三維記憶體之動態考量散熱與存取模式的配置方法

Thermal-aware Dynamic Page Allocation Policy Considering Future Access Patterns for Hybrid Memory Cube (HMC)

梁凱澤(Liang, Kai Tse)

指導教授 ： 黃婷婷

國立清華大學/電機資訊學院/資訊工程學系/碩士(2015年)

https://doi.org/10.6843/NTHU.2015.00383

摘要

在多核心處理器的運作下，對於記憶體的頻寬需求量也隨之提高。所以透過矽穿孔堆疊動態隨機存取的記憶體能夠滿足高頻寬的需求。然而，這堆疊的結果會導致功率密度提高，連帶產生高溫散熱的問題，舉凡資料的可靠度、記憶體效能以及額外必須負擔的冷卻成本。除了散熱的問題，在多核心的環境下，程序間的記憶體干擾亦會有損系統效能。為了達到最佳的效能，我們提出動態考量散熱與存取模式的配置方法，方法中考慮到資料的使用頻率、各層的位置分佈、散熱影響、頻寬變化量、記憶體干擾。我們利用數學分析模型估算系統效能，並考量到上述的所有影響因素。實驗結果顯示，我們提出的記憶體配置方法效能上比起傳統配置方法快12.7%。此外，我們的數學分析模型平均錯誤比率只有0.86%。

關鍵字

三維記憶體 ； 散熱 ； 存取模式

並列摘要

Three-dimensional (3-D) memory stacking like Hybrid Memory Cube (HMC) can resolve
memory bandwidth challenges for multi-core system where stacked multiple DRAM dies are
connected by Through Silicon Vias (TSVs). However, high power density due to the high in-
tegration incurs temperature related problems in reliability, performance, and system cooling
cost. In addition to thermal issues, in multi-core system, memory interference between pro-
cesses may degrade system performance. In order to achieve better performance, we propose
a dynamic page allocation policy considering access frequency of pages, physical locations of
DRAM dies, thermal impacts, bandwidth variation of each process, and memory interference
among processes. We also propose an analytical model to estimate the system performance
considering the above factors. Experimental results show that our proposed memory mapping
policy can outperform MCP [1] 12.7% on average. The average error rate of our analytical
model is only 0.86%.

並列關鍵字

HMC ； Thermal ； Access patterns

參考文獻

partitioning. In MICRO'11, pages 374-385, 2011.

national Conference on Networking, Architecture and Storage (NAS), pages 268-274,

on Computer Architecture (ISCA'07), pages 312-322, 2007.

60-71, 2009.

on High-Performance Computer Architecture (HPCA-17), pages 515-525, 2011.