A guide to help choose a Category of access level for HPCVL.

The following discussion assumes that the Central site main production pool has over 968 CPUs as at Aug., 2006. It is only meant to illustrate the basic idea of resource usage. There are other resources that HPCVL may offer to groups that may come into play as well.

General Institutional Access (Category 1)

Category 1 access level should allow a PI (with a small group) to access the resources of HPCVL in a low-key manner. That is, small numbers of CPUs per job (say up to 12) and not burdening the systems all the time. So, should a Cat.1 PI have 3 graduate students, each submitting jobs that take 4 to 12 CPUs, the total usage might be using 32 CPUs 50% of the time. Alternatively, a Cat. 1 PI may have only 1 graduate student but this graduate student requires 32 CPUs for 12 hours each day. This is not yet approaching the point where resources might be denied other users and this PI would be okay here but perhaps would be asked to move to Category 2, the Enhanced Institutional Access Category, if their average monthly usage approached 2% or more of the cluster's resources.

Another example is a group that starts out using 6 to 16 CPUs with 1 user. More members of the group sign up and 3 users end up collectively using 48 CPUs 60% of the time (roughly 3.0% of the available resource). This is a situation that would require the PI move from a Cat. 1 level to a Cat. 2 level. Flexibility should be preserved as one day a group might have 3 people needing 48 CPUs altogether but 90% of the time they use only 12. Constant monitoring is required and we understand that minor bumps need not mean that people have entered into a more computationally demanding phase of their research programs. We are also keeping in mind that research capability is why HPCVL exists.

Enhanced Institutional Access (Category 2)

The Category 2 access level should allow a PI and their group to access a larger amount of resource. Although the restriction on numbers may be implicit for the Cat. 1 level, for the Cat. 2 level there should be no such implicit restriction. These groups could involve 1 or more users whose computing needs exceed 48 CPUs say 60% of the available time. Say a group uses 60 CPUs on average 60% of the available time. At present, that would constitute 3.7% of the available 968 CPUs.

With the installation of additional and more powerful CPUs, the percentages may drop and we will be updating these guidelines as needed. Throughout all of this is the understanding that the Cat. 2 access level has a slightly higher level of priority. Should a user in a group need even higher priority and more resources (say 160 CPUs for 6 weeks), a special fee may need to be paid to administer this service. However, provided the Cat. 2 PI whose group may need say the 160 CPUs for 6 weeks is making every effort to optimize their code, this type of work should be encouraged. It is an example of capability computing that we strive for. We may be resource bound at times and be unable to provide this service. We may also have a special pool of CPUs available at times to facilitate this type of work at some point.