You can say anything you want about why IBM may be buying Sun Microsystems, but I think it's all about storage.
In the storage world, Sun has continued to innovate while IBM has continued to stagnate. IBM largely sells its own old technology and resells storage technology built by others, such as NetApp and LSI. Sun-although it's milked the technology acquired with the purchase of StorageTek-has recently returned to innovating.
IBM has been steadily losing market share in the storage market while Sun has been shaking things up with the new Storage 7000 Unified Storage System (look for my review this May), which pools multiple devices together to form a kind of network-based RAID and facilitates data migration across various hardware technologies housed in separate physical units. Coupling the Open Storage initiative with both commodity and high-performance storage devices in a tiered fashion has boosted Sun's revenue in recent quarters.
One aspect that is worthy of consideration is the mainframe storage play. Believe it or not, there is still a lot of big iron out there that costs companies a small fortune to maintain. Replacing a single storage platter on a mainframe can run thousands of dollars and does not move the company forward. Instead, it further locks it into continuing to run mission-critical apps on legacy (and expensive) hardware. Every company I've spoken to about this has said it can't wait to get off the mainframe, but at the same time simply can't migrate due to the mission-critical nature of the apps.
I strongly suspect that if this scenario is true, then IBM is making the move to acquire Sun because Big Blue is sick of Fortune 500 companies migrating off of the mainframes and onto someone else's (read: Sun, Hewlett-Packard, Dell) commodity hardware.
Think about this pathway: Company X is throwing money out every year struggling to keep that mainframe running optimally because it continues to host mission-critical apps. The writing is on the wall and they know they need to get off the mainframe. Now, they can hire IBM Global Services to come in, partition the mainframe to also run OpenSolaris, and use Open Storage to add primary, secondary and tertiary storage using commodity storage hardware instead of expensive proprietary mainframe storage.
This applies especially to backup and archive, where needs continue to grow. While mainframe storage hardware prices go up over time, prices of commodity storage that can be accessed using Open Storage come down.
Matthew D. Sarrel is executive director of Sarrel Group, an IT test lab, editorial services and consulting firm in New York.
More eWEEK Labs analyses about the effects of a possible IBM/Sun merger: