The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
A tecnologia multithreading simultânea (SMT) pode efetivamente melhorar o rendimento geral e a justiça, melhorando a eficiência do uso de recursos dos processadores. Trabalhos tradicionais propuseram algumas métricas para avaliação em sistemas reais, cada uma das quais estabelece um compromisso entre justiça e rendimento. Como escolher uma métrica adequada para atender a demanda ainda é controverso. Portanto, apresentamos sugestões sobre como selecionar as métricas adequadas através da análise e comparação das características de cada métrica. Além disso, para o novo cenário de aplicações da computação em nuvem, os data centers possuem alta demanda pela qualidade de serviço para aplicações matadoras, o que traz novos desafios ao SMT em termos de garantias de desempenho. Portanto, propomos uma nova métrica P-desaceleração para avaliar a qualidade das garantias de desempenho. Com base em dados experimentais, mostramos a viabilidade da desaceleração P na avaliação de desempenho. Também demonstramos o benefício da desaceleração P por meio de dois casos de uso, nos quais não apenas melhoramos o nível de garantia de desempenho dos processadores SMT por meio da cooperação da desaceleração P e da estratégia de alocação de recursos, mas também usamos a desaceleração P para prever a ocorrência de comportamento anormal contra ataques de segurança.
Xin JIN
Xi'an University of Technology
Ningmei YU
Xi'an University of Technology
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copiar
Xin JIN, Ningmei YU, "Which Metric Is Suitable for Evaluating Your Multi-Threading Processors? In Terms of Throughput, Fairness, and Predictability" in IEICE TRANSACTIONS on Fundamentals,
vol. E103-A, no. 9, pp. 1127-1132, September 2020, doi: 10.1587/transfun.2019EAL2155.
Abstract: Simultaneous multithreading technology (SMT) can effectively improve the overall throughput and fairness through improving the resources usage efficiency of processors. Traditional works have proposed some metrics for evaluation in real systems, each of which strikes a trade-off between fairness and throughput. How to choose an appropriate metric to meet the demand is still controversial. Therefore, we put forward suggestions on how to select the appropriate metrics through analyzing and comparing the characteristics of each metric. In addition, for the new application scenario of cloud computing, the data centers have high demand for the quality of service for killer applications, which bring new challenges to SMT in terms of performance guarantees. Therefore, we propose a new metric P-slowdown to evaluate the quality of performance guarantees. Based on experimental data, we show the feasibility of P-slowdown on performance evaluation. We also demonstrate the benefit of P-slowdown through two use cases, in which we not only improve the performance guarantee level of SMT processors through the cooperation of P-slowdown and resources allocation strategy, but also use P-slowdown to predict the occurrence of abnormal behavior against security attacks.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.2019EAL2155/_p
Copiar
@ARTICLE{e103-a_9_1127,
author={Xin JIN, Ningmei YU, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Which Metric Is Suitable for Evaluating Your Multi-Threading Processors? In Terms of Throughput, Fairness, and Predictability},
year={2020},
volume={E103-A},
number={9},
pages={1127-1132},
abstract={Simultaneous multithreading technology (SMT) can effectively improve the overall throughput and fairness through improving the resources usage efficiency of processors. Traditional works have proposed some metrics for evaluation in real systems, each of which strikes a trade-off between fairness and throughput. How to choose an appropriate metric to meet the demand is still controversial. Therefore, we put forward suggestions on how to select the appropriate metrics through analyzing and comparing the characteristics of each metric. In addition, for the new application scenario of cloud computing, the data centers have high demand for the quality of service for killer applications, which bring new challenges to SMT in terms of performance guarantees. Therefore, we propose a new metric P-slowdown to evaluate the quality of performance guarantees. Based on experimental data, we show the feasibility of P-slowdown on performance evaluation. We also demonstrate the benefit of P-slowdown through two use cases, in which we not only improve the performance guarantee level of SMT processors through the cooperation of P-slowdown and resources allocation strategy, but also use P-slowdown to predict the occurrence of abnormal behavior against security attacks.},
keywords={},
doi={10.1587/transfun.2019EAL2155},
ISSN={1745-1337},
month={September},}
Copiar
TY - JOUR
TI - Which Metric Is Suitable for Evaluating Your Multi-Threading Processors? In Terms of Throughput, Fairness, and Predictability
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 1127
EP - 1132
AU - Xin JIN
AU - Ningmei YU
PY - 2020
DO - 10.1587/transfun.2019EAL2155
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E103-A
IS - 9
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - September 2020
AB - Simultaneous multithreading technology (SMT) can effectively improve the overall throughput and fairness through improving the resources usage efficiency of processors. Traditional works have proposed some metrics for evaluation in real systems, each of which strikes a trade-off between fairness and throughput. How to choose an appropriate metric to meet the demand is still controversial. Therefore, we put forward suggestions on how to select the appropriate metrics through analyzing and comparing the characteristics of each metric. In addition, for the new application scenario of cloud computing, the data centers have high demand for the quality of service for killer applications, which bring new challenges to SMT in terms of performance guarantees. Therefore, we propose a new metric P-slowdown to evaluate the quality of performance guarantees. Based on experimental data, we show the feasibility of P-slowdown on performance evaluation. We also demonstrate the benefit of P-slowdown through two use cases, in which we not only improve the performance guarantee level of SMT processors through the cooperation of P-slowdown and resources allocation strategy, but also use P-slowdown to predict the occurrence of abnormal behavior against security attacks.
ER -