NetApp streaks ahead, avoids trip-up by bum flash leg

NetApp is a veritable money machine these days, and is currently clocking in growing fourth quarter and annual results, but a blip may have appeared on the horizon as it searches for a way to bolster its flash offering. The tech giant has signalled it will partner up to expand on flash technology and work on a few in-house …

[Anonymous Coward]

Why do you keep bringing up flash drives outside of the array?

Flash outside of the array makes no sense - you raise it as an issue repeatedly in your article. It's not relevant to enterprise storage. Flash storage as main storage is a relatively new technology that's had a few teething pains. There have been quite a few well publicized cases (Intel's self bricking drives, poor wear leveling algorithms from a number of manufacturers) that have resulted in some very high failure rates for what is otherwise an extremely low failure rate technology.. Using unprotected (non-raided) flash drives in enterprise systems (NetApp's market) makes no sense. Is there some reason you keep harping on this?

[Diskcrash]

Re: Why do you keep bringing up flash drives outside of the array?

Because it is what customers are demanding and where performance is best added and ignoring this is a quick way to lose those customers?

If you look at flash outside of a HDD based storage array as just an SSD then you are being a bit simplistic and most of the higher end products have ways to protect the data and the user with some better than others. EMC's VFCache does this by doing all writes as write through to the external storage which means that it doesn't improve write performance but neither does it reduce data reliability and you still get accelerated read performance which in most work flows is 80 to 90 percent of the traffic.

Not too shabby. For a small amount you get to boost your performance dramatically and don't reduce your reliability.

So why wouldn't you talk about flash outside of the array? Unless you don't have it as a product.....

[Anonymous Coward]

Re: Why do you keep bringing up flash drives outside of the array?

NetApp does have flash outside the array. You just described a NetApp PAM card.

[scorched_cpu]

Re: Why do you keep bringing up flash drives outside of the array?

PAM is inside the array as part of the read cache mechanism.

[Tom Maddox]

Re: Why do you keep bringing up flash drives outside of the array?

EMC's Project Lightning flash and HP's rebranded TMS flash drives are about onboard server read-only flash cache, not flash drives per se. The value in it is that it allows read traffic to be cached, potentially taking signicant i/o load off the array. The reason to put flash cache on the server as opposed to or in addition to the array is that it allows much more effective acceleration of that particular server's workload, which is good for your critical systems. NetApp's FlashCache is an on-array technology (not off-array as incorrectly stated above), which is good for some workloads.

Both NetApp and EMC are behind the curve with flash technology, and they can afford to be, for the moment. Eventually, both NetApp and EMC need more compelling flash offerings; EMC seems to be on the ball, finally, with its XtremeIO acquisition, which leaves NetApp looking a little clueless on that front.

[Diskcrash]

E series

NetApp says that the E-Series exceeded their target expectations but doesn't say what those were. Considering that the impetus for LSI to sell the division to NetApp was how badly they screwed up the relationship with IBM and the IBM DS sales were going to fall off a cliff, so it would be interesting to see if NetApp has been able to repair that. IBM still has the E-Series for sale in the middle and entry level area and they are nice solid systems if a bit simple in their feature set and integration.

For something that should be big news the release of Ontap 8.1 Cluster Mode is a bit a failure both as an actual release and as a marketing event. For a company that keeps saying it is an innovator in the storage market you don't see that much innovation actually making it out and 8.1 has virtually no innovation and comes with a lot of potential pitfalls.

Is NetApp a one trick pony that needs to be put to pasture? Quite possibly but they still have a lot of potential but like my nephew they seem to be wasting it. Shame really.

[flashguy]

Re: E series

I don't remember if it was in the conference call or supplementary commentary with the earnings release, but NetApp said E-Series OEM exceeded forecasts (700 mil as opposed to 600 mil forecast); but it also said E-Series as a whole (OEM+NetApp branded) was a slight miss. I've never been able to track down what the forecast for NetApp branded E-Series was, but the numbers indicate NetApp itself didn't sell a whole lot. Still it appears revenue was slightly higher than it was under LSI-- just under 800 mil I'ld estimate, with most of that coming from Teradata in the midrange and IBM with the entry level.

I'ld agree OnTap 8.1 hasn't made much of splash; but I don't know why. Are there problems with it?

[flashguy]

weird conference call

I suppose given the IT sector's general bearish forecasts of late, NetApp may not have had much choice, but it was a weird conference call given Georgens confident assertion ntap would grow another point in TAM this year, while at the same time claiming they didn't have enough visibility to forecast yearly revenue plans, and forecast an extremely conservative Q1. The plan must either be to sell a whole bunch of FAS and E-Series low-end 2xxx boxes or buy another billion dollars in revenue? Is there another Engenio out there?

[scorched_cpu]

Chris Mellor has forgotten

Funny but the author has forgotten his own article about netapp's off array flash and integrated links to ONTAP, way better than EMC will have.

http://www.theregister.co.uk/2011/06/17/netapp_project_mercury/

So while NetApp has yet to acquire a company similar to xtremeIO, you can rest assured they are looking at lots of technology. And the OEM E-series deals, while limited had one of the biggest wins, the 55PB lustre FS for the Sequoia Supercomputer at LLNL.

[Tom Maddox]

Re: Chris Mellor has forgotten

That's nice. Wake me when they have a shipping product. It's way too late for NetApp to be "looking at lots of technology;" they need real product on the roadmap, or they're going to get their lunch eaten by EMC on the one hand and an assortment of leading-edge smaller vendors on the other.