Storagebod is an Enterprise end user - I have discussions with him via the blogosphere and twitter. He is in the trenches and has hands-on knowledge of the demands of high end storage. He recently responded in his blog - Virtualisation 2.Oh Give It A Rest - to Hu Yoshida’s blog - “Where will storage virtualization take us?” Since I find it hard to resist inserting myself into this kind of dialog - here is my take.
Storagebod made a few important points - some of which I agree and others I do not:
“Redefining Storage Virtualisation, declaring Virtualisation 2.0; these are the acts of a desperate man.” (And there is more of that type of commentary that you can read on Storagebod’s blog.)
As you can imagine I don’t agree with this on a number of levels. HDS is doing very well with the USP-V and therefore there is no reason to be desperate - Hu least of all. More importantly, Hu is very passionate about the USP-V and its implementation of external storage virtualization - the man lives and breathes this stuff - it is “religious” in a sense. Also - if you read Hu’s blog - he isn’t really all that extreme about it - he mentions that SNIA’s view of Storage Virtualization 2.0 is “The application of virtualization to storage services or devices for the purposes of aggregating, hiding complexity, or adding new capabilities to lower level storage resources.” Therefore within this context - it pretty much is one of the core value propositions of the USP-V.
I don’t understand the harshness of Storagebod’s comments of Hu Yoshida - Hu is a gentleman and is very polite and not one to partake in sarcasm or counter-attacks. There are certainly other vendor bloggers that are far more mean-spirited and issue forth major FUD that don’t inspire this level of harshness.
Storagebod makes the following point:
“But give the virtualisation thing a rest! It’s not the solution to all the storage issues in the world! If it were, we’d all be rushing to put it in (or are we just too stupid to see that it is the solution to all the storage problems in the world?).”
I do agree that external storage virtualizaiton is not the solution to all storage issues in the world. However, it is valuable and probably not fully understood by everyone that could be positively impacted by it. HDS is partly to blame for this - they have not clearly educated the market on real world implementations and quantified the resulting value for everyone to get that “aha” moment. I vehemently disagree that HDS should “give it a rest” - it is one of their competitive advantages and provides unique value and capabilities to its customers. My advice to HDS is to RAISE THE VOLUME - and not just be louder but also be smarter about how you talk about it.
Storagebod adds:
“And this insistence that USP is somehow different to SVC is rubbish! SVC and USP are extremely similar conceptually just implemented in slightly different ways!”
I agree that the SVC is a good product and provides a ton of value to IBM customers - and there are conceptual comparisons to be made. However, one of the most important things about the USP is that it is a standalone Enterprise-class storage system. It lives within the most intense mission-critical environments with the highest levels of intelligence, scalability and reliability. External storage virtualization is just another feature of the USP-V, albeit an important one - that you can turn on as needed. The implications of just this fact are multi-fold:
1. Let me ask Storagebod a hypothetical question: Would he seriously consider using the USP-V for a tier one mission-critical application using a combination of local disks in the USP-V and tier 2 storage virtualized behind the USP-V? I would suspect the answer would be yes. Then I would ask - would he seriously consider using the SVC for a tier one mission-critical application using only tier 2 storage behind it? I suspect the answer would be no. The USP-V is the highest order storage system providing its capabilities either on a peer level or downward.
2. In some cases IT professionals don’t have budget for storage virtualization solutions but they do have budget for storage systems. In large and bureaucratic organizations this can mean the difference between getting access to this technology or not. One system administrator I spoke with wanted to implement storage virtualization but was shot down at budget review. However he was able to buy the USP-V and get storage virtualization under the radar.
3. Since the USP-V is both a storage system and an external storage virtualization platform in one it enables you to roll out external storage virtualization at your own pace - getting value from the USP-V as a standalone storage system and also for external storage virtualization.
4. The USP-V is HDS’ flagship product and therefore the greatest and smartest amount of resources goes to this product. That cannot be said of any competitive external storage virtualization solution.
I do agree with Storagebod’s sentiment - which I interpret as “enough high level rhetoric - let’s get to the nuts and bolts”. Give us facts, give us use cases, give us quantifiable value, give us education based on real world results. And all of this needs to be mixed in with the “gotchas” and the “challenges” and an understanding of the “right fit”. I think external storage virtualization 2.0 isn’t about product or technology but rather a deeper knowledge of how it applies to the real world. Since I have some insight to what HDS marketing has planned - I know that this is the goal and objective going forward.
Comments (6 )
marc farley on 16 Jan 2009 at 12:55 am
I think Hu is a pretty good guy,, but his blog writing doesn’t expose his best qualities. Hu tends to ignore comments and does not really participate in online conversations (or smackdowns) - which makes a lot of people suspicious - or dismissive.
tony on 16 Jan 2009 at 7:18 am
Marc - all of that may be true(although I don’t agree with the suspicious aspect) but I am perplexed and curious about the harshness. There are others in the vendor commmunity that probably deserve relentless scrutiny based on their behavior. In any case, my blog wasn’t just about Hu - in fact it is mostly about external storage virtualization and the USP-V. I think that HDS needs to keep talking about it but perhaps in different ways. Just imagine if someone told you guys to stop talking about thin provisioning. And I did just see you write another blog on Chunklets. It is core and part of your essential value proposition.
Daniel on 18 Jan 2009 at 5:34 am
I like seeing vendors being brought down to earth, in some respects its what keeps you all ahead of the customer need by taking on feedback. If you are willing to take realistic sensible (not negative) feedback on the chin and amicably then you gain more respect in my book as a vendor.
I think Hu being in the blogosphere is great for the industry and I enjoy reading most of his material, taking the high ground on smackdowns and vendor smacking is i expect his tactic and is the sign of a good leader…
tony on 19 Jan 2009 at 4:20 pm
Daniel - I agree that vendors should take feedback and input. I also believe that there should be a discussion - and that positions should be made on both sides of the virtual table. Thanks for the input.
Barry Whyte on 21 Jan 2009 at 4:04 pm
My take, and I think what Martin was getting at is that 2.0 is not needed. With SVC we have been providing all of what Hu and yourself talk about since 2003.
The ability to take modular ‘unintelligent’ storage and provide the intelligence and enhance the underlying storage is what storage virtualization is all about. You don’t need to buy an expensive USP-V to do that. Heck, you guys sell it too as USP-VM - but Hu seems to forget you sell that, and tries to attack SVC by saying its not storage virtualization. Well, I do find that hard to swallow, when EVERYTHING he says that virtualization 2.0 is provided by SVC.
Its been good to see Hu engage a bit more in the community recently, and I’ve not ’spat my coffee’ for a few months
tony on 21 Jan 2009 at 9:48 pm
Barry - to me the SVC does provide external storage virtualization - and you have thousands of implementations to stand behind it. I think that it is actually in the common interest for HDS and IBM to both continue to drive adoption and education since the value proposition is so powerful - especially since the competition is lacking in this technology.
Hu didn’t bring up SVC in the blog I am referencing but I do understand that the debate is a long standing one. There are important differences but that doesn’t mean that both products don’t have a signficant place in the market. I believe they do. However - it is important to highlight the differences - I happen to think the points I make are the importants differences.