RT Conference Proceedings
T1 Spatial and Temporal Cache Sharing Analysis in Tasks
A1 Ceballos, Germán
A1 Black-Schaffer, David
A2 Carretero Pérez, Jesús
A2 García Blas, Javier
A2 Petcu, Dana
AB Understanding performance of large scale multicore systems is crucial for getting faster execution timesand optimize workload efficiency, but it is becoming harder due to the increased complexity of hardwarearchitectures. Cache sharing is a key component for performance in modern architectures, and it has beenthe focus of performance analysis tools and techniques in recent years. At the same time, new programmingmodels have been introduced to aid the programmer dealing with the complexity of large scale systems,simplifying the coding process and making applications more scalable regardless of resource sharing. Taskbasedruntime systems are one example of this that became popular recently. In this work we develop modelsto tackle performance analysis of shared resources in the task-based context, and for that we study cachesharing both in temporal and spatial ways. In temporal cache sharing, the effect of data reused over time bythe tasks executed is modeled to predict different scenarios resulting in a tool called StatTask. In spatialcache sharing, the effect of tasks fighting for the cache at a given point in time through their execution isquantified and used to model their behavior on arbitrary cache sizes. Finally, we explain how these toolsset up a unique and solid platform to improve runtime systems schedulers, maximizing performance ofexecution of large-scale task-based applications.
SN 978-84-608-6309-0
YR 2016
FD 2016-02
LK https://hdl.handle.net/10016/22886
UL https://hdl.handle.net/10016/22886
LA eng
NO Proceedings of the First PhD Symposium on Sustainable UltrascaleComputing Systems (NESUS PhD 2016) Timisoara, Romania. February 8-11, 2016.
NO European Cooperation in Science and Technology. COST
NO The work presented in this paper has been partially supported by EU under the COST programme ActionIC1305,‘Network for Sustainable Ultrascale Computing (NESUS)’, and by the Swedish Research Council, carried out within the Linnaeus centre of excellence UPMARC, Uppsala Programming for Multicore Architectures Research Center.
DS e-Archivo
RD 17 jul. 2024