检索结果-内蒙古大学图书馆

4th acm sigplan symposium on principles and practice of parallel programming, PPOPP 1993

作者： Chatterjee, Siddhartha Gilbert, John R. Long, Fred J.E. Schreiber, Robert Teng, Shang-Hua Research Institute for Advanced Computer Science Mail Stop T045-1 NASA Ames Research Center Moffett FieldCA94035-1000 United States Xerox Palo Alto Research Center 3333 Coyote Hill Road Palo AltoCA943041314 United States Department of Computer and Information Science University of California Santa CruzCA95064 United States Department of Mathematics Massachusetts Institute of Technology CambridgeMA02139 United States

ISBN: (纸本)0897915895

Generating local addresses and communication sets is an important issue in distributed-memory implementations of data-parallel languages such as High Performance Fortran. We show that for an array A affinely aligned to a template that is distributed across p processors with a cyclic(κ) distribution, and a computation involving the regular section A(: h: s), the local memory access sequence for any processor is characterized by a finite state machine of at most κ;states. We present fast algorithms for computing the essential information about these state machines, and extend the framework to handle multidimensional arrays. We also show how to generate communication sets using the state machine approach. Performance results show that this solution requires very little runtime overhead and acceptable preprocessing time.

关键词： Memory architecture

来源：评论

学校读者我要写书评

暂无评论

Preliminary Results on NB-FEB, a Synchronization Primitive for parallel programming

引用

acm sigplan NOTICES 2009年第4期44卷 295-296页

作者： Ha, Phuong Hoai Tsigas, Philippas Anshus, Otto J. Univ Tromso N-9001 Tromso Norway Chalmers Univ Technol Gothenburg Sweden

We introduce a non-blocking full/empty bit primitive, or NB-FEB for short, as a promising synchronization primitive for parallel programming on may-core architectures. We show that the NB-FEB primitive is universal, scalable and feasible. NB-FEB, together with registers, can solve the consensus problem for an arbitrary number of processes (universality). NB-FEB is combinable, namely its memory requests to the same memory location can be combined into only one memory request, which consequently mitigates performance degradation due to synchronization "hot spots" (scalability). Since NB-FEB is a variant of the original full/empty bit that always returns a value instead of waiting for a conditional flag, it is as feasible as the original full/empty bit, which has been implemented in many computer systems (feasibility).

关键词： Algorithms Reliability theory many-core architectures non-blocking synchronization full/empty bit universal primitives combinability

来源：评论

学校读者我要写书评

暂无评论

Heuristics Entwined with Handlers Combined From Functional Specification to Logic programming Implementation 14

Heuristics Entwined with Handlers Combined From Functional S...

引用

16th International symposium on principles and practice of Declarative programming (PPDP)

作者： Schrijvers, Tom Wu, Nicolas Desouter, Benoit Demoen, Bart Univ Ghent Ghent Belgium Univ Oxford Oxford England Katholieke Univ Leuven Leuven Belgium

ISBN: (纸本)9781450329477

A long-standing problem in logic programming is how to cleanly separate logic and control. While solutions exist, they fall short in one of two ways: some are too intrusive, because they require significant changes to Prolog's underlying implementation;others are lacking a clean semantic grounding. We resolve both of these issues in this paper. We derive a solution that is both lightweight and principled. We do so by starting from a functional specification of Prolog based on monads, and extend this with the effect handlers approach to capture the dynamic search tree as syntax. Effect handlers then express heuristics in terms of tree transformations. Moreover, we can declaratively express many heuristics as trees themselves that are combined with search problems using a generic entwining handler. Our solution is not restricted to a functional model: we show how to implement this technique as a library in Prolog by means of delimited continuations.

关键词： monads effect handlers free monad transformer delimited continuations logic programming heuristic tree search

来源：评论

学校读者我要写书评

暂无评论

Training one DeePMD Model in Minutes: a Step towards Online Learning 24

Training one DeePMD Model in Minutes: a Step towards Online ...

引用

29th acm sigplan Annual symposium on principles and practice of parallel programming (PPoPP)

作者： Hu, Siyu Zhao, Tong Sha, Qiuchen Li, Enji Meng, Xiangyu Liu, Liping Wang, Lin-Wang Tan, Guangming Jia, Weile Chinese Acad Sci Inst Comp Technol State Key Lab Proc Beijing Peoples R China Univ Chinese Acad Sci Beijing Peoples R China China Univ Petr Qingdao Inst Software Coll Comp Sci & Technol Qingdao Peoples R China Chinese Acad Sci Inst Semicond Beijing Peoples R China

ISBN: (纸本)9798400704352

Neural Network Molecular Dynamics (NNMD) has become a major approach in material simulations, which can speedup the molecular dynamics (MD) simulation for thousands of times, while maintaining ab initio accuracy, thus has a potential to fundamentally change the paradigm of material simulations. However, there are two time-consuming bottlenecks of the NNMD developments. One is the data access of ab initio calculation results. the other, which is the focus of the current work, is reducing the training time of NNMD model. the training of NNMD model is different from most other neural network training because the atomic force (which is related to the gradient of the network) is an important physical property to be fit. Tests show the traditional stochastic gradient methods, like the Adam algorithms, cannot efficiently deploy the multisample minibatch algorithm. As a result, a typical training (taking the Deep Potential Molecular Dynamics (DeePMD) as an example) can take many hours. In this work, we designed a heuristic minibatch quasi-Newtonian optimizer based on Extended Kalman Filter method. An early reduction of gradient and error is adopted to reduce memory footprint and communication. the memory footprint, communication and settings of hyper-parameters of this new method are analyzed in detail. Computational innovations such as customized kernels of the symmetry-preserving descriptor are applied to exploit the computing power of the heterogeneous architecture. Experiments are performed on 8 different datasets representing different real case situations, and numerical results show that our new method has an average speedup of 32.2 compared to the Reorganized Layer-wised Extended Kalman Filter with 1 GPU, reducing the absolute training time of one DeePMD model from hours to several minutes, making it one step toward online training.

关键词： parallel training Molecular dynamics First principle ab initio GPU

来源：评论

学校读者我要写书评

暂无评论

Application-Aware Management of parallel Simulation Collections

引用

acm sigplan NOTICES 2009年第4期44卷 35-44页

作者： Yau, Siu-Man Karamcheti, Vijay Damevski, Kostadin Parker, Steven G. Zorin, Denis NYU Courant Inst Math Sci New York NY 10003 USA Univ Utah Dept Comp Sci Salt Lake City UT 84112 USA

this paper presents a system deployed on parallel clusters to manage a collection of parallel simulations that make up a computational study. It explores how such a system can extend traditional parallel job scheduling and resource allocation techniques to incorporate knowledge specific to the study. Using a UINTAH-based helium gas simulation code (ARCHES) and the SimX system for multi-experiment computational studies, this paper demonstrates that, by using application-specific knowledge in resource allocation and scheduling decisions, one can reduce the run time of a computational study from over 20 hours to under 4.5 hours on a 32-processor cluster, and from almost 11 hours to just over 3.5 hours on a 64-processor cluster.

关键词： Design Performance parallel System High-throughput computing

来源：评论

学校读者我要写书评

暂无评论

Formal semantics for expressing optimism: the meaning of HOPE

Formal semantics for expressing optimism: the meaning of HOP...

引用

Proceedings of the 14th Annual acm symposium on principles of Distributed Computing

作者： Cowan, Crispin Lutfiyya, Hanan Oregon Graduate Inst Portland United States

Optimism is a powerful technique for increasing concurrency. A program can increase concurrency by making an optimistic assumption about its future state, and verifying the assumption in parallel with computations based on the optimistic assumption. the use of optimism has been restricted to specialized systems due to the difficulty of writing and understanding optimistic programs. In this paper, we define optimism as any computation that uses rollback. We present a formal semantics for expressing optimism by defining operations for concurrent programs to specify which optimistic computations to roll back, while automating the dependency tracking. We prove that these semantics guarantee some intuitively desirable behaviors. Finally, we describe our programming language implementation based on the semantics.

关键词： Computer systems programming

来源：评论

学校读者我要写书评

暂无评论

the integration of application and system based metrics in a parallel program performance tool 91

The integration of application and system based metrics in a...

引用

3rd acm sigplan symposium on principles and practice of parallel programming, PPOPP 1991

作者： Hollingsworth, Jeflrey K. Bruce Irvin, R. Miller, Barton P. Computer Sciences Department University of Wiseonsin-Madison 1210 W. Dayton Street MadisonWI53706 United States

ISBN: (纸本)0897913906

the IPS-2 parallel program measurement tools provide performance data from application programs, the operating system, hardware, network, and other sources. Previous versions of IPS-2 allowed programmers to collect information about an application based only on what could be collected by software instrumentation inserted into the program (and system call libraries). We have developed an open interface, called the "external time histogram", providing a graceful way to include external data from many sources. the user can tell IPS-2 of new sources of performance data through an extensible metric description language. the data from these external sources is automatically collected when the application program is run. IPS-2 provides a library to simplify constructing the external data collectors. the new version of IPS-2 can measure sharedmemory and message-passing parallel programs running on a heterogeneous collection of machines. Data from C or Fortran programs, and data from simulations ean be processed by the same tool. As a result of including the new external performance data, IPS-2 now can report on a whole new set of performance problems. We describe the results of using IPS-2 on two real applications: A shared-memory database join utility, and a multi-processor interconnection network simulator. Even though these applications previously went through careful tuning, we were able to precisely identify performance problems and extract additional performance improvements of about 30%. © 1991 Association for Computing Machinery. All Rights Reserved.

关键词： Application programs

来源：评论

学校读者我要写书评

暂无评论

Real-Time Matching of Antescofo Temporal Patterns 14

Real-Time Matching of <i>Antescofo</i> Temporal Patterns

引用

16th International symposium on principles and practice of Declarative programming (PPDP)

作者： Giavitto, Jean-Louis Echeveste, Jose IRCAM UMR STMS 9912 CNRS 1 Pl Igor Stravinsky F-75004 Paris France INRIA MuTant Project 1 Pl Igor Stravinsky F-75004 Paris France Univ Paris 06 Sorbonne Univ 1 Pl Igor Stravinsky F-75004 Paris France

ISBN: (纸本)9781450329477

online matching. Antescofo is a real-time system for performance coordination between musicians and computer processes during live music concert. ATP are used to define complex events that correspond to a combination of perceived events in the musical environment as well as arbitrary logical and metrical temporal conditions. the real-time recognition of such event is used to trigger arbitrary actions in the style of event-condition-action rules. the musical context, the rationales of temporal patterns and several illustrative examples are introduced to motivