May 23, 2002

Sun vs. Linux on mainframes: Round V (and VI?)

- By Grant Gross -

If you think the argument over Linux on mainframes has cooled, think again. This week, Alexander Tormasov, the chief scientist at Linux virtualization company SWsoft, has crafted a response to Sun Microsystems chief competitive officer Shahin Khan's response to Tormasov's first letter that responded to Khan's "Linux on the Mainframe -- Not a Good Idea." Tormasov's new letter questions Sun's business model, and suggests that's why Khan is working so hard to discredit Linux on mainframes.
Okay, for everyone who's confused, let's try this again, working forward instead of backward:

Khan publishes his Linux-on-mainframe critique Feb. 20.

Tormasov and Linux coder Andrey Savochkine respond to Khan on March 5, in a letter endorsed by Red Hat, Turbolinux, Caldera and SuSE. Ironically, SWsoft doesn't work with Linux on mainframes -- it uses virtual machine technology pioneered on Intel-based machines.

IBM, which helped bring Linux to the mainframe, defends itself in a letter by Mark Cathcart of the IBM Academy of Technology, IBM eServer Architecture and Design.

Khan fires back April 9 with "Mainframes Running Linux, Wolves in Sheep's Clothing," taking on both the Tormasov/Savochkine and Cathcart letters. Khan defends Sun's place in the Open Source community by saying the company's "donation of millions of lines of code and intellectual property to the open source community over the years surely qualifies it as a member of the community." But the letter still prompts yet another round of questions about Sun's new-found commitment to Linux.

Then on Tuesday, a couple of days before Sun announced Solaris 9, Tormasov writes again, offering a technical critique of Khan's arguments while dissecting Sun's business model.

In addition, a couple of tech analyst companies are now jumping into the fight. A week ago, ZDNet published an article featuring the conflicting views of Giga and META. "The META report says that mainframe configurations of Linux will fall out of favor as soon as 2005, by which time Unix and Intel-based Windows and Linux solutions will have enough mainframe-like features without the mainframe-like costs to make those options the better choice."

Back to the newest Tormasov letter: The associate professor at the Moscow Institute of Physics and Technology lists a number of technical points where he sees Khan's argument failing. Tormasov takes issue with Khan's assertion that "mainframe systems still only go up to 16 CPUs, when Sun made 64-way systems five years ago and has 106-way single-cabinet systems today."

Tormasov writes: "The 'Amdahl law' states that any workflow can be broken into parallelalizeable and serial (non parallelalizeable) parts. This shows that CPU number is not as important as how efficiently the server is used."

If all that stuff is just a bit too technical for your taste, Tormasov also questions whether Khan's objections to Linux on the mainframe are because of Sun's own "weak" market position. "If someone like Sun says this is technically wrong, it just means that Sun is afraid of such things, because their own efforts in the area may have some problems," he says.

From the Tormasov letter: "For years, Sun was content to dominate the market for general purpose open systems for mid-to-large enterprises. Unfortunately for Sun, Linux emerged as a formidable opponent. Linux on Intel servers are now winning in the mid-enterprise market for general purpose 'Open Systems.' Mainframes have long dominated the enterprise market and Linux on the mainframe suddenly turned the previously closed mainframe into the perfect 'Open System' for large enterprises. Customers wanting immediate access to Linux on enterprise hardware did not have to wait the three years technologists predicted it would take Intel-based servers to catch up to the mainframes high availability features. These large enterprise customers rapidly rushed to embrace Linux on the mainframe."

He continues: "Sun's position will continue to worsen as multi-processor Intel servers improve and new dynamic partitioning technologies like carrier-grade Virtuozzo for Linux on Intel enable Linux to beat Sparc/Solaris in both the mid enterprise and large enterprise markets. Due to its open nature, Linux already dominates in the hosting market and with rock-solid service provider solutions like HSPcomplete for Linux that enable much better cost management and hundreds of CPUs than a comparable Solaris solution ... "

If all this sounds a bit self-serving (Virtuozzo and HSPcomplete are SWsoft products, after all), Tormasov and SWsoft's PR rep admit that this pissing match with Sun doesn't hurt their company's profile in the Linux community and with its enterprise customers, just as it doesn't hurt Sun among customers interested in its non-Linux enterprise products.

But Tormasov says he and SWsoft, as members of the Linux community, also have an interest in seeing Linux succeed on the whole. "We're alerting our customers to the fact that Linux is now a viable business tool across the board, in every aspect of the data center," says Alex Plant, SWsoft's PR person.

Plant says SWsoft doesn't feel like it's carrying the torch for IBM on this issue. The latest Tormasov letter has been informally endorsed by IBM, but Plant says the inspiration for the letters has come from individual developers inside the company who've been offended by Sun's position.

Asked how long this debate has legs, Tormasov chuckles and says that depends on Sun. Adds Plant: "It seems that it's maintained a certain level of interest around this process. When that's going to subside, I don't know, but it doesn't seems that all this has derived from Sun or us, but from a communal perspective. People are talking about it ... and that to me is the real life of this debate."

Any bets on how long it'll take for Khan to respond, or when this debate will end?


  • Linux
Click Here!