« Raising Kids In The Information Age | Main | Five Impressions From Mumbai »

November 16, 2011

Comments

nate

Good point chuck! I couldn't agree more.. Though in reading your post I found myself pondering whether or not "file based" auto tiering can significantly lessen the impact of maintaining multiple file systems with the global name space. Your mentioning of 16TB FS makes me think your aimed at NetApp but am not sure. I don't believe NetApp has any sort of tiering technology at this point (outside of PAM), so this speculation wouldn't apply to their devices.

What I think is even more important though is the disclosure of the costs of the system itself, something I've harped on a lot as to one of the main reasons I like SPC-1, to me it's more about the disclosures than the results. How did they get to those numbers, what is the specific configuration, what is the cost, support etc.

SPEC SFS for example discloses NetApp intentionally cripples their 3000-series arrays with dual core processors.

You should mention the EMC Celerra results which exported 18TB of data across *8* file systems.
http://www.spec.org/sfs2008/results/res2010q1/sfs2008-20100106-00155.html
(I don't see mention of global name space either)

Would be nice to see if EMC/Isilon can augment the isilon technology to be able to run as a more traditional gateway against the higher end EMC back end storage in addition to the scale-out design that it has now. Since of course the market for gateway-type designs (like Celerra) is much bigger than than that of scale out designs.

HP did this with their X9000 boxes, NetApp of course has the V-series(which I think requires NetApp disks if you want to use cluster mode unless that has changed recently). It doesn't appear that Dell is willing to go that route with Exanet technology (despite Exanet being a great gateway product). HDS does it with BlueArc (assuming BlueArc still runs on LSI disks as their standard back end since being acquired I don't know).

Can Isilon run more than one file system on the cluster? One limitation on the earlier generations of Exanet technology for example was you were limited to 1 file system per cluster. For the most part this was fine but there was cases where I would of really liked to have had a few TB in another file system(on another tier of back end storage) for certain workloads, would of gone a long way. They were working on this when the company collapsed, not sure where Dell is at with it now.

I shot an email over to a friend of mine who runs field services for Isilon asking him some questions on Isilon (e.g. is it limited to RAID 5 13+1, can RAID span controller nodes etc), still awaiting a reply.

The comments to this entry are closed.

Chuck Hollis


  • Chuck Hollis
    SVP, Oracle Converged Infrastructure Systems
    @chuckhollis

    Chuck now works for Oracle, and is now deeply embroiled in IT infrastructure.

    Previously, he was with VMware for 2 years, and EMC for 18 years before that, most of them great.

    He enjoys speaking to customer and industry audiences about a variety of technology topics, and -- of course -- enjoys blogging.

    Chuck lives in Vero Beach, FL with his wife and four dogs when he's not traveling. In his spare time, Chuck is working on his second career as an aging rock musician.

    Warning: do not ever buy him a drink when there is a piano nearby.

    Note: these are my personal views, and aren't reviewed or approved by my employer.
Enter your Email:
Preview | Powered by FeedBlitz

General Housekeeping

  • Frequency of Updates
    I try and write something new 1-2 times per week; less if I'm travelling, more if I'm in the office. Hopefully you'll find the frequency about right!
  • Comments and Feedback
    All courteous comments welcome. TypePad occasionally puts comments into the spam folder, but I'll fish them out. Thanks!