检索结果-内蒙古大学图书馆

Speculative pre-execution assisted by compiler (SPEAR)

JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING 2006年第8期66卷 1076-1089页

作者： Ro, Won W. Gaudiot, Jean-Luc Univ Calif Irvine Henry Samueli Sch Engn Dept Elect Engn & Comp Sci Irvine CA 92697 USA Calif State Univ Northridge Dept Elect & Comp Engn Northridge CA 91330 USA

Speculative pre-execution achieves efficient data prefetching by running additional prefetching threads on spare hardware contexts. Various implementations for speculative pre-execution have been proposed, including compiler-based static approaches and hardware-based dynamic approaches. A static approach defines the p-thread at compile time and executes it as a stand-alone running thread. Therefore, it cannot efficiently take dynamic events into account and requires a higher fetch bandwidth. Conversely, a hardware approach is, by essence, able to dynamically make use of run-time information. However, it requires more complex hardware and also lacks global information on data and control flow. This paper proposes Speculative Pre-Execution Assisted by compileR (SPEAR), a pre-execution model which is a hybrid of the two approaches. It relies on a post-compiler to extract the p-thread code from program binaries and uses custom-designed hardware to execute the p-thread. (C) 2006 Elsevier Inc. All rights reserved.

关键词： data prefetching speculative pre-execution simultaneous multi-threading instruction-level parallelism

来源：评论

学校读者我要写书评

暂无评论

APS: adaptable prefetching scheme to different running environments for concurrent read streams in distributed file systems

引用

JOURNAL OF SUPERCOMPUTING 2018年第6期74卷 2870-2902页

作者： Lee, Sangmin Hyun, Soon J. Kim, Hong-Yeon Kim, Young-Kyun Korea Adv Inst Sci & Technol Dept Sch Comp Daejeon South Korea ETRI High Performance Comp Res Grp Daejeon South Korea

Distributed file systems (DFSs) are widely used in various areas. One of the key issues is to provide high performance of concurrent read streams (i.e., multiple series of sequential reads by concurrent processes) for their applications. Despite the many studies on local file systems (LFSs), research has seldom been done on concurrent read streams in DFSs with different running environments (i.e., different types of storage devices and various network delays). Furthermore, most of the existing DFSs have a sharply degraded performance compared with a LFS (i.e., EXT4). Therefore, to achieve high performance in concurrent read streams, this study introduces a populating effect that keeps sending subsequent reads to a storage server and then proposes an adaptable prefetching scheme (APS) to obtain the effect even in different running environments. Hence, our APS resolves all the problems that we identified as dramatically degrading the performance in existing DFSs. In three different types of storage devices and in various network delays, the evaluation results show that our prefetching scheme (1) achieves almost the same performance as a LFS from an individual server and (2) minimizes the performance degradation of random reads.

关键词： Distributed file system Concurrent read streams data prefetching Device type Network delay

来源：评论

学校读者我要写书评

暂无评论

A performance study on bounteous transfer in multiprocessor sectored caches

引用

JOURNAL OF SUPERCOMPUTING 1997年第4期11卷 405-420页

作者： Liu, KC King, CT Natl Tsing Hua Univ Dept Comp Sci Hsinchu 30043 Taiwan

In a sectored cache, a cache line is divided into several subblocks. Each subbIock is a basic coherence unit. In this way partial block invalidation can be done on the cache lines in order to eliminate false sharing on invalidate-based multiprocessors. Sectored caches often include a facility, called bounteous transfers, to supply extra subblocks after transferring the missed subblock on a read miss. Unfortunately, previous works on sectored caches concentrated mainly on solving the false sharing problem, while overlooked the prefetching effects of bounteous transfer. In this paper, we evaluate the performance impacts of bounteous transfer based on a MESI-based sectored cache. Three different types of bounteous transfer are evaluated: bounteous transfer with valid subblocks(BT-V), bounteous transfer with clean subblocks (BT-C). and bounteous transfer disabled (No-BT). We simulated the execution of typical benchmarks, FFT, LU, Radix, SOR, on the MESI-based sectored cache. Two metrics Lf-rate and R-rate are proposed to help observe the sharing granularities and coherence overhead. Evaluation results show that different benchmarks work better with different kinds of bounteous transfer and using bounteous transfer carelessly may result in performance degradation.

关键词： cache coherence multiprocessor sectored cache partial block invalidation data prefetching false sharing

来源：评论

学校读者我要写书评

暂无评论

Secondary cache enhancement using a novel tagged prefetching method

引用

MICROPROCESSORS AND MICROSYSTEMS 1999年第4期23卷 245-253页

作者： Ki, A Elect & Telecommun Res Inst Comp Syst Dept Comp & Software Technol Lab Taejon 305350 South Korea

data prefetching, which issues data fetch requests prior to actual use, is an effective technique to reduce the effects of memory access latency. In this paper, we propose an implementation of tagged data prefetching that needs no cache modification and fairly simple hardware. Using a program driven simulation of scientific applications in the context of shared-memory multiprocessors, it is shown that the proposed method can result in performance equal to or better than that of conventional tagged prefetching. (C) 1999 Elsevier Science B.V. All rights reserved.

