Argonne National Laboratory

Application Power Profiling on IBM Blue Gene/Q

TitleApplication Power Profiling on IBM Blue Gene/Q
Publication TypeConference Paper
Year of Publication2013
AuthorsWallace, S, Vishwanath, V, Coghlan, SM, Tramm, JR, Lan, Z, Papka, ME
Conference NameCluster Computing (CLUSTER), 2013 IEEE International Conference
Conference LocationIndianapolis, IN
Other NumbersANL/MCS-P5067-0114

The power consumption of state of the art supercom­ puters, because of their complexity and unpredictable workloads, is extremely difficult to estimate. Accurate and precise results, as are now possible with the latest generation of supercomputers, are therefore a welcome addition to the landscape. Only recently have end users been afforded the ability to access the power consumption of their applications. However, just because it's possible for end users to obtain this data does not mean it's a trivial task. This emergence of new data is therefore not only understudied, but also not fully understood.

In this paper, we provide detailed power consumption analysis of microbenchmarks running on Argonne's latest generation of ffiM Blue Gene supercomputers, Mira, a Blue Gene/Q system. The analysis is done utilizing our power monitoring library, MonEQ, built on the ffiM provided Environmental Monitoring (EMON) API. We describe the importance of sub-second polling of various power domains and the implications they present. To this end, previously well understood applications will now have new facets of potential analysis.