Optimization of Uncore Data Flow on NUMA Platform - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2014

Optimization of Uncore Data Flow on NUMA Platform

Qiuming Luo
  • Function : Author
  • PersonId : 994300
Chang Kong
  • Function : Author
  • PersonId : 994301
Mei Wang
  • Function : Author
Ye Cai
  • Function : Author
  • PersonId : 994302


Uncore part of the processor has a profound effect, especially in NUMA systems, since it is used to connect cores, last level caches (LLC), on-chip multiple memory controllers (MCs) and high-speed interconnections. In our previous study, we investigated several benchmarks’ data flow in Uncore of Intel Westmere microarchitecture and found that the data flow of Global Queue (GQ) and QuickPath Home Logical (QHL) has serious imbalance and congestion problem. This paper, we aims at the problem of entries’ low efficiency in GQ and QHL we set up an M/M/3 Queue Model for GQ and QHL’s three trackers’ data flow, and then design a Dynamic Entries Management (DEM) mechanism which could improve entries’ efficiency dramatically. The model is implemented in Matlab to simulate two different data flow pattern. Experiment results shows that DEM mechanism reduces stall cycles of trackers significantly: DEM reduces almost 60% stall cycles under smooth request sequences; DEM mechanism reduces almost 20~30% stall cycles under burst request sequences.


Fichier principal
Vignette du fichier
978-3-662-44917-2_7_Chapter.pdf (608.6 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-01403066 , version 1 (25-11-2016)





Qiuming Luo, Yuanyuan Zhou, Chang Kong, Mei Wang, Ye Cai. Optimization of Uncore Data Flow on NUMA Platform. 11th IFIP International Conference on Network and Parallel Computing (NPC), Sep 2014, Ilan, Taiwan. pp.72-83, ⟨10.1007/978-3-662-44917-2_7⟩. ⟨hal-01403066⟩
40 View
130 Download



Gmail Facebook X LinkedIn More