SGI announced this afternoon that it has purchased the “software, patents, technology and expertise” of Linux Networx Inc. (LNXI) through an all-stock transaction. Although there has been no official statement saying so, LNXI is believed to have shut it doors. However, some of LNXI’s technology, particularly the Clusterworx Advanced software suite, appears to have a new lease on life.

This transaction returns both companies to center stage for inspection. SGI struggled famously for 10 years and emerged from bankruptcy vowing to reclaim its prior HPC status, and indeed, with former LNXI CEO Bo Ewald in charge, SGI is showing signs of improvement. LNXI was the poster child of the Linux cluster revolution, but the company struggled to retain valuable differentiation once larger vendors arrived and increased competition.

Tabor Research examines the potential impact of this announcement on SGI’s future, LNXI’s present, and the face of the HPC market.

What’s in it for SGI?

Tabor Research believes that there are three major reasons for a company to purchase all or part of another company:

1. Buy vs. build — It may be cheaper or faster to buy a product or technology rather than reproduce it in-house. In the case of technology owned by smaller companies the transaction may entail buying the entire company.
2. Buy vs. compete — Companies can purchase their competitors rather than compete with them. In this case the buyer acquires market share and access to a new client base.
3. Buy and sell — Very large companies will purchase and sell small companies as a financial investment strategy. In this case the purchasing company may believe that it can increase the value of the target company through better management, or economies of scale available through the purchasing company.

We believe that SGI moved to purchase LNXI’s assets primarily to acquire that company’s software suite, and hopefully to be able to retain the engineering staff. SGI has made employment offers to many LNXI engineers, but there is no indication yet as to how many offers will be accepted. We see the most attractive asset in LNXI’s portfolio as the Clusterworx Advanced cluster management suite, a comprehensive set of administration tools. Clusterworx Advanced could provide significant components for SGI’s Industrial Strength Linux Environment (ISLE) strategy.

Secondary benefits may come from an increase in services revenue and access to customers. Interestingly, there is marginal overlap in customers allowing SGI to potentially expand its footprint through helping LNXI customers through this transition.  SGI has stated it will no longer maintain the LNXI cluster products, alleviating the potential for product overlap and customer confusion. 

What Happens to LNXI?

LNXI appears to be making the best of a bad situation. The company was an early leader in the highly competitive, standards based cluster market. Product differentiation in this market was initially based on vendor expertise rather than hardware features. As more and larger companies entered into the market, and gained experience, LNXI’s ability to deliver clusters that could be brought into production quickly declined in relative value, and margin pressures subsequently increased. The company has been struggling for sometime.

After the purchase, LNXI will be able to show some form of return to its investors in the form of SGI stock, and keep at least some of its people and technology active.

At this point, the fate of LNXI’s customers is unclear. Although SGI offers heterogeneous system support, they will not inherit LNXI’s outstanding contracts, leaving existing users to buy services from SGI (or elsewhere) or go it alone.

Who’s in Charge? A Potential Family Reunion

An interesting side story on the purchase is that SGI’s CEO Bo Ewald once held the same position with LNXI, and LNXI CTO Dave Morton once held a senior product engineering role at SGI. Both executives are familiar with each side of the fence, although it is not known at this time whether Morton will rejoin SGI. These relationships may help explain the deal in that the cross-company experience may have shortened the negotiation and asset check-out periods. It may also bode well for the purchase in that SGI may have a good idea (i.e., few surprises) of what it is purchasing and how the purchase will fit into its overall product strategy.

Tabor Research Analysis

SGI is racing to get back into a leadership HPC role, and this acquisition could help. As one of the very few companies dedicated to HPC Linux clusters, LNXI developed technology assets that it did not succeed in delivering to a wide enough audience. These will provide some missing pieces in SGI’s product portfolio.

The transaction also may be a sign of healthy growth for SGI. A smaller company could not have managed the transaction, and the larger players (IBM, HP, Dell, Sun) may not have appreciated the value of LNXI’s bargain-priced assets. SGI has a long way to go, but they may now be pointed in the right direction.