There are nice earners, and then there is VMware. In 2004 EMC paid $635m to buy the processor virtualisation specialist, a price that some thought a little too high. How wrong they were. This month EMC pocketed nearly $1bn by selling off just 13% of VMware.
VMware was one of the early purchases in the shopping spree that saw EMC buy over 30 companies between 2002 and 2006. As it did for all those other purchases, EMC claimed a technological motivation for its acquisition of VMware, and said it would 'advance the convergence of server and storage virtualisation'.
The storage giant even talked of combining VMware's software with EMC's data mirroring systems to provide combined server and storage failover systems.
Do not hold your breath waiting for any such products or convergence to be realised. Four years on, the closest that the technology streams of EMC and VMware have come is with EMC's assurance to customers that its relevant storage hardware and software now connects to VMware virtual servers just as well as it does to real servers. A lot of other storage suppliers have made exactly the same assurance, and they did not have to buy VMware to do so. (Although they might wish they had done, simply for the cash.) In reality EMC was probably a little uncertain of how the processor technology would fit with the rest of its products. But it must have decided that the purchase would at least make sound financial sense.
The storage giant also very likely reasoned that the acquisition would upset the long-term plans of its rival Veritas Software, by keeping VMware out of Veritas' hands. Veritas had almost certainly been eyeing VMware as an excellent way to plug a major gap in its utility computing plans.
VMware was fairly typical of the type of EMC purchase that led many to accuse it of becoming a second CA, of the old model when CA was led by Charles Wang and simply milked a series of software acquisitions for their maintenance revenues.
It not a valid comparison. EMC has grown the revenues of its larger acquisitions, and unlike CA, it has just reported its 16th quarter of double-digit revenue growth. There is no sign of any intention by EMC to do anything but preserve the health of its acquisitions.
But if EMC is not like the old CA, is it like the new CA? Under new management, CA is continuing to buy up smaller companies, but this time with a much stronger emphasis on merging their technologies with its own.
The answer is that EMC is not entirely like that either. EMC's much larger purchases of Legato and Documentum were also pitched as leading the way to some sort of technology convergence that would fulfill EMC's then fashionable ILM strategy. To create systems that move data down through tiers of storage as it ages, EMC needed to add content management, backup management, and e-mail archiving software to its portfolio.
Has EMC merged all those products into one automated ILM system or integrated them more closely with its hardware products? No. Integration has mostly been limited to convergence of Documentum and Legato so that the two products store data in a common repository.
There is a very similar story to be told about EMC's other big purchase, RSA Security. For all of EMC's larger acquisitions, the products have been little changed or integrated with the rest of EMC's portfolio.
Customers might just as well still be buying them from the original, pre-merger vendors.
But the real point is that customers prefer to buy them from EMC, because they want deal with as few vendors as possible. They want to eliminate the complexity of handling multiple suppliers and support contracts, and they want as few throats as possible to have to choke when technical problems occur. It is the eternal appeal of the one-stop shop over the row of boutiques.
Size has always counted, but over the last few years it has become more important than ever in the IT industry, as the number of server makers dwindles and only a few giants consolidate their grip on the market. IBM and HP are close to becoming $100bn companies.
There is another very big reason why EMC has bought those large software companies: it needs to sell more software to compensate for the continuing erosion of the profit margins it sees on its storage hardware products.
But the practice of claiming a technical basis for a merger or acquisition that is in reality driven by the desire for bulk or diversification is not limited to EMC, or to companies expanding from hardware into software.
The merger of software giants Symantec and Veritas was promised to deliver multiple benefits by combining security and storage technology.
That was two years ago, since when we have seen only limited integration of the two companies' products.
It was never obvious how Symantec's firewalls and virus checkers could be combined with Veritas's backup management and storage virtualisation tools. But it was obvious that Symantec was facing a very serious challenge to its business from Microsoft, and Veritas had saturated its markets and could find no easy way to continue its growth.
Expect this pattern to continue. Citrix has announced a plan to buy fledgling VMware rival XenSource. According to Citrix, there are distinct technical reasons why as a thin client specialist it will be able to exploit the technology of a server virtualisation provider.