Network latency equalization-oriented switch allocation strategy for networks-on-chip

被引：0
|
作者：
Wang Z. [1 ]
Chen X. [1 ]
Guo Y. [1 ]
机构：
[1] College of Computer, National University of Defense Technology, Changsha
来源：
Chen, Xiaowen (xwchen@nudt.edu.cn) | 2018年 / National University of Defense Technology卷 / 40期
关键词：
Equalization; Networks-on-chip; Switch allocation;
D O I：
10.11887/j.cn.201806012
中图分类号：
学科分类号：
摘要：
According to the problem of switch allocation, a network latency equalization-oriented switch allocation strategy was proposed to improve system performance by designing a more fairness switch allocation strategy and improving the degree of network latency equalization. The evaluation of the design with SPEC CPU2006 benchmarks was performed in a full-system simulator. Compared with the canonical separable switch allocator and the TS-Router (recently proposed switch allocator), the experiments show that the approach decreases LSD (latency standard deviation) by 13.8% and 3.9% respectively, as well as ML (maximum latency) by 45.6% and 15.1% respectively. The approach improves system throughput by 0.8% over that of TS-Router. Finally, the design is implemented based on the separable switch allocator, and it can perform the evaluation in speed, area and power. © 2018, NUDT Press. All right reserved.
引用
收藏
页码：82 / 88
页数：6
相关论文
共 12 条
  • [1] Jerger N.D.E., Peh L.S., On-chip Networks, Synthesis Lectures on Computer Architecture, (2009)
  • [2] Dally W., Towles B., Principles and Practices of Interconnection Networks, (2003)
  • [3] Hsieh W.C., Wang P., Weihl W.E., Computation migration: enhancing locality for distributed-memory parallel systems, Proceedings of the Fourth ACM SIGPLAN Symposium on PRINCIPLES and Practice of Parallel Programming, pp. 239-248, (1993)
  • [4] Michaud P., Exploiting the cache capacity of a single-chip multi-core processor with execution migration, Proceedings of 10th International Symposium on High Performance Computer Architecture, (2004)
  • [5] Chakraborty K., Wells P.M., Sohi G.S., Computation spreading: employing hardware migration to specialize CMP cores on-the-fly, ACM SIGOPS Operating Systems Review, 40, 5, pp. 283-292, (2006)
  • [6] Mullins R., West A., Moore S., Low-latency virtual-channel routers for on-chip networks, ACM SIGARCH Computer Architecture News, 32, 2, (2004)
  • [7] Park D., Das R., Nicopoulos C., Et al., Design of a dynamic priority-based fast path architecture for on-chip interconnects, Proceedings of 15th Annual IEEE Symposium on High-Performance Interconnects, pp. 15-20, (2007)
  • [8] Mckeown N., The iSLIP scheduling algorithm for input-queued switches, IEEE/ACM Transactions on Networking, 7, 2, pp. 188-201, (2002)
  • [9] Chang Y.Y., Huang S.C., Poremba M., Et al., TS-Router: on maximizing the quality-of-allocation in the on-chip network, Proceedings of 19th International Symposium on High Performance Computer Architecture, pp. 390-399, (2013)
  • [10] Agarwal N., Krishna T., Peh L.S., Et al., GARNET: a detailed on-chip network model inside a full-system simulator, Proceedings of International Symposium on Performance Analysis of Systems and Software, pp. 33-42, (2009)
12