Inside System Storage -- by Tony Pearson

Tony Pearson Tony Pearson is a Master Inventor and Senior IT Specialist for the IBM System Storage product line at the IBM Executive Briefing Center in Tucson Arizona, and featured contributor to IBM's developerWorks. In 2011, Tony celebrated his 25th year anniversary with IBM Storage on the same day as the IBM's Centennial. He is author of the Inside System Storage series of books. This blog is for the open exchange of ideas relating to storage and storage networking hardware, software and services. You can also follow him on Twitter @az990tony.
(Short URL for this blog: ibm.co/Pearson )
  • Add a Comment
  • Edit
  • More Actions v
  • Quarantine this Entry

Comments (4)

1 localhost commented Trackback

Fact Check - you wrote:<div>&nbsp;</div> "Thanks to its clever RAID-X implementation, losing data on an IBM XIV is less likely than losing data on any RAID-5 based disk array from any storage vendor."<div>&nbsp;</div> That is blatantly inaccurate and a total misrepresentation of facts, although you possibly don't understand the probability mathematics well enough to realize that.<div>&nbsp;</div> Question: Would IBM actually make that guarantee in a legally-binding contract with a prospect or customer, with monetary penalties if proven wrong?<div>&nbsp;</div> I seriously doubt it, but please - prove me wrong.<div>&nbsp;</div> Also, if you truly are interested in honest and factual assessment, how about you guys shipping a fully loaded XIV array to my office at 176 South St., Hopkinton. Last we asked, we were told "sorry, we can't spare you one right now" - even when we offered to buy it outright (and become your second hard-cash-paying customer).<div>&nbsp;</div> I promise we'll only run real-world benchmarks on it.<div>&nbsp;</div> Oh - and one more thing. IBM and XIV people are out there bad mouthing me personally in front of customers...which probably explains why my readership continues to climb (and XIV sales continue to be flatlined at 0). Seems your folks still haven't come up with credible responses to my observations.<div>&nbsp;</div> Thanks for the publicity, by the way. Just wish you'd all stay focused on what I said instead of attacking me personally.<div>&nbsp;</div> TTFN - I'll be waiting for that RAID-X vs. RAID-5 guarantee - you know where to reach me...

2 localhost commented Trackback

So, BarryB, for you to convert this quote from ESG:"Traditional benchmarks running a single application workload can’t help IT managers understand what happens when a mix of applications are deployed together in a virtual server environment."<div>&nbsp;</div> into your interpretation:<div>&nbsp;</div> "SPC's are a meaningless tool by which to measure or compare enterprise storage arrays."<div>&nbsp;</div> ...would require that all IT departments run all workloads as a mix of applications in virtual server environments.<div>&nbsp;</div> So, first, not every customer runs VMware or any other virtual server technology. Those that do, don't do it for all of their workloads. Both IBM and VMware agree that some applications or workloads are better served on native machines instead of virtualized environments.<div>&nbsp;</div> So, SPC benchmarks are helpful to decide which vendors to choose from, which models might best fit a particular amount of workload, which models have the sufficient scalability to handle anticipated growth, and which models might make the "short list" for further evaluation, including in-house proof-of-concept analysis in a mixed application / virtual server environment.<div>&nbsp;</div> So, I dispute the term "meaningless", removing this makes the statement correct:<div>&nbsp;</div> "SPC's are a tool by which to measure or compare enterprise storage arrays."<div>&nbsp;</div> -- Tony

3 localhost commented Trackback

IBMer Barry Whyte weighs in here:<div>&nbsp;</div> http://www.ibm.com/developerworks/blogs/page/storagevirtualization?entry=missing_the_point_misdirections_misinterpretations<div>&nbsp;</div> Steve Duplessie from ESG talks about the ESG Lab here:<div>&nbsp;</div> http://esgblogs.typepad.com/steves_it_rants/2008/10/competitive-bus.html<div>&nbsp;</div> Marc Farley from 3PAR<div>&nbsp;</div> http://www.storagerap.com/2008/10/anarchist-punches-out-ibm-and-tms-benchmarks---more.html<div>&nbsp;</div> ...posted his comment on BarryB's page. Here's an excerpt:<div>&nbsp;</div> But I don't understand why EMC has it in for the SPC so badly. Sure there are imperfections in their system, but they do provide a vendor-independent workload to measure and 3rd party verification of the results. Those are two important steps towards objective, meaningful results. My personal complaint is that the information in Appendix C where the details of the configuration are listed is tough plowing and takes more effort than many customers can afford to give it. Heck, even you didn't have time to chase it down - somebody had to dig into it an email you what they found. That's obviously not working. My guess is that the SPC would like to figure out how to make this info more transparent and easier to digest, but that is very difficult considering the multiple configuration and tuning options used by all the various vendors.<div>&nbsp;</div> So, I understand your call for an overhaul, but I'm not sure that starting from scratch is the best way. From your perspective, what's keeping the SPC from being more effective and useful for EMC?<div>&nbsp;</div> Also, do you really think the problem of defining a "representative" workload will ever be resolved? I have a hard time believing vendors agreeing on something like this. That's the value of having an independent entity do it - warts and all.

4 localhost commented Trackback

Tony - I see you haven't lost your skill in twisting words, even though it appears you are no longer in marketing.<div>&nbsp;</div> But please, do explain to me how running a mix of applications on virtual servers against a single storage array differs from running a mix of applications on multiple physical servers? Or how about simply running a mix of applications on a single server, be it UNIX, Windows or Mainframe? From the storage perspective, are the I/O patterns any different between the three?<div>&nbsp;</div> I think not, and thus by derivation, the SPCs can't help IT managers understand what happens when a mix of applications are deployed together in a server environment - the word "virtual" is an unnecessary qualifier!<div>&nbsp;</div> And while you're at tell - tell me exactly how many customers you know who run a single application on a single server against a DMX-4 or a USP-V? Now, given the performance limitations of the mid-tier architecture DS8300 Turbo, you might actually have more than a few on that platform. But I can assure you, that would be the extremely rare exception for a REAL enterprise storage array.<div>&nbsp;</div> And what - no update on that guarantee I asked about?

Add a Comment Add a Comment