This page archives the workload logs and problem descriptions presented at JSSPP workshops.

• List of all workloads
• Parallel Workloads Archive [external archive]

---

CERIT-SC workload

This workload comes from the 2017 paper "Analysis of Mixed Workloads from Shared Cloud Infrastructure" by Dalibor Klusáček and Boris Parák.
It contains real-life workload traces from the CERIT-SC system. Importantly, the traces describe a "mixed" workload consisting of both cloud VMs and grid jobs executed over a shared computing infrastructure.
Detailed description of the workload can be found in the aforementioned paper. The workload contains three major traces:

• Description of cloud VMs in JSON format [download log].
• Description of "grid worker" VMs in JSON format [download log].
• Description of grid jobs executed within "grid worker" VMs in extended SWF format (Standard Workload Format) [download log].
• Number of available CPUs in the shared infrastructure (per cluster and day) in Excel sheet [download sheet].

Newly added fields into the base SWF format represent: execution node(s) name, queue name, submit arguments of qsub command and the number of requested GPUs, respectively. SWF format description is available here.

Usage and Acknowledgments

The MetaCentrum workload log was graciously provided by the CERIT-SC and the Czech National Distributed Infrastructure MetaCentrum. If you use this log in your work, please use a similar acknowledgment. To acknowledge authors' work please consider citing the paper that introduced this log:

Dalibor Klusáček and Boris Parák, "Analysis of Mixed Workloads from Shared Cloud Infrastructure". In Job Scheduling Strategies for Parallel Processing, June 2017.