关键词： data prefetching tagged prefetching memory latency multiprocessors

来源：评论

学校读者我要写书评

暂无评论

Exploring prefetching, Pre-Execution and Branch Outcome Streaming for In-Memory database Lookups

IEEE COMPUTER ARCHITECTURE LETTERS

引用

IEEE COMPUTER ARCHITECTURE LETTERS 2020年第1期19卷 5-8页

作者： Cavus, Mustafa Shatnawi, Mohammed Sendag, Resit Uht, Augustus K. Univ Rhode Isl Dept Elect Comp & Biomed Engn Kingston RI 02881 USA

Lookup operations for in-memory databases are heavily memory-bound because they often rely on pointer-chasing linked data structure traversals. They are also branch heavy with branches that are hard-to-predict due to random key lookups. In this study, we show that although cache misses are the primary bottleneck for these applications, without a method for eliminating the branch mispredictions, only a small fraction of the performance benefit is achieved through prefetching. We propose the Node Tracker (NT), a novel programmable prefetcher/pre-execution unit that is highly effective in exploiting inter key-lookup parallelism to improve single-thread performance. We extend the NT with branch outcome streaming (BOS) to reduce branch mispredictions and show that NT with BOS can achieve an extra 3x speedup. Finally, we evaluated the NT as a pre-execution unit and show that we can further improve the performance in both single- and multi-threaded execution modes.

关键词： prefetching Task analysis Hardware Parallel processing Multicore processing databases data prefetching branch prediction pre-execution hardware acceleration key-value lookups

来源：评论

学校读者我要写书评

暂无评论

Adaptive Cooperation of prefetching and Warp Scheduling on GPUs

引用

IEEE TRANSACTIONS ON COMPUTERS 2019年第4期68卷 609-616页

作者： Oh, Yunho Kim, Keunsoo Yoon, Myung Kuk Park, Jong Hyun Park, Yongjun Annavaram, Murali Ro, Won Woo Yonsei Univ Sch Elect & Elect Engn Seoul 03722 South Korea Hanyang Univ Div Comp Sci & Engn Seoul 04763 South Korea Univ Southern Calif Ming Hsieh Dept Elect Engn Los Angeles CA 90007 USA

This paper proposes a new architecture, called Adaptive prefetching and Scheduling (APRES), which improves cache efficiency of GPUs. APRES relies on the observation that GPU loads tend to have either high locality or strided access patterns across warps. APRES schedules warps so that as many cache hits are generated as possible before the generation of any cache miss. Without directly predicting future cache hits/misses for each warp, APRES creates a warp group that will execute the same static load shortly and prioritizes the grouped warps. If the first executed warp in the group hits the cache, grouped warps are likely to access the same cache lines. Unless, APRES considers the load as a strided type and generates prefetch requests for the grouped warps. In addition, APRES includes a new dynamic L1 prefetch and data cache partitioning to reduce contentions between demand-fetched and prefetched lines. In our evaluation, APRES achieves 27.8 percent performance improvement.

关键词： GPU cache warp scheduling data prefetching performance

来源：评论

学校读者我要写书评

暂无评论

Fast data Delivery for Many-Core Processors

引用

IEEE TRANSACTIONS ON COMPUTERS 2018年第10期67卷 1416-1429页

作者： Bakhshalipour, Mohammad Lotfi-Kamran, Pejman Mazloumi, Abbas Samandi, Farid Naderan-Tahan, Mahmood Modarressi, Mehdi Sarbazi-Azad, Hamid SUT Tehran Iran Inst Res Fundamental Sci IPM Sch Comp Sci Tehran Iran Univ Tehran Tehran Iran Univ Calif Riverside Dept Comp Sci Riverside CA 92521 USA Shahid Chamran Univ Ahvaz SCU Dept Comp Engn Fac Engn Ahvaz Khuzestan Iran Univ Tehran Sch Elect & Comp Engn Tehran Iran SUT Dept Comp Engn Tehran Iran

Server workloads operate on large volumes of data. As a result, processors executing these workloads encounter frequent L1-D misses. In a many-core processor, an L1-D miss causes a request packet to be sent to an LLC slice and a response packet to be sent back to the L1-D, which results in high overhead. While prior work targeted response packets, this work focuses on accelerating the request packets. Unlike aggressive OoO cores, simpler cores used in many-core processors cannot hide the latency of L1-D request packets. We observe that LLC slices that serve L1-D misses are strongly temporally correlated. Taking advantage of this observation, we design a simple and accurate predictor. Upon the occurrence of an L1-D miss, the predictor identifies the LLC slice that will serve the next L1-D miss and a circuit will be set up for the upcoming miss request to accelerate its transmission. When the upcoming miss occurs, the resulting request can use the already established circuit for transmission to the LLC slice. We show that our proposal outperforms data prefetching mechanisms in a many-core processor due to (1) higher prediction accuracy and (2) not wasting valuable off-chip bandwidth, while requiring significantly less overhead. Using full-system simulation, we show that our proposal accelerates serving data misses by 22 percent and leads to 10 percent performance improvement over the state-of-the-art network-on-chip.

关键词： Memory system network-on-chip circuit switching data prefetching

来源：评论

学校读者我要写书评

暂无评论

Research on data pre-deployment in information service flow of digital ocean cloud computing

引用

Acta Oceanologica Sinica 2014年第9期33卷 82-92页

作者： SHI Suixiang XU Lingyu DONG Han WANG Lei WU Shaochun QIAO Baiyou WANG Guoren National Marine Data and Information Service State Oceanic Administration Key Laboratory of Digital Ocean State Oceanic Administration College of Computer Engineering and Science Shanghai University College of Information Science and Engineering Northeastern University

data pre-deployment in the HDFS （Hadoop distributed file systems） is more complicated than that in traditional file systems. There are many key issues need to be addressed, such as determining the target location of the data prefetching, the amount of data to be prefetched, the balance between data prefetching services and normal data accesses. Aiming to solve these problems, we employ the characteristics of digital ocean information service flows and propose a deployment scheme which combines input data prefetching with output data oriented storage strategies. The method achieves the parallelism of data preparation and data processing, thereby massively reducing I/O time cost of digital ocean cloud computing platforms when processing multi-source information synergistic tasks. The experimental results show that the scheme has a higher degree of parallelism than traditional Hadoop mechanisms, shortens the waiting time of a running service node, and significantly reduces data access conflicts.

关键词： HDFS data prefetching cloud computing service flow digital ocean

来源：评论

学校读者我要写书评

暂无评论

TPF: TCP plugged file system for efficient data delivery over TCP

引用

IEEE TRANSACTIONS ON COMPUTERS 2007年第4期56卷 459-473页

作者： Lim, Sang Seok Park, Kyu Ho Korea Adv Inst Sci & Technol Dept Elect Engn Taejon 305701 South Korea

Most Internet services, including Web, FTP, and streaming, have been realized on top of TCP, which is the de facto protocol for data delivery over the Internet. Therefore, in order to achieve high-performance data delivery over TCP, we thoroughly analyze TCP-based data delivery and identify three critical mismatches in a general file system design while supplying data to TCP. The first is the frequent sleeping/ waking up of a server process that accompanies excessive context switching overheads due to processing TCP data and ACK segments in different contexts. The second is the inefficient uniform data prefetching for TCP connections, irrespective of their characteristics such as bandwidth, latency, and the status of a send buffer. The third is inefficient disk access due to the ignorance of abrupt changes in TCP connections. As a remedy to these mismatches, we newly design a TCP-plugged file system (TPF) which is comprised of three novel mechanisms, each of which relieves the identified mismatches, integrated data sending routine, TCP aware data prefetching, and eager disk request cancellation. With these mechanisms, TPF becomes capable of supplying data managed by a file system to TCP connections timely and seamlessly and becomes reactive to abrupt changes in TCP connections. As a consequence, TPF provides minimal context switching overhead, high buffer utilization, and highly effective disk access utilization. We have implemented and tested the mechanisms in Linux 2.4. The experimental results show that the number of context switching is reduced by up to 40 percent and the overall system performance is improved by 3-34 percent.

关键词： operating systems file systems TCP data prefetching disk scheduling

来源：评论

学校读者我要写书评

暂无评论

prefetching on Storage Servers through Mining Access Patterns on Blocks

引用

IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS 2016年第9期27卷 2698-2710页

作者： Liao, Jianwei Trahay, Francois Gerofi, Balazs Ishikawa, Yutaka Southwest Univ China Coll Comp & Informat Sci Chongqing 400715 Peoples R China Nanjing Univ State Key Lab Novel Software Technol Nanjing 210023 Jiangsu Peoples R China Telecom SudParis Paris France RIKEN Adv Inst Computat Sci Kobe Hyogo Japan Univ Tokyo Tokyo 1138654 Japan

Distributed file systems have been widely deployed as back-end storage systems to offer I/O services for parallel/distributed applications that process large amounts of data. data prefetching in distributed file systems is a well-known optimization technique which can mask both network and disk latency and consequently boost I/O performance. Traditionally, data prefetching is initiated by the client file systems, however, conventional prefetching schemes are not well suited for client machines that have limited memory and computing capacity. To offer an efficient prefetching approach for resource-limited client machines, this paper proposes a novel server-side prefetching mechanism. Specifically, we propose to piggyback client identification to I/O requests so that server side block access history can be put into context. On the server side, we utilize the horizontal visibility graph technique to transform perclient time series of block access sequences into a connected graph for which we employ Tarjan's algorithm to disclose cut points in the connected graph. We express these patterns with feature tuples and we propose the X-step pattern matching algorithm to find a matching access pattern (i.e., a feature tuple) for a given block access history. Experimental results indicate that our newly proposed prefetching mechanism can ease client machines and their applications from the process of data prefetching, boosting client performance accordingly, and that it yields an attractive increase in data throughput as well.

关键词： Storage servers distributed file systems data prefetching block access patterns horizontal visibility graph

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：