No big ideas in this post ...
No, this one is just various odds and ends I've been meaning to write about, but just haven't gotten around to it.
Sorry for the bouncy nature of the post ...
Meet Kathrin Winkler
Yes, EMC is getting very serious about sustainability -- not only our products, but how we conduct our business. You've probably read Dick Sullivan's blog Energy Matters at some point.
Well, a while back EMC promoted Kate to take on the humungous and mind-bending challenge of transforming EMC into a world leader on sustainability best practices. Trust me, this ain't about greenwashing, or slick PR. If you've ever met Ms. Winkler, you'd understand why I'm saying this.
Kate had a nice interview recently, and has just started her own blog on her personal journey in this new role. We're all rooting for you, Kate!
Pillar Trounces SPC-1, But Why?
I, for one, had hoped that the whole SPC diatribe was dead and buried. I know I've had my hand on the shovel to help out any way I can.
Well, no such luck. Pillar (via Mike Workman's excellent blog) announced best-in-class set of results on the controversial tests. Fortunately, this time we largely avoided the traditional back-and-forth on SPC relevancy, methodology, motivations, etc. that usually erupts every time a storage vendor reaches for this marketing play.
Although, I have to say, I really enjoyed Byte and Switch's comparison to cheating at poker, since they're all cheating using basically the same rules.
C'mon, IBM, NetApp, HP et. al. -- those SPC results you've got posted are looking really long in the tooth! I mean, do they make 36GB drives anymore? How about spending big bucks on some fresh results with your latest products to keep the whole thing relevant and meaningful?
You're not gonna let Pillar run away with this thing, are you?
Or, perhaps, have you come around to my personal view that it's all a massive waste of time?
BTW, among all the smaller storage vendors, I do tend to like Pillar, and I do like Mike Workman -- nothing personal, guys!
Seagate Has Firmware Problem
You're going to think I'm a sick puppy, but I actually smiled to myself when this story broke.
For years, I've been trying to tell people that most drive failures are really firmware and protocol failures; that extensive drive testing really matters, that the purported "double drive failure" scenario has little to do with hardware and everything to do with firmware protocols, that this issue affects how we build storage arrays like the DMX and CX, and much more.
For those of you who are enamored with the allure of do-it-yourself storage, here's yet another thing to consider, yes?
My apologies to all of you who are having a really bad day with your Seagate disk drives for this very reason, but it does illustrate a key aspect of storage technology that many people tend to overlook.
Seagate Has Other Problems
You probably noticed the management shakeup at Seagate as well. I'm sure that there's lots of things going on there, but I can't help but thinking that their cash cows -- high performance FC drives -- are in a bad way due to the advent of enterprise flash drives -- and, from what I can tell, Seagate doesn't have a visible play here yet.
Seagate did, however, chose to invest in providing various storage bricks that used their disks -- you have to admit, investing in competing with your direct customers (array vendors) is certainly a bold move.
In this business, strategy is everything. And one can speculate that one of the root causes here might be a strategic goof of substantial signficance.
Where Is Oracle Exadata?
Oracle's DW behemoth was announced with great fanfare last September. My pointed skepticsm even resulted in a minor war of words with some of the high priests of Oracle's DW community.
Well, here we are in January, and it looks like the whole thing is losing momentum -- as predicted. Sure, throwing a ton of hardware at the problem can make Oracle run much faster, but there are more elegant ways to get the same results -- which customers are starting to realize.
And, let's face it, who's got the money and the power and the cooling and the space for one of these monsters these days? Not only that, I've had customers tell me that they really didn't think it was a good idea to have DBAs administering servers and storage in a one-off manner. I would agree with that view.
Any way you look at it, I bet it's rough going. I wouldn't be surprised if they haven't sold (as in bought and paid for) a single one of these in a decent configuration. Or, if they have sorta sold a few, they're smaller configs.
The Oracle sales force has always struck me as very pragmatic; if they realize this is tough sledding, they'll go focus on something else.
Anyone heard anything similar or different?
Cisco Will Enter The Server Market
And, by far the most impactful news has to be this tidbit from the New York Times, detailing how Cisco plans to enter the server market before too long.
My challenge is that I'm so wrapped up in so many different NDAs that I'm almost scared to comment on this one, but -- you have to agree -- it should be very interesting, shouldn't it?
And ... That's All For Now
I need to go home and shovel snow again -- talk to you later.

Chuck, the post is bouncy, but you have covered quite a few very important topics.
The seagate problems have started to surface lately with a bunch of news coming out there: management shakeups to the recent drive issues, to the vision and the direction of seagate and really the future.....
The most interesting one is Cisco in this post. I can read in between lines you want to say something but i guess can't comment it.
Cisco's great plan / agenda is to be datacenter work horse??? The important things now with Cisco is every packet in a datacenter whether it is IP, FC or ISCSI (network oriented, san oriented) is passing through there switches. What are they missing today?????
Lets not talk about UPS's and Generators, but from being the first entry point in a network to being the last exit point (users that are connected to workgroup switches, aironet or webex) cisco dominates it all. With VMware and virtual switching (probably running cisco IOS in the near future), they are everywhere.
They do have some missing components in the equations, like???
Servers????? is SUN a good fit for CISCO????
Storage????? there have been rumors since 2005 about an acquisition in the storage market by Cisco???? I am not going to name it!!!!!
We have hear customer use terms like we are a SUN shop or an EMC shop or an IBM shop, I have seldom heard people saying we are Cisco shop!!! If they manage to put one of the above components together like Host or Storage, that would great a big impact in the market.
Again i dont think we are too far from it.......
Posted by: Devang Panchigar | January 19, 2009 at 11:31 PM
Chuck,
A couple of comments, well three...
Re Pillar. Yes, interesting technology, but largely irrelevant for majority of customers. Furthermore, the product line is missing key technology that is a must have these days in the storage business: array based software functionality (without having to rely upon 3rd party bolt-ons).
Re Oracle DW, we're seeing a fair bit of activity in the market from the Oracle sales force - they are compensated for the whole sale and not just the Oracle software components, so obviously there's interest in positioning this. The product is good and does offer an ROI over doing Oracle the traditional way. Having said that, there are other ways of finding that ROI by not implementing Oracle the traditional way too.
Re Cisco, this is a very interesting step for the company. If Cisco's strategic concept for getting into the server business of "compute-virtualization-connect" congruence is to be believed, this would also require Cisco to get into the storage business. Without this discipline, I think that much of Cisco's effort will be stillborn in the broader market and this will appeal to a select niche customer set. Perhaps we should once again dust off the annual rumour of Cisco buying EMC?
Geoff @ Dell.
Posted by: Geoff Mitchell | January 20, 2009 at 09:27 AM