Two-Level Hybrid Sampled Simulation of Multithreaded Applications

	Two-Level Hybrid Sampled Simulation of Multithreaded Applications
	Chuntao Jiang; Zhibin Yu; Lieven Eechhout; Hai Jin, Xiaofei Liao; Cheng-Zhong Xu
刊名	ACM Transactions on Architecture and Code Optimization
	2015
英文摘要	Sampled micro-architectural simulation of single-threaded applications is mature technology for over a decade now. Sampling multi-threaded applications on the other hand, is much more complicated. Not until very recently have researchers proposed solutions for sampled simulation of multi-threaded applications.Time-Based Sampling (TBS) samples multi-threaded application execution based on time — not instructions as is typically done for single-threaded applications — yielding estimates for a multi-threaded application’s execution time. In this paper, we revisit and analyze previously proposed TBS approaches (periodic and cantor fractal based sampling), and we obtain a number of novel and surprising insights, such as (i)accurately estimating fast-forwarding IPC, i.e., performance in-between sampling units, is more important than accurately estimating sample IPC, i.e., performance within the sampling units; (ii) fast-forwarding IPC estimation accuracy is determined by both the sampling unit distribution and how to use the sampling units to predict fast-forwarding IPC; and (iii) cantor sampling is more accurate at small sampling unit sizes,whereas periodic is more accurate at large sampling unit sizes. These insights lead to the development of Two-level Hybrid Sampling (THS), a novel sampling methodology for multi-threaded applications that combines periodic sampling’s accuracy at large time scales (i.e., uniformly selecting coarse-grain sampling units across the entire program execution) with cantor sampling’s accuracy at small time scales (i.e., ability to accurately predict fast-forwarding IPC in-between small sampling units). The clustered occurrence of small sampling units under cantor sampling also enables shortened warmup and thus enhanced simulation speed. Overall, THS achieves an average absolute execution time prediction error of 4% while yielding an average simulation speedup of 40 compared to detailed simulation,which is both more accurate and faster than the current state-of-the-art. Case studies illustrate THS’ability to accurately predict relative performance differences across the design space.
收录类别	SCI
原文出处	http://dl.acm.org/citation.cfm?id=2818353
语种	英语
内容类型	期刊论文
源URL	[http://ir.siat.ac.cn:8080/handle/172644/6883]
专题	深圳先进技术研究院_数字所
作者单位	ACM Transactions on Architecture and Code Optimization
推荐引用方式 GB/T 7714	Chuntao Jiang,Zhibin Yu,Lieven Eechhout,et al. Two-Level Hybrid Sampled Simulation of Multithreaded Applications[J]. ACM Transactions on Architecture and Code Optimization,2015.
APA	Chuntao Jiang,Zhibin Yu,Lieven Eechhout,Hai Jin, Xiaofei Liao,&Cheng-Zhong Xu.(2015).Two-Level Hybrid Sampled Simulation of Multithreaded Applications.ACM Transactions on Architecture and Code Optimization.
MLA	Chuntao Jiang,et al."Two-Level Hybrid Sampled Simulation of Multithreaded Applications".ACM Transactions on Architecture and Code Optimization (2015).