article · 2015
GOWNIVARIPALLI, YASASWINI JYOTHI. Hybrid Cache, Bank, and Controller Aware Coloring
Cited by 0Abstract and details only
Abstract
Real-time systems require tight bounds on the Worst Case Execution Time (WCET) of a task. Modern multicore CPUs provide multiple levels of storing data (caching) and differ-ent paths to access memory (banks) that improve the performance but also introduce unpre-dictability in task execution time. To obtain predictable task execution times, developers have to consider the WCET, which may void the benefits of modern CPUs and may lead to low utilization of resources. On multicore platforms, caches present a source of unpredictability as standard heap al-located regions do not provide guarantees on the cache set that will hold a particular page translation. This unpredictability can lead to cache misses and consequently inter-task inter-ference resulting in loose bounds on the WCET. Memory access latency varies significantly depending on where the data is located in Non-Uniform Memory Access (NUMA) systems and how banks are interleaved. Data allocations without locality awareness may experience high memory latency so that execution times become highly unpredictable, resulting in loose bounds on the WCET of tasks and overly conservative scheduling with low utilization of modern CPUs.
Cite this paper
(2015). GOWNIVARIPALLI, YASASWINI JYOTHI. Hybrid Cache, Bank, and Controller Aware Coloring.
“GOWNIVARIPALLI, YASASWINI JYOTHI. Hybrid Cache, Bank, and Controller Aware Coloring.” 2015.
2015. “GOWNIVARIPALLI, YASASWINI JYOTHI. Hybrid Cache, Bank, and Controller Aware Coloring.”
(2015) ‘GOWNIVARIPALLI, YASASWINI JYOTHI. Hybrid Cache, Bank, and Controller Aware Coloring’.
@article{anon2015gownivaripalli,
title = {{GOWNIVARIPALLI, YASASWINI JYOTHI. Hybrid Cache, Bank, and Controller Aware Coloring}},
year = {2015}
}Made from the index record. Check names, capitalisation and page numbers against the paper. Need another style? AI for Research has over 10,000.