Innovative Architecture for Future Generation High-Performance Processors and Systems最新文献

筛选
英文 中文
ASCI application performance and the impact of commodity processor architectural trends ASCI应用程序性能和商品处理器体系结构趋势的影响
O. Lubeck, A. Hoisie, F. Bassetti, K. Cameron, Yong Luo, H. Wasserman
{"title":"ASCI application performance and the impact of commodity processor architectural trends","authors":"O. Lubeck, A. Hoisie, F. Bassetti, K. Cameron, Yong Luo, H. Wasserman","doi":"10.1109/IWIA.1998.779067","DOIUrl":"https://doi.org/10.1109/IWIA.1998.779067","url":null,"abstract":"The purpose of this paper is to summarize recent performance results from an important ASCI-related application and to speculate on how trends within the computer industry and in computer architecture relate to these results.","PeriodicalId":418008,"journal":{"name":"Innovative Architecture for Future Generation High-Performance Processors and Systems","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1999-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115615180","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 4
New methods for exploiting program structure and behavior in computer architecture 计算机体系结构中开发程序结构和行为的新方法
A. Roth, G. Sohi
{"title":"New methods for exploiting program structure and behavior in computer architecture","authors":"A. Roth, G. Sohi","doi":"10.1109/IWIA.1998.779075","DOIUrl":"https://doi.org/10.1109/IWIA.1998.779075","url":null,"abstract":"Micro-architectural techniques of the next decade will have to be more efficient and scalable in order to handle growing workloads and longer communication and memory latencies. We believe that information about program structure, the data and control relationships between instructions, can be used as a powelful framework for new techniques. We argue that program structure information has several inherent advantages over frameworks that associate information either with instructions in isolation or with data. We present summaries of four novel methods that apply program structure information to memory system problems from disambiguation and data cache bandwdith to. prefetching and coherence optimization.","PeriodicalId":418008,"journal":{"name":"Innovative Architecture for Future Generation High-Performance Processors and Systems","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126610991","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 1
First step to combining control and data speculation 将控制和数据推测结合起来的第一步
T. Sato
{"title":"First step to combining control and data speculation","authors":"T. Sato","doi":"10.1109/IWIA.1998.779073","DOIUrl":"https://doi.org/10.1109/IWIA.1998.779073","url":null,"abstract":"Recently there are many studies of data value prediction for increasing instruction level parallelism, and it is found that data speculation affects branch prediction accuracy. Even when data dependences are speculated successfully, processor performance would be degraded if branch prediction accuracy were decreased. On the other hand, branch prediction studies are nearly matured. While it becomes very difficult to increase the accuracy, there is still a vast gap between actual and ideal processor performance. From these considerations, we investigate to combine control and data speculation. Here, we evaluate the correlation between control speculation and data speculation and then propose to predict branch outcomes using data value prediction. Keywords: instruction level parallelism, speculative execution, branch prediction, value prediction, speculative verification.","PeriodicalId":418008,"journal":{"name":"Innovative Architecture for Future Generation High-Performance Processors and Systems","volume":"46 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"117331313","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 4
Development of biological and chemical applications on a 64-node PC cluster 在64节点PC集群上开发生物和化学应用
Y. Akiyama, K. Onizuka, T. Noguchi, M. Ando
{"title":"Development of biological and chemical applications on a 64-node PC cluster","authors":"Y. Akiyama, K. Onizuka, T. Noguchi, M. Ando","doi":"10.1109/IWIA.1998.779070","DOIUrl":"https://doi.org/10.1109/IWIA.1998.779070","url":null,"abstract":"We describe the development of two parallel applications running on our PC cluster which consists of 64 Pentium Pro 200MHz microprocessors. One is an integrated parallel calculation system, called the PAPIA system, dedicated to protein information analysis. The PAPIA system enables very fast protein database searches (3-0 structure matching and sequence homology search) by fully utilizing the power of the 64 local hard-disks, as well as fast parallel calculations supported by the high-bandwidth and low-latency communication driver developed by RWCR We have starteda WWWsewice (http://www. rwcp.oKjp/papid) so any biologist can easily submit jobs to the PAPIA system through a WWW browse,: The service has been accessed from 55 countries in the world. The other application is parallel molecular dynamics simulation. We have eficiently ported the AMBER program to the PC cluster; and then appended an optional functionality of the Barnes-Hut tree code for quick and accurate calculation of Coulomb potentials without a cut-off approximation (PPPC method). The RWC PC Cluster shows good scalability in these two real parallel applications.","PeriodicalId":418008,"journal":{"name":"Innovative Architecture for Future Generation High-Performance Processors and Systems","volume":"190 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124215256","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Increasing the lookahead of multilevel branch prediction 增加多级分支预测的前瞻性
