计算机研究与发展
計算機研究與髮展
계산궤연구여발전
JOURNAL OF COMPUTER RESEARCH AND DEVELOPMENT
2009年
11期
1949-1955
,共7页
李胜梅%程步奇%高兴誉%乔林%汤志忠
李勝梅%程步奇%高興譽%喬林%湯誌忠
리성매%정보기%고흥예%교림%탕지충
性能分析%cache缺失%主成分分析%线性回归%SPEC CPU2006
性能分析%cache缺失%主成分分析%線性迴歸%SPEC CPU2006
성능분석%cache결실%주성분분석%선성회귀%SPEC CPU2006
performance analysis%cache miss%principal component analysis%linear regression analysis%SPEC CPU2006
应用程序的性能分析能够给体系架构设计者和性能优化者提供有效的参考和指导.采用主成分线性回归模型分析了SPEC CPU2006的整型程序性能.模型选取性能监测单元采样到的事件为自变量,每条指令的时钟周期数(CPI)作为因变量.模型中采用主成分分析法消除了性能事件之间的相关性.实验结果表明,模型的拟合优度在90%以上,对性能进行预测的平均相对误差为15%.模型从量化上分析了L1,L2高速缓存缺失作为影响性能的关键因素是怎样影响程序性能的.
應用程序的性能分析能夠給體繫架構設計者和性能優化者提供有效的參攷和指導.採用主成分線性迴歸模型分析瞭SPEC CPU2006的整型程序性能.模型選取性能鑑測單元採樣到的事件為自變量,每條指令的時鐘週期數(CPI)作為因變量.模型中採用主成分分析法消除瞭性能事件之間的相關性.實驗結果錶明,模型的擬閤優度在90%以上,對性能進行預測的平均相對誤差為15%.模型從量化上分析瞭L1,L2高速緩存缺失作為影響性能的關鍵因素是怎樣影響程序性能的.
응용정서적성능분석능구급체계가구설계자화성능우화자제공유효적삼고화지도.채용주성분선성회귀모형분석료SPEC CPU2006적정형정서성능.모형선취성능감측단원채양도적사건위자변량,매조지령적시종주기수(CPI)작위인변량.모형중채용주성분분석법소제료성능사건지간적상관성.실험결과표명,모형적의합우도재90%이상,대성능진행예측적평균상대오차위15%.모형종양화상분석료L1,L2고속완존결실작위영향성능적관건인소시즘양영향정서성능적.
The factors influencing application performance are various and the extents of influence are different. Analyzing and distinguishing the extents of influence caused by various factors can guide the architects in the architecture design and help programmers in the optimization. However, it is not easy to distinguish the extents of influence because the factors may correlate each other themselves. In this paper, a principal component linear regression model aiming at performance of SPEC CPU2006 integer benchmarks is set up. Cycles per instruction (CPI) is used to represent the application performance and the performance events monitored by performance monitor unit (PMU) are used to represent the influencing factors. Principal component analysis is implemented to eliminate the linear correlation among performance events. Then linear regression model is set up which uses CPI as the dependent variable and principal components as the independent variables. This model can analyze the influence on CPI caused by the performance events i. e. L1 data cache miss, L2 cache miss, DTLB miss, branch mis-prediction, micro-fusion, memory disambiguation events quantitatively. The model is validated by the t test and F test with goodness of fit over 90%. The average relative prediction error of the model is 15%. The results show quantitatively how L1 and L2 cache misses dominate the performance of the applications.