Original URL: https://www.theregister.com/2010/01/28/oracle_sun_storage/

Oracle: Storage trouble in store

Sun's 7000 top of the class - who's getting expelled?

By Chris Mellor

Posted in Channel, 28th January 2010 16:37 GMT

Comment Oracle is putting a heavy focus on Sun's 7000, tape and flash storage products, implying the rest of the bought-in storage line is not so well-regarded and may be under threat.

Larry E and the new exec team at the big O have spoken and the storage winners and losers are becoming clear. In the winners' enclosure are the StorageTek tape line, the 7000 open storage line with ZFS, and flash-based storage products. Outside we see the ho-hum drive arrays, the 6000 modular line and 2000 workgroup storage line, the confused-looking 4000 set of storage servers and the 9000 line of resold HDS USP-V enterprise arrays, plus a few bits and pieces of storage management and protection software products.

Not everyone can be top, but does this mean there must be relegations? In the harsh light shining out from Redwood Shores it probably does. Oracle has little tolerance for second and third-rate products.

The StorageTek mainframe-class tape products are leading that market but the LTO products are not so well-regarded, and Sun is outside the HP, IBM and Quantum trio now controlling the LTO consortium and setting its direction. With tape being one the three specialities of the Oracle-ised Sun sales force, the StorageTek mainframe tape staff will be riding high and the LTO ones less so.

The second speciality is servers and the third is, well, storage, meaning flash and disk drive arrays and the storage software. Only the 7000 line of storage arrays and its software, including the Apple-rejected ZFS file system, has been mentioned at a top level.

That's not because of a high sales count, but because of the huge momentum the products have built up since they were developed. They are hot, whereas you get the distinct impression that the other storage arrays are not.

Oracle wants to sell integrated Oracle app-to-server-and-disk flatpacks. It hasn't said it wants to be a successful storage product supplier outside that remit, and with regard to the non-7000 storage arrays it isn't. Sun's rankings in the externally-attached storage market have been falling and it is a big supplier with ageing, across-the-line kit fallen on hard times, not a new and nimble startup with focussed hot boxes, like Compellent or 3PAR.

The high-end 9000s are sold in competition with HDS itself and with HP, which also OEMs them from Hitachi. There is a USP-V rev coming along which Sun may take, but you get the feeling Sun could bulk up the 7000 to fulfil the 9000's role in its product line. It might even look at Pillar's recently hardware-boosted Axiom and judge it could replace both the 9000 and modular 6000 arrays.

6000 line looking a bit peaky

The latter, the 6000 line, looks underwhelming as it faces intense competition from every other storage vendor and his brother. The gist of Sun's system storage development has been to closely integrate storage with servers and software, the 7000s and the 4000 storage servers for example. That leaves externally-attached dumb arrays like the 6000 out in the cold.

It is hard not to conclude that, until Oracle/Sun says convincingly otherwise, the 6000 line is treading water, and it's the same with the workgroup 2000 line. These two lines face intense competition from Dell, EMC, HDS, HP and NetApp, not to mention Compellent and Pillar and 3PAR's F Series for the 6000.

The Sun pair don't have any stand-out features and, with Dell and HP surging ahead with iSCSI array sales and HP having an EVA refresh coming, it's easy to see a cold-eyed Oracle exec staffer saying we're not in the volume x86 server business, so why should we be in the volume mid-range and low-end storage array business, especially when we don't have the volume?

What about the storage servers, the 4000 products like Thumper? Will Sun want to sell such products to companies wanting to run apps on them that compete with Oracle ones, particularly if they have no place in Oracle's own integrated IT flatpack? The Oracle line here might well be shape up and ship profitably without damaging Oracle's own sales - or die.

We might also say that Sun's preferred storage software is already in the 7000 software set, so everything else in the storage software line is not. Chances are that all these non-focussed products are going to get less investment, will fall further behind the competition and then wither. Oracle backs winners - it's not a support agency for products that aren't match fit. ®