is great technology but it doesn't apply in all situations (high transient data environments, such as those where large media files are downloaded/created/deleted/ repeated).  However, wide striping applies to just about every application and system you can think of, all the time.

If you had wide striping today, the re-provisioning puzzle pratically goes away - it's a non-issue.  

If you don't have it today and you want to make your data migrations a whole lot easier look at putting in wide striping to reduce the amount of puzzling you have to do to figure out where everything goes.  Why not spread it over a whole lot of disks where the capacity planning is a LOT simpler and the performance is a LOT more predictable.

And as the Storage Anarchist writes in his post on XIV,  why would you want to do this on a system with only SATA drives? 

A sound byte from President Barack Obama's inaugural address.

(the transcription is below)

Our challenges may be new. The instruments with which we meet them may be new. But those values upon which our success depends - honesty and hard work, courage and fair play, tolerance and curiosity, loyalty and patriotism – these things are old.  These things are true.  They have been the quiet force of progress throughout our history.

What is demanded then is a return to these truths. What is required of us now is a new era of responsibility.  A recognition on the part of every American that we have duties to ourselves, our nation and the world.  Duties that we do not grudgingly accept but rather seize gladly, firm in the knowledge that there is nothing so satisfying to the spirit, so defining of our character, than giving our all to a difficult task.

This is the price and the promise of citizenship.  This is the source of our confidence. The knowledge that God calls on us to shape an uncertain destiny. This is the meaning of our liberty and our creed why men and women and children of every race and every faith can join in celebration across this magnificent mall.

And why a man whose father less than 60 years ago might not have been served in a local restaurant can now stand before you to take a most sacred oath.