@article{8b5eeee167164dff9e6427dcade99368,
title = "Mitigating execution unit contention in parallel applications using instruction-aware mapping",
abstract = "Parallel applications running on simultaneous multithreading (SMT) processors naturally compete for execution units when their threads are mapped to the same core. This issue is further aggravated when such threads execute similar instructions that stress the same execution unit type, making their execution to behave very similarly as if the threads were running sequentially. This, in turn, will lead to performance degradation and underutilization of hardware resources. This work proposes a completely transparent framework (no modifications to the source code are necessary) that automatically maps threads of multiple parallel applications on SMT processors. The framework focuses on improving performance by mitigating the contention on execution units, considering each thread's instruction types, which are detected at runtime by our framework. Results show performance gains of 21% (geometric mean), compared to the native scheduler of the operating system.",
keywords = "execution unit contention, SMT processors, resource sharing, performance degradation",
author = "Serpa, {Matheus S.} and Cruz, {Eduardo H.M.} and Matthias Diener and Lorenzon, {Arthur F.} and Beck, {Antonio C.S.} and Navaux, {Philippe O.A.}",
note = "This work has been partially supported by Petrobras (2016/00133‐9, 2018/00263‐5) and Green Cloud project (2016/2551‐0000 488‐9), from FAPERGS, CNPq Brazil, PRONEX 12/2014 program and CAPNS (001). Experiments presented in this article were carried out using the Grid'5000 experimental testbed, being developed under the INRIA ALADDIN development action with support from CNRS, RENATER and several Universities as well as other funding bodies (see https://www.grid5000.fr ). Some experiments in this work used the PCAD infrastructure, http://gppd‐hpc.inf.ufrgs.br , at INF/UFRGS. information Conselho Nacional de Desenvolvimento Cient{\'i}fico e Tecnol{\'o}gico, Grant/Award Number: PRONEX 12/2014; Funda{\c c}{\~a}o de Amparo {\`a} Pesquisa do Estado do Rio Grande do Sul, Grant/Award Numbers: 2016/2551-0000 488-9; Petrobras, 2016/00133-9; 2018/00263-5; Coordena{\c c}{\~a}o de Aperfei{\c c}oamento de Pessoal de N{\'i}vel Superior, Grant/Award Number: 001This work has been partially supported by Petrobras (2016/00133-9, 2018/00263-5) and Green Cloud project (2016/2551-0000 488-9), from FAPERGS, CNPq Brazil, PRONEX 12/2014 program and CAPNS (001). Experiments presented in this article were carried out using the Grid'5000 experimental testbed, being developed under the INRIA ALADDIN development action with support from CNRS, RENATER and several Universities as well as other funding bodies (see https://www.grid5000.fr). Some experiments in this work used the PCAD infrastructure, http://gppd-hpc.inf.ufrgs.br, at INF/UFRGS.",
year = "2023",
month = aug,
day = "1",
doi = "10.1002/cpe.6819",
language = "English (US)",
volume = "35",
journal = "Concurrency and Computation: Practice and Experience",
issn = "1532-0626",
publisher = "John Wiley & Sons, Ltd.",
number = "17",
}