OS vendors on the Merced starting grid

1999-09-01

We have said repeatedly in the past that Merced is likely a test bed technology, enabling the potential of the Intel 64-bit architecture to be established and paving the way for McKinley. However this is unlikely to prevent the inevitable operating system wars.

Merced silicon had barely left the fabrication plants before the rumour-mill started up. Currently the lines are being drawn between a 64-bit version of Linux and a Windows 2000 beta for IA-64. HP-UX and Monterey, both Unix flavours, are next on the list.

What can we expect to see? The most likely event is that Intel’s own release plan will determine the release plans for the operating systems. Microsoft has already announced, for example, that its 64-bit operating system will be available at the same time as Merced, “before the end of 2000.” Likely there will be similar announcements for HP-UX, Monterey and even Linux which will need a marketing push to position itself effectively against the Microsoft and Monterey spinning machine.

Then, there will be the inevitable betas and delays. It is Microsoft who are the most disadvantaged here, as Monterey, HP-UX and Linux development for IA-64 is already underway but Microsoft will not be able to fully commit resources until after the release of Windows 2000. Sure, developers will be working on 64-bit prototypes, but Gates and Ballmer are unlikely to risk W2K delays by reallocating developers elsewhere. Finally, we will see the inevitable benchmarks and bakeoffs, with each vendor, consortium or community demonstrating beyond reasonable doubt that their OS is the best.

There are several factors which will decide the success of the operating systems in the IA-64 space. The first is time to market: any delays will severely jeopardise the market’s interest. The second is application availability, which is less and less of an issue even for Linux. The third is down to marketing and acceptability. Here Microsoft is very strong but Linux is carving its own credibility. HP-UX is likely to have support, not least from the Hewlett Packard user community, but it is unlikely to be “the one”. The riskiest player has to be Monterey, which despite strong backing from the vendor community, may well end up as the Unix also-ran.

Overall, then, interesting times ahead. Obviously for us commentators, who capitalise on the gossip that this situation will generate. Also, though, for the vendor community, who even now may be staking the future of the company on which of the operating systems is likely to dominate. Finally, the user organisations have a clear decision to make, but will probably prefer to wait until the dust settles before linking their IT strategies to any specific operating system.

(First published 1 September 1999)