According to the Briefing.com
rumor mongering financial news site, software giant Oracle - not even finished with its $7.4bn acquisition of Sun Microsystems - is sniffing around middleware and virtualization maker Citrix Systems and may be interested in acquiring the company.
In the same report, you'll find that Briefing.com is also hearing chatter about someone wanting to acquire Rambus, with HP being mentioned as the possible buyer of that memory company.
That Oracle would be interested in buying Citrix is totally plausible. Oracle has its own implementation of the open source Xen hypervisor, which was based largely on the work done by Red Hat as it commercialized Xen in Red Hat Enterprise Linux 5. When it became clear that Red Hat was going to shift to the KVM hypervisor when it bought Qumranet for $107m in September 2008, Oracle looked around for another set of Xen tools and snapped up Virtual Iron, which had its own variation on the Xen theme. But in August 2007 Citrix bought XenSource, the controller of the Xen project, for $500m - a very large sum for a company that had only $1m in sales at the time.
Oracle likes to be in control, as it soon will be with Java, so why not with x64-based server virtualization as well as application streaming and desktop virtualization? Oracle is a good fit for Citrix. But then again, Novell and Citrix could also merge. And as El Reg has pointed out more than once, IBM needs Novell and Citrix as much as Oracle might. Perhaps more.
On December 24, Citrix had a market capitalization of $7.7bn, which would make the company much more expensive to acquire than Sun. And considering that even as messed up as Sun currently it is, the Sun biz can generate at least $10bn and maybe even $12bn in sales along with perhaps $1bn in profits - and if Oracle slashes the heck out of the payroll, maybe even a little more than that. Citrix generated $1.6bn in sales in 2008 and brought $178.3m to the bottom line, which is not too shabby.
But it is hard to imagine anyone getting ahold of Citrix for less than $9bn to $10bn. Citrix will have to have a very bad 2010 and 2011 before Oracle could pay only $3bn for the company - which is arguably all it is worth based on revenues and profits. (I can see paying two times annual revenues for a reasonably profitable company, as Citrix is.) But you can't argue with Wall Street, which is why Sun, at the height of the dot-com boom, had a market cap over $200bn. Funny, isn't it? Not if you are a Sun ex-employee.
On the Rambus front, it is hard to see what HP would want with the company. HP used to be in the disk and chip businesses, but ditched them more than a decade ago. Maybe HP uses so much memory that it wants to be its own supplier, but if it does that, it will have exactly the same problem Big Blue had with its disk drives: no one wants to buy components from a systems competitor.
Rambus has settled its antitrust issues with the European Commission by dropping the licensing fees for its chip technology, which makes it a target. But with a $2.5bn market cap, Rambus would be a pricey acquisition relative to very small revenues. In the past five years, Rambus has raked in only $819.8m in cumulative revenues, and has booked $186.1m in losses. Why anyone would pay more than maybe $400m for Rambus is a mystery, unless it has some memory technology that HP thinks it can deploy in its servers. And make buckets of money.
This Rambus-HP talk sounds a lot like someone has been hitting the egg nog a little too hard, or someone is floating a trial balloon or engaging in a little wishful thinking. ®