publication . Conference object . 2018

System Monitoring with lo2s: Power and Runtime Impact of C-State Transitions

Andreas Gocht; Robert Schöne; Philipp Joram; Thomas Ilsche; Mario Bielert;
Restricted
  • Published: 07 Aug 2018
  • Publisher: IEEE
Abstract
In this paper, we present improvements to the low-overhead performance monitoring tool lo2s. We combine a detailed recording of system events with information from a high-resolution power measurement, to record the scheduling of applications and C-state transitions. These sub-millisecond transitions are difficult to observe with traditional approaches. Our methodology provides us with insights into the behavior of the system before, during, and after C-state transitions of processors cores.
doi: 10.1109/ipdpsw.2018.00114
Subjects
free text keywords: Real-time computing, Performance monitoring, Kernel (linear algebra), System monitoring, Power demand, Scheduling (computing), Computer science
Related Organizations
Funded by
EC| READEX
Project
READEX
Runtime Exploitation of Application Dynamism for Energy-efficient eXascale computing
  • Funder: European Commission (EC)
  • Project Code: 671657
  • Funding stream: H2020 | RIA
Communities
FET H2020FET HPC: HPC Core Technologies, Programming Environments and Algorithms for Extreme Parallelism and Extreme Data Applications
FET H2020FET HPC: Runtime Exploitation of Application Dynamism for Energy-efficient eXascale computing
Download fromView all 2 versions
http://dx.doi.org/10.1109/ipdp...
Conference object
Provider: Sygma
https://dblp.uni-trier.de/db/c...
Conference object
Provider: Microsoft Academic Graph
http://xplorestaging.ieee.org/...
Conference object . 2018
Provider: Crossref
IEEE
Powered by OpenAIRE Research Graph
Any information missing or wrong?Report an Issue