Joerg, over at c0t0d0s0.org pointed to a good article by Paul Murphy about the Economics of ZFS over at ZDnet.
He is responding to another article that proposed to estimate the OLTP price/performance of various boxes, including Solaris 10/ZFS boxes, without taking into account the cost advantages of fast software RAID.
The money quote
Murphy says the
. . . ratings ignore important differentiating technology – had he bench-ti-mated his Solaris costs and performance on a ZFS/JBOD basis, Sun would have placed at the top of every grouping.
All of which suggests an important question for readers of this blog: how many of you are making the same mistake? How about your colleagues and bosses?
As I recall, it took customers almost 8 years before RAID controllers and arrays became the dominant data storage hardware choice. How long will it take to unmake that decision?
Comments welcome as always, though my moderation may be slower than usual.
For those of you who celebrate Christmas, Merry Christmas. For those who celebrate other holidays – and I can think of several – happy holidays. For everyone else, take some time and smell the roses.
Thanks for the leadership and enlightenment you have provided in 2006.
In one of our environment we’ve been using middle-level arrays with SATA drives. Well it turned out that x4500 (aka Thumper) + ZFS is a killer platform when it comes to price/performance and price/storage – so we’re buying more. In another environment with external IBM’s entry level array we moved from Linux+PCI RAID controller to Solaris + ZFS and the array connected as JBOD. At the end performance is at least the same if not better but we did it mostly due to ZFS manageability (and overall system observability).
There is an interesting scenario playing out in Storage.
Take this phrase from the reply above, “we did it mostly due to ZFS manageability”.
Contrast this to Marc Farley’s comments, “Wade O’Harrow, one of our SE’s from the southeast was there and completely in his element as a VMWare animal. He kept describing our technology as “doing the same thing for storage as VMWare does for servers.” I admit it took me awhile to figure it out, but I finally did get it. In an EqualLogic SAN, the individual storage systems are really secondary to the logical storage pools that customers create. Volumes can be moved among the underlying storage hardware resources – more or less independently of what those resources happen to be. You need disk drives for virtual storage just like you need processors for virtual servers, but each disk drive just gets less and less important in the scheme of things.”
And the summation,
“I talked to customers from Europe, Asia, South America, Canada and the US who are creating new, large server infrastructures on top of VMWare because it saves huge amounts of administrative time. I’m not joking when I say that we haven’t seen a technology shift this radical since refrigeration became widespread.”
“Full speed ahead into V-Land”
Jeremiah Owyang’s writes about the convergence of “enabling” Enterprise Architecture and Web Strategy in a space named the Community.
Jesse James Garrett echoes this in his writings about the User Experience.
As does Peter Morville in Information Architect 3.0.
We are finally arriving at the “We no longer manage at the spindle level” destination. Seamless, transparent and invisible “enabling Technology.