本文選題：ccNUMA系統(tǒng) + 稀疏目錄��； 參考：《計算機研究與發(fā)展》2014年09期

【摘要】：稀疏目錄技術(shù)在緩存一致性非一致存儲訪問(cache coherent non-uniform memory access,ccNUMA)系統(tǒng)中有廣泛應(yīng)用.但是,稀疏目錄技術(shù)的一個主要缺陷在于目錄項替換在目錄熱點存在的情況下會嚴(yán)重降低系統(tǒng)的性能.針對此問題,研究如何提升稀疏目錄性能.首先,從定性和定量的角度對主流的稀疏目錄替換算法進(jìn)行了分析,并且發(fā)現(xiàn)最為原始的最久未使用(least-recent-used,LRU)算法的性能實際上要優(yōu)于一些近期提出的目錄替換算法,如最少共享者(least-sharer-count,LSC)算法.其次,將victim cache的思想應(yīng)用到稀疏目錄上,提出了victim目錄(victim directory),該技術(shù)在主稀疏目錄模塊上加入了一個小的全相聯(lián)二級目錄存儲.最后,提出了選擇性victim目錄(selective victim directory)技術(shù),使得victim目錄選擇性地僅存儲有用的目錄項,從而進(jìn)一步減少了目錄替換的數(shù)量.選擇性victim目錄通過向所有節(jié)點的cache發(fā)出探查消息(probe messages),從而選擇性地存儲較為有用的目錄項.實驗證明,選擇性victim目錄取得了比簡單victim目錄更好的性能;而且在僅增加了1KB左右的硬件開銷的代價下,選擇性victim目錄節(jié)省了35.7%的程序運行時間.
[Abstract]:Sparse directory technology is widely used in cache consistent storage access coherent non-uniform memory access nu MAs. However, one of the main defects of sparse directory technology is that the substitution of directory items in the presence of directory hotspots will seriously reduce the performance of the system. To solve this problem, we study how to improve the performance of sparse directories. First of all, the main sparse directory replacement algorithms are analyzed qualitatively and quantitatively, and it is found that the performance of the most original algorithm, which has not been used for a long time, is actually better than that of some recently proposed directory replacement algorithms. For example, least-share-count LSC algorithm. Secondly, the idea of victim cache is applied to the sparse directory, and the victim directory is proposed, which adds a small fully associative secondary directory storage to the main sparse directory module. Finally, the selective victim directory selective victim directorytechnique is proposed, which makes the victim directory selectively store only useful directory entries, thus further reducing the number of directory replacements. The selective victim directory selectively stores useful directory entries by sending a probe message to the cache of all nodes. Experiments show that the selective victim directory achieves better performance than the simple victim directory, and that the selective victim directory saves 35.7% of the program running time at the cost of only increasing the hardware overhead about 1KB.
【作者單位】： 中國科學(xué)院計算技術(shù)研究所高性能計算機研究中心;中國科學(xué)院大學(xué);
【基金】：國家“九七三”重點基礎(chǔ)研究發(fā)展計劃基金項目(2011CB302501) 國家自然科學(xué)基金項目(61173007,61100013,61100015,61202059)
【分類號】：TP333

【相似文獻(xiàn)】

相關(guān)碩士學(xué)位論文 前2條

1 陳超;智能網(wǎng)絡(luò)存儲系統(tǒng)（INSS）客戶端技術(shù)研究[D];華南理工大學(xué);2011年

2 李志;面向閃存的緩沖區(qū)管理算法研究[D];中國科學(xué)技術(shù)大學(xué);2010年

，

本文編號：1912173