Improving co-running program's performance on CMP

被引:0
|
作者
Tian, Dawei [1 ]
Li, Hongkui [2 ]
Niu, Wenhui [2 ]
Deng, Ying [2 ]
Li, Fujian [2 ]
机构
[1] State Grid Shandong Elect Power Co, Jinan 250001, Shandong, Peoples R China
[2] State Grid Shandong Heze Elect Power Co, Heze 274000, Shandong, Peoples R China
关键词
D O I
10.1051/matecconf/201712804011
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
Chip multi-processor (CMP) has become the most common processor in the current cluster and desktop computer, and it is also the current development direction. On CMP, programs usually co-running with each other. However, programs commonly interfere with each other. Some time the interference takes big effect, which cause serious drop down of performance. In order to avoid the serious performance interference, programs should be scheduled reasonably to different socket to improve the program's performance and system's utilization. In this paper we propose a new scheduling method to realize a more reasonable scheduling. We do not only consider the LLC miss rate, but also consider the LLC reference of the programs. By the information of LLC reference and LLC miss rate, we schedule programs to different sockets, which realize a reasonable scheduling. The experiment result show that making use of the scheduling method proposed by our paper, program's performance can improve 4%, because the performance improve is realized by on-chip resource, which is a big contribution.
引用
收藏
页数:4
相关论文
共 50 条
  • [1] Guaranteeing Co-running Program'S Performance in Data Center
    Guan, Ti
    Liu, Yong
    Wu, Guanbin
    You, Daning
    Zhang, Xuekai
    2017 INTERNATIONAL CONFERENCE ON CLOUD TECHNOLOGY AND COMMUNICATION ENGINEERING (CTCE2017), 2017, 910
  • [2] Contention-aware prediction for performance impact of task co-running in multicore computers
    Ren, Shenyuan
    He, Ligang
    Li, Junyu
    Chen, Zhiyan
    Jiang, Peng
    Li, Chang-Tsun
    WIRELESS NETWORKS, 2022, 28 (03) : 1293 - 1300
  • [3] Contention-aware prediction for performance impact of task co-running in multicore computers
    Shenyuan Ren
    Ligang He
    Junyu Li
    Zhiyan Chen
    Peng Jiang
    Chang-Tsun Li
    Wireless Networks, 2022, 28 : 1293 - 1300
  • [4] Ensuring the Fairness of program's performance on CMP
    Wang, Qilong
    Gao, Jun
    Hou, Guangsong
    Li, Hongkui
    Xu, Ke
    2017 INTERNATIONAL CONFERENCE ON ELECTRONIC INFORMATION TECHNOLOGY AND COMPUTER ENGINEERING (EITCE 2017), 2017, 128
  • [5] Understanding Co-Running Behaviors on Integrated CPU/GPU Architectures
    Zhang, Feng
    Zhai, Jidong
    He, Bingsheng
    Zhang, Shuhao
    Chen, Wenguang
    IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 2017, 28 (03) : 905 - 918
  • [6] Resource Isolation Method for Program'S Performance on CMP
    Guan, Ti
    Liu, Chunxiu
    Xu, Zheng
    Li, Huicong
    Ma, Qiang
    2017 INTERNATIONAL CONFERENCE ON CLOUD TECHNOLOGY AND COMMUNICATION ENGINEERING (CTCE2017), 2017, 910
  • [7] Learning based Memory Interference Prediction for Co-running Applications on Multi-Cores
    Saeed, Ahsan
    Mueller-Gritschneder, Daniel
    Rehm, Falk
    Hamann, Arne
    Ziegenbein, Dirk
    Schlichtmann, Ulf
    Gerstlauer, Andreas
    2021 ACM/IEEE 3RD WORKSHOP ON MACHINE LEARNING FOR CAD (MLCAD), 2021,
  • [8] Impact of Data Sharing on Co-Running Embedded Applications in Multi-Core System
    Korotaeva, Anna
    Nebel, Wolfgang
    23RD EUROMICRO INTERNATIONAL CONFERENCE ON PARALLEL, DISTRIBUTED, AND NETWORK-BASED PROCESSING (PDP 2015), 2015, : 716 - 720
  • [9] Dynamic Budgeting for Settling DRAM Contention of Co-running Hard and Soft Real-time Tasks
    Flodin, Jonas
    Lampka, Kai
    Yi, Wang
    2014 9TH IEEE INTERNATIONAL SYMPOSIUM ON INDUSTRIAL EMBEDDED SYSTEMS (SIES), 2014,
  • [10] A SPATIAL RESONANCE UNDER 4-PHOTON INTERACTION BETWEEN CO-RUNNING WAVES IN A CUBIC MEDIUM
    BOLSHOV, LA
    VLASOV, DV
    GARAEV, RA
    KVANTOVAYA ELEKTRONIKA, 1982, 9 (01): : 83 - 91