You are here
Service Unit (SU) Accounting
The Supercomputing Institute keeps track of machine usage using Service Units (SUs). General Service Units awarded to a group can be used on any high performance system except Koronis, while Koronis SUs can only be used on Koronis. There are two allocation types to help a PI distribute SUs to their group: Group Allocation, in which SUs are made available to all group members in common, and User Allocation, in which SU amounts are assigned to individual users. The default setting is for Group Allocation. If you wish to change your allocation type, or have questions about accounting and monitoring usage, send an email to MSI support.
You can monitor your usage by using the command acctinfo. This command provides a summary of SU usage for the user executing it.
More information is provided below.
There are two types of allocation ceilings as specified by a Principal Investigator: Group Allocation and User Allocation.
In Group Allocation SUs for the entire group are pooled together — when one person runs out the entire group runs out. This is the default setting.
In User Allocation the PI may assign SU amounts to individual members of their research group. To do so, or to move SUs from one member to another, please contact user support.
More about acctinfo
Users and Principal Investigators can see their combined usage for the current allocation period with the command acctinfo.
Usage for previous allocation periods can be checked by specifying the year, e.g. to check 2014-2015: acctinfo -y 2014
Along with SU usage acctinfo will report the percentage of usage remaining, days remaining in the allocation period and the percentage of days remaining in the allocation period.
Your group's fairshare targets and fairshare usage are also reported. For more information see: Fairshare Scheduling.
Example acctinfo Output
Below is an example acctinfo output for a Group Allocated non-PI user.
Determining/ Calculating SU charges
Service Units (SUs) are charged for computer time on HPC resources.
One SU will provide a fixed number of hours of CPU time as follows and for detailed hardware specifications, click on the system name in the following table:
|HPC Resource||CPU hours/SU|
To determine the number of SUs you require you will need to know:
a. how many processors (cores) your program runs on,
b. how long each run takes (in hours), and
c. how many runs you plan to do.
The product of these three numbers determines how many CPU hours you will need. Dividing the required CPU hours by the appropriate CPU-to-SU conversion factor found above will determine the number of SUs needed on the chosen machine. The table below provides a few examples for the Itasca and Calhoun systems:
|Application Information||CPU Hour Calculation||Itasca SUs||Calhoun SUs|
|a single-core application that takes 5.5 hours per run; 190 runs will be needed||1 core x 5.5 hours x 190 runs = 1,045 CPU-HRs||Not Allowed||1,045 CPU-HRs / 3.5 CPU-HRs per SU = 299 SUs|
|a 128-core application that takes 19 hours per run; 500 runs will be needed||128 cores x 19 hours x 500 runs = 1,216,000 CPU-HRs||1,216,000 CPU-HRs / 1.5 CPU-HRs per SU = 810,667 SUs||1,216,000 CPU-HRs / 3.5 CPU-HRs per SU = 347,429 SUs|
|a 2,048-core application that takes 24 hours per run; 75 runs will be needed||2,048 cores x 24 hours x 75 runs = 3,686,400 CPU-HRs||3,686,400 CPU-HRs / 1.5 CPU-HRs per SU = 2,457,600 SUs||3,686,400 CPU-HRs / 3.5 CPU-HRs per SU = 1,053,143 SUs|
Please see the Allocations page for details on Service Unit allocation eligibility, renewal, and the peer review allocation process.