A. Veidenbaum
{"title":"Increasing the lookahead of multilevel branch prediction","authors":"A. Veidenbaum","doi":"10.1109/IWIA.1998.779074","DOIUrl":"https://doi.org/10.1109/IWIA.1998.779074","url":null,"abstract":"Many techniques have been proposed for tolemting memory latency in future systems, including prefetching and Decoupled-Access DRAM (DA-DRAM) architectures. In order for these techniques to be effective they need to have a suficient lookahead, i.e. to be far enough ahead of processor execution in requesting data. Bmnch prediction has been utilized before to achieve this but only small degrees of lookahead have been studied. This is not enough for latencies of tens to hundreds of clock cycles. In particular, DA-DRAM systems need a higher degree of decoupling because the access processor is placed in memory. This paper investigates the potential of multilevel prediction for increased lookahead, beyond the 2-3 level predictors that have been reported. A new multilevel predictor is defined and up to 8 levels of lookahead are studied. We find that the predictor has a high degree of accuracy, over 80%, even for integer benchmarks such as gcc and Oracle. Branch outcome as well as the address of the next instruction after the branch are predicted. A comparison of singleand multiple lookup prediction is presented. In general, accuracy decreases with lookahead. However, given sufficient predictor size it can be made quite high. With the DA-DRAM predictor located in DRAM, a purely software predictor with storage in the DRAM can be implemented. A similar predictor can be used for instruction and L2 cache prefetching or trace scheduling.","PeriodicalId":418008,"journal":{"name":"Innovative Architecture for Future Generation High-Performance Processors and Systems","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131465166","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
A speculative multithreading with selective multi-path execution 具有选择性多路径执行的推测多线程
O. Kanemitsu, Y. Wataru, F. Fumihito
{"title":"A speculative multithreading with selective multi-path execution","authors":"O. Kanemitsu, Y. Wataru, F. Fumihito","doi":"10.1109/IWIA.1998.779072","DOIUrl":"https://doi.org/10.1109/IWIA.1998.779072","url":null,"abstract":"Recent microprocessors' performance has been improved by their high-speed clock frequency and by their exploiting instruction-level parallelism (ILP). Physical limitations of clock speed and semantical limitations of control dependencies impede the improvement of performance. To overcome this dificulty, it is indispensable to make use of the thread-level parallelism. This paper proposes a speculative thread execution model that aims at a speed-up of sequential program execution by selective multi-path thread execution.","PeriodicalId":418008,"journal":{"name":"Innovative Architecture for Future Generation High-Performance Processors and Systems","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127843696","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 1
Cc-coma: the compiler-controled coma as a framework for parallel computing 作为并行计算框架的编译器控制的昏迷
S. Saito, T. Uehara, K. Joe, Y. Kunieda
{"title":"Cc-coma: the compiler-controled coma as a framework for parallel computing","authors":"S. Saito, T. Uehara, K. Joe, Y. Kunieda","doi":"10.1109/IWIA.1998.779082","DOIUrl":"https://doi.org/10.1109/IWIA.1998.779082","url":null,"abstract":"In order to provide a fully-autosmutic parullelzing compiler with the furiction of both automatic datu partitioning and distribution, we propose a compilercontrolled cache only memory architecture (cc-COMA). The cc-COMA runtime system as based on a softwure-emulated COMA which covers a variety of parallel architectures from NUMA to NOW. The compiler generates both the user code and control code for the emulated cache based on the Entry Consistency model. The runtime system of cc-COMA executes the generated code following the suggestions of the compiler. In this paper we present an outline of what we call the cc-COMA system. Also described is an implementation of the cc-COMA, runtime system for networks of workstations, named Laurasia.","PeriodicalId":418008,"journal":{"name":"Innovative Architecture for Future Generation High-Performance Processors and Systems","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125227272","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 1
The effects of predicated execution on architectures supporting dynamic speculation 预测执行对支持动态推测的体系结构的影响
S. Mantripragada, A. Nicolau
{"title":"The effects of predicated execution on architectures supporting dynamic speculation","authors":"S. Mantripragada, A. Nicolau","doi":"10.1109/IWIA.1998.779071","DOIUrl":"https://doi.org/10.1109/IWIA.1998.779071","url":null,"abstract":"Branch instructions pose a serious problem in achieving a good instruction level pamllelism (ILP) from a program. Modern microprocessors have attempted to alleviate this problem with the support of sophisticated branch prediction schemes. Dynamic speculation, as a hardware feature, is used to execute instructions out-of-order (OOO) guided by the outcomes of such prediction schemes. Previous branch outcomes are used to speculate an instruction before the branch is fully resolved. Studies have shown that inspite of such complex prediction schemes, there still exist many frequently executed branches which are hard to predict. Predicated execution has been proposed as an alternative technique by researchers to eliminate some of these hard to predict bmnches. Previous studies have analyzed the effects of applying predicated execution with branch prediction. Here, we study the effects of predicated execution on architectures supporting dynamic speculation. Branch classification, as a technique, is used to group branches with similar dynamic execution frequencies. The effects of different predication models on dynamic speculation are then studied for each of these branch classes. An in-depth analysis of useful and useless instructions as a result of applying these different models is also summarized.","PeriodicalId":418008,"journal":{"name":"Innovative Architecture for Future Generation High-Performance Processors and Systems","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130171248","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
Achieving high performance via co-designed virtual machines 通过共同设计的虚拟机实现高性能
J.E. Smith, S. Sastry, T. Heil, T. Bezenek
{"title":"Achieving high performance via co-designed virtual machines","authors":"J.E. Smith, S. Sastry, T. Heil, T. Bezenek","doi":"10.1109/IWIA.1998.779076","DOIUrl":"https://doi.org/10.1109/IWIA.1998.779076","url":null,"abstract":"A virtual machine (VM) uses software to support a virtual instruction set architecture on a hardware platjonn executing a native instruction set. By co-designing the hardware and software elements of a VM, and by using an implementation-dependent native instruction set, there will be many new opportunities for improved performance and flexibility. Because the hardware-supported instruction set is implementation dependent, performance optimizations can be more easily passed from software through to hardware, and performance feedback infonnation can be more easily passed from hardware up to the software. Furthermore, optimizations can be performed by software dynamically, as the program runs. A codesigned virtual machine may include adaptive hardware Performance features, continuous hardware performance feedback, and on-the-fly optimizing re-compilation by the VM. Hardware and software can cooperate in jinding instruction level parallelism across large blocks of dynamic instructions, and can eflciently implement of a number of advanced microarchitecture techniques involving control independence, prediction, speculation, and cache hierarchy management.","PeriodicalId":418008,"journal":{"name":"Innovative Architecture for Future Generation High-Performance Processors and Systems","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123791342","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 37
System support for dynamic optimization of application performance 系统支持动态优化应用程序性能
Shih-Hao Hung, E. Davidson
{"title":"System support for dynamic optimization of application performance","authors":"Shih-Hao Hung, E. Davidson","doi":"10.1109/IWIA.1998.779068","DOIUrl":"https://doi.org/10.1109/IWIA.1998.779068","url":null,"abstract":"Today's high performance and parallel computer systems provide substantial opportunities for concurrency of execution and scalability that is largely untapped by the applications that run on them. Under traditional frameworks, developing efficient applications can be a labor-intensive process that requires an intimate knowledge of the machines, the applications, and many subtle machine-application interactions. Optimizing applications so that they can achieve their full potential on target machines is often beyond the programmer's or the compiler's ability or endurance. This paper argues for addressing the performance optimization problem by providing support for tuning application codes with substantially reduced human intervention by exploiting dynamic information gathered at run time so that the optimization is responsive to actual run time behavior as data sets change and installed systems evolve.","PeriodicalId":418008,"journal":{"name":"Innovative Architecture for Future Generation High-Performance Processors and Systems","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1900-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130306645","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 1
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信
小红书