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

Geez, Tony. Obviously you are still "Green" (With Envy) over VMware, huh? And that IPO date is rapidly approaching!<div>&nbsp;</div> Readers Note: that "other vendor" who provides "operational improvements" by "virtualizing x86 servers" happily counts IBM as a major reseller of said "server virtualization technology."<div>&nbsp;</div> Go figure.<div>&nbsp;</div> And for the record: apples-to-apples, drive-for-drive, and port-to-port, the DMX easily requires less power to deliver more IOPS and MB/s than the equivalent DS8000. SATA drives are not the secret to using less power than the power-hungry P5-powered storage furnace (but, hey, they are a nice added bonus).<div>&nbsp;</div> And we're talking real, measurable power and cooling advantage, without the need to buy into a "sole-sourced" "mainframe" "dinosaur" just to be able to safely run multiple Linux applications on a server platform.<div>&nbsp;</div> Oh, and "all" "quotation" "marks" "are" "indeed" "intentional" :)

2 localhost commented Trackback

Gosh, I love it when all the bloggers blog about each other's blogs.<div>&nbsp;</div> I have to note that IBM's recent effort to replace lots of servers with a mainframe tickled me. I can't tell you how many large enterprise shops I have been to lately where the same idea is under consideration.<div>&nbsp;</div> Not necessarily to green the server environment, but to get control over storage. In one utility company, the CIO told me she was so sick of the obfuscation of her visibility into and management of EMC storage that she was contemplating deploying an IBM mainframe to serve as a front end for all of her distributed systems storage. Leveraging DFSMS and DFHSM, she could make short work of management and reduce costs enormously.<div>&nbsp;</div> I have also noted that MF are coming back into vogue as companies are encountering problems with performance of large databases, even in "virtualized" server environments. Bigger DBs belong on mainframes.<div>&nbsp;</div> As for the comments on competitive green features of EMC this and that versus IBM this and that, my comment is simple: green IT has nothing to do with hardware, but with data management.<div>&nbsp;</div> Thanks for the friendly links, by the way. I just paid you a kudo on DrunkenData.

3 localhost commented Trackback

The choice between big iron or x86 as a virtualization platform is never clear cut- IBM did good with their in house job, but that scale is rarely seen in the real (non-fortune 500) world. <div>&nbsp;</div> There are pros and cons to each side:<div>&nbsp;</div> Mainframes (or even Unix on Power) typically have a large minimum configuration, are more limited in what type of workloads/OSs they can host, but have better reliability than x86. This comes from having decades of experience building virtualization into the machines themselves.<div>&nbsp;</div> x86 virtualization comes with a hypervisor that can cost you up to 40% overhead and is still in the process of gaining acceptance for mission critical loads, however offers features that big iron with its decades of virtualization experience never even thought of- like moving virtual machines from one resource to another without bringing them off line.<div>&nbsp;</div> It comes down to (like so many things in the IT world) a company's individual situation. Figure out the variables, set goals, and decide what's best in your particular situation.

4 localhost commented Trackback

Changed to "rack-optimized" servers to reflect that there are multiple manufacturers of processors (Intel, AMD, IBM), multiple architectures (x86, x64, POWER3, POWER4) and do not want to sound exclusive.

Add a Comment Add a Comment