2.025: r.i.p. ds8300

The pain and agony is finally over.

After nearly 3 years of denial, we now have proof-positive that the IBM DS8300 has been unceremoniously removed from life support. I've been told of numerous prospects whose IBM account teams vehemently denied the impending introduction of the DS8700 during the last quarter, even as EMC account teams asserted (with confidence) that the DS8300 was indeed on its deathbed.

True to my prediction back in February, the DS8700 intro is now inarguably imminent…see for yourself with this Google search. And the word is that many customers actually received quotes for the new DS8700 over the past several weeks.

Pity those who were suckered into buying a DS8300 this year (remember, I tried to warn you!)

This time I'll not play the role of truth-in-marketing (as I did for the XIV intro), so you'd better buckle your seat belts for another round of Meaningless Marketing coming from Big Blue as they try to convince you that the aging and decrepit Sharkitecture has been resuscitated with the magic face cream of P6 processors and the life-giving breath of flash drives.

i sincerely doubt that any of these will even come close to overcoming the inherent shortcomings of that architecture, though. Already I'm seeing outlandish claims that the DS8700 has ASTONISHING improvements for "distinct" workloads – which probably means they found some benchmark that looks good, even as the non-distinct workloads realize little or no new value.

Word is the DS8700 is part of next Tuesday's set of weekly IBM announcements, so it should be a fun couple of weeks in the blogosphere sifting through the FUD and marketing misrepresentations.

Cue TonyP!

 

technorati tags: IBM,XIV,DS8700,DS8300,Shark

Comments

[TimC]

Huh, that's a funny take on "what you said". I don't recall you EVER saying the DS8300 was going to be REPLACED, you did however claim the DS8000 line was "dead".

http://thestorageanarchist.typepad.com/weblog/2007/10/0045-pushing-da.html

Congratulations on now changing your story to "I told you the DS8000 was getting a hardware refresh". Here, let me make an astounding prediction as well: sometime in the next 3 years, the V-Max will be replaced by an updated piece of hardware". Were you also "warning" your own customers a year ago not to buy a DMX because you had a hardware replacement in the works?

[SRJ]

TimC - couldn't have said it better myself!

BarryB - Hilariously ridiculous.

[the storage anarchist]

Yuk it up, guys, but it's going to take more than a new processor, undersized (and over-priced) Flash Drives, a pay-for-use pricing scheme and a shiny new model number to make this aging architecture competitive.

If it weren't for mainframe attach, the DS8K line would have been shut down ages ago. And even with that lifeline, the z Team probably wish they'd put the old girl out of her misery - they probably could have avoided this latest round of charges of monopolistic behaviours if they weren't lugging around a clearly inferior storage platform to boot.

Face it, the Shark has clearly lost its bite.

[SRJ]

Yep - IBM's "aging" "inferior" POWER architecture just isn't suited for high performance storage...

Oh...wait...new SPC-1 benchmark world record?! With that same "aging" and "inferior" POWER architecture?!?!

This can't BE!!! DS8000 has been dying/dead for YEARS!!! Must. listen. to. EMC. FUD.

[william bishop]

Dunno Barry, you did walk right into that one....

[SRJ]

Forget about my comment?

[the storage anarchist]

Sorry, SRJ, been a busy few days.

The DS8000 series architecture is just a wee bit different than the SPC-killer P6 system. Aside from using short-stroked SSDs on a direct connect set of SAS bridges, and the fact that a DS8K can only be configured into TWO partitions, that AIX config avoided totally the overheads associated with external storage arrays.

As an example of how to optimize a server platform for maximum I/Os, it ain't bad, (although I suspect Sun's F5100 might give it a run), but the fact that the configuration left so many cores unused demonstrates that the arrangement was less than optimal.

I'll go out on a limb here, but I seriously doubt that same server could get the same results with any configuration of a DS8000 storage array as external storage.

But then, I've been proven wrong before...

[JosephZhou]

I dunno what you people talking about here --

About the new SPC-1 result, Power 595 with 32-way dual-core Power6. The test utilized 48 cores on 24 processors.

The DS8700 uses dual 4-way Power6 controllers (8 processors).
So DS8700 should do about 1/3 of the Power 595 SPC-1 result.
But that would be lower than DS8300 SPC-1 result!

Hence, this Power 595 number is meaningless for DS8700, because there is just no way Power5 outperforms Power6.

[the storage anarchist]

@JosephZou -

I've never seen anyone actually try to compare two obviously different implementation architectures by using simple math...and I honestly don't know what you are trying to say.

I will observe this, though, IBM is claiming that the P6's benefit the DS8700 to the tune of 50% more IOPS than the DS8300Turbo, and the PCIe interconnect replacement of the 8300's RIO-G results in 150% MB/s improvement.

Clearly, the performance of a storage array is not limited by CPU power alone...and I thus maintain my assertion that the POWER Server SPC-1 results bear no indication of the expected performance for the DS8700.

And in fact, those results should be disallowed by the SPC, since they are not a benchmark of external storage, it's a measure of a server with direct-attached JBOD.

It's an apples and bubble gum comparisons...

[JosephZhou]

@theStorageAnarchist

Yeah, that's exactly what I wanted to say - the new Power595 benchmark does not relate to DS8700 performance. :-)

I thought you were trying to relate - you said "but I seriously doubt that same server could get the same results with any configuration of a DS8000 storage array as external storage"

We are actually in agreement that there is no relation. Cool.