From trace generation to visualization: a performance framework for distributed parallel systems
Proceedings of the 2000 ACM/IEEE conference on Supercomputing
A Dynamic Periodicity Detector: Application to Speedup Computation
IPDPS '01 Proceedings of the 15th International Parallel & Distributed Processing Symposium
A new data compression technique for event based program traces
ICCS'03 Proceedings of the 2003 international conference on Computational science: PartIII
Scalable parallel trace-based performance analysis
EuroPVM/MPI'06 Proceedings of the 13th European PVM/MPI User's Group conference on Recent advances in parallel virtual machine and message passing interface
Introducing the open trace format (OTF)
ICCS'06 Proceedings of the 6th international conference on Computational Science - Volume Part II
Hi-index | 0.00 |
We are developing a novel performance measurement technique to address the scalability challenges of event-based tracing on high-end computing systems. We collect the information needed to diagnose performance problems that traditionally require traces, but at a greatly reduced data volume. Performance analysis working on today's high-end systems require event-based measurements to correctly identify the root cause of a number of the complex performance problems that arise on these highly parallel systems. These high-end-architectures contain tens to hundreds of thousands of processors, pushing application scalability challenges to new heights. Unfortunately, the collection of event-based data presents scalability challenges itself: the added measurement instructions and tool activities perturb the target application; and the large volume of collected data increases tool overhead, and results in data files that are difficult to store and analyze.