Charter for Discovery Project on Athena/SGI Strategy aka CEASE (Convening to Examine the Athena SGI Effort) Discovery Question ------------------ Should we continue devoting the fiscal and human resources that are now going toward purchase and support for the SGI/Athena platform, or free them up for other uses? Scope ----- This project is essentially a fact-finding mission, to gather information necessary for deciding on our future SGI/Athena strategy. In particular, this project will focus on: - whether to freeze SGI support at IRIX 6.5 - viability and timeline of SGI/Athena support, including options for FY2001 renewal - scenarios for supporting academic use of SGIs in individual departments if SGI/Athena support is scaled back Because this question impacts Athena development, cluster machine renewals, and departmental proposals, it is important to do this work now so that we are prepared to allocate staff resources efficiently, avoid disrupting the Athena release cycle, and generally be prepared for next year's machine purchases. We have frozen this year's departmental purchases for SGI O2s, and are accepting no new requests pending the outcome of this discovery effort. Current environment ------------------- Athena first started supporting SGIs in response to particular departments that use specific software applications only available on this platform (MSI and BioSym used by Chemistry, Chemical Engineering, and Material Science, and Alias Wavefront used by Architecture). There continues to be a small subset of our customer base who depend on these applications. Historically, there has also been a small amount of courseware ported only to the SGI and some reliance on SGI-specific graphics capabilities in MIT subjects. In the public clusters, there are currently 90 SGIs (out of 416 total, or roughly 22%). The breakdown by machine type and renewal date in the clusters is: type number purchased 4-year renewal ---- ------ --------- -------------- Indy 44 FY1997 FY2001 O2 20 FY1999 FY2003 O2 26 FY2000 FY2004 With purchases about to be made this FY, there will be a total of 60 SGIs in academic departments (36 up for renewal in FY2001), and 31 SGIs for IS staff and SIPB (17 up for renewal in FY2001). The support costs of each IRIX release are disproportionately high relative to the ratio of SGIs to Suns. Added support costs related to SGI/Athena include development (Athena port and OS troubleshooting), SGI liaison efforts, SGI-specific 3partysw installation and maintenance; there also may be higher SGI-specific costs in terms of Cluster Services and OLC support. Because the SGI occupies more of a niche market than do Suns, there is less general expertise on our staff for supporting them, and particular idiosyncracies of IRIX tend to make troubleshooting less transparent. The purchase price per machine is also higher (slightly less than $5000 for an SGI O2 compared to slightly less than $3000 for a Sun Ultra5). The future direction of IRIX and SGI is something Athena developers have been watching for some time, and is always fodder for debate during annual release planning and machine renewal discussions. Although SGI states that they will continue to support IRIX for the next 4 years, the direction of the company remains unclear. Past experience has been that once a hardware vendor states its intention to move away from a particular platform or OS, third party software vendors tend to stop producing new releases for that platform. Given the spate of recent announcements from SGI, the time seems ripe to make a concerted effort to gather as much information as possible about our customer's needs, SGI's plans for IRIX support, and alternatives for the specialized third-party software (some vendors are looking at migrating to other platforms, some professors are looking at other applications that now run on other platforms). Discovery Project Work and Team Membership ------------------------------------------ The team is led by Abby Fox and Naomi Schmidt. Other team members are Bob Basch, Bill Cattey, and Alex Prengel. We anticipate this work will take roughly 4-6 weeks. The anticipated weekly time commitment for team members will be a 1 hour team meeting, plus an average of 3 additional hours for research and other tasks. The discovery effort will include interviewing representatives of other IS units who currently support SGI/Athena (in particular, Athena Consulting and Cluster Services) and faculty currently using SGIs in courses, as well as discussions with SGI and 3partysw vendors. Risks and Constraints --------------------- Risks of continuing to support IRIX: * commitment of resources that might be better spent elsewhere * implicit overcommitment of support for Athena SGI users in the event that SGI ends (or phases out) IRIX, or 3partysw vendors stop releasing new versions for IRIX Constraints for freezing IRIX support: * courses relying on software unavailable on other platforms * potential resistance from the community of SGI users Team Sponsor and Commitment Path -------------------------------- Vijay Kumar, IS Director for Academic Computing, is the sponsor for this project. Recommendations will be presented to ACMG and given to Vijay for final decision.