What is the future for HP's high-end XP storage arrays? Which way will incoming HP server-networking-storage boss David Donatelli jump? Donatelli left EMC to run a combined server, storage, and networking operation at HP. The company is busy creating an integrated IT stack consisting of servers, storage, networking, and system …
another thought, and some corrections...
another possibility is that HP thinks this traditional monolithic high-end array market will shrink away to nothing over the next few years to be replaced by a combination of mid-range offerings and a "storage grid". I seem to recall some folks in HP talking about this a few years ago. In that scenario I could see HP continuuing its relationship with Hitachi right up until that market becomes unsustainable whilst delivering their own storage grid capabilities on commodity hardware.
and HP has no OEM relationship with HDS. Both HP and HDS OEM the USP-V from Hitachi Japan. Sun then OEM the OEM'd HDS version of the array. The point is that HP are not dependent on HDS for their XP business. Hitachi is the parent company of HDS.
don't give them any ideas
I want 3par to stay independent! damnit.
what about Z
If HP bought up 3PAR and hence probably butcher their relationship with HDS, then they'd lose Z attach - therefore I don't see it happening.
HP correction: it's Hitachi and not HDS
Sent in by HP's XP product manager:-
With respect to your article on XP and Hitachi, let me make a couple of important distinctions. HP has no relationship with HDS. HP has both an Engineering partnership with Hitachi Japan and an OEM sales relationship.
This distinction is significant to a premise of your article that HP prefers to have control over the technology it ships. HP participates very actively in the design and core technology of the XP product family. We do not control it per se, but we have a very active involvement in the creation of the products. This engineering relationship has existed from the begining of the XP product introduction by HP, and was in fact a key component of our decision to form a relationship with Hitachi. The engineering relationship is directly with Hitachi Japan, the creators and owners of the technology, not through HDS. Sun, by contrast, is simply a reseller of the HDS technology and has no such input or connection with Hitachi Japan.
Regards, James R. Wilson XP Product Manager Hewlett Packard
This clarification is appreciated,
RE: another thought, and some corrections...
Storage pooling would seem to be the way to go, but as hp says, it needs to be scale up as well as scale out. The idea of pooling several mid-range arrays such as the EVAs into one SAN device, with front-end bandwidth to match the XPs, would probably get us to ditch the XPs we have. I love 'em, but they cost two arms and three legs! But storage virtualisation tech like SVC and SVSP just isn't quite there yet, so we'll probably end up buying at least one more gen of XP for our really core mission crit stuff.
FC vs NFS
I think the days of the fat FC array are coming to an end. FCoE is a stepping stone to using NFS pretty much everywhere in the distributed space. Why bother with all that extra FC complexity & cost when you can run directly over NFS for the vast majority of applications? 10GbE and DCE gives you the bandwidth and and low latency... IP networks can be designed every bit as reliable as FC and you don't get that hideous interoperability mess that FC forces on you. So, who needs FC except for non-virtualised Windows apps? It's all going IP & NFS folks.
Bye bye big arrays
As a couple of people have alluded to, the end of the high end enterprise class array is nigh for all but a few requirements. I'd imagine HP would keep the XP in their portfolio as it is unrivaled in the enterprise space, but you are likely to see more scale-out, grid like solutions from them ... and who is better positioned to deliver commoditised storage hardware acting like and delivering enterprise class functionality and performance than HP? With their supply chain, they could be in a position to really drive the direction of things to come.
3PAR is Enterprise High-End Storage?
Sorry, most customers do not consider 3PAR to be the same class of storage as HDS USP-V EMC DMX/V-Max or IBM DS-8000. The features which matter in this market are 100% uptime, replication which works, and responsive, on-site vendor provided service.
Looking at 3PAR's web site, I do not see an asychronous replication solution, just sychronous, which limits disaster recover options.
HP EVA and NetApp both have more enterprise data protection features than 3PAR. Many customers consider NetApp to be class-leading in midrange storage, but most customers do not consider NetApp in the same class as USP-V, EMC DMX, or IBM DS-8000.
Perhaps HP, like IBM did with the ESS, can create an enterprise class high-end solution. Maybe 3PAR's technology can help HP create a super EVA platform. But the market does not consider 3PAR's T-Series to be a high-end enterprise solution today.
Agree with Nate wholeheartedly!!
I would imagine quite a few of the top storage vendors would like to either purchase or kill off 3PAR! But please don't give HP any ideas! We much prefer 3PAR to remain independent.
HP has done very little with PolyServe since its purchase (in South Africa at any rate) - a fantastic solution particularly when it comes to highly available SQL consolidation on industry standard servers - they prefer to push SQL on the Superdome, a very expensive, not so great solution!! Most of the market here doesn't even know what PolyServe is!
In terms of the enterprise debate, our customers are very happy to adopt 3PAR in the enterprise within open systems and run fully replicated systems in Production & DR. 3PAR offers both synchronous & asynchronous replication, is definitely highly available, and offers the best in class performance - particularly within virtualized environments and for mixed workloads. Many customers are seeing 3x the performance with mixed workloads on the T-series compared to that achieved on traditional enterprise storage arrays! That's 300%, not 30%! Who else can offer that sort of performance improvement!!
Amongst many other industry leading solutions (Thin Provisioning, Thin Persistence, Thin Reclamation), 3PAR's ease of management and fully automated provisioning truly sets it apart from traditional storage arrays!! No highly paid storage specialists needed to manage this environment nor exorbitant vendor's professional fees!
Best in class performance? Is that enterprise class or mid-range then?
If you are saying that the 3PAR array is enterprise class then it really doesn't have best in class performance ... the USP-V / XP will out perform anything out there in the enterprise space. I don't think any of the big players are in that much of a rush to kill off 3PAR as we never come up against them! They really aren't on many people's radar!
RE: FC vs NFS
Been there, done that! First, when 100Base-T was finally standardised, everyone was running around pronouncing that NFS was going to trounce the new boy fibre channel. Well, that didn't quite happen! Instead, we got severly congested networks, and then along came Gigabit and everyone was sprouting the same story again, only we didn't fall for it the second time. Meanwhile, dedicated SAN networks had come of age, FC swept the board and is now so deeply entrenched even FCOE/"Converged-Ethernet"/whatever-it's-being-flogged-as-tomorrow still is unlikely to replace many dedicated SANs in the immediate future.
Our network bods are not keen on FCOE. They don't "do" storage, they don't particularly want to, and they don't want to share their toys with the SAN guys. Meanwhile, the SAN guys don't want to lose precious bandwidth to Internet downloads, email (BIG network hog) and the threat of more VOIP apps. It's going to be fun to watch!
- +Comment Trips to Mars may be OFF: The SUN has changed in a way we've NEVER SEEN
- Vid Google opens Inbox – email for people too stupid to use email
- Back to the ... drawing board: 'Hoverboard' will disappoint Marty McFly wannabes
- Pic Forget the $2499 5K iMac – today we reveal Apple's most expensive computer to date
- Google+ goes TITSUP. But WHO knew? How long? Anyone ... Hello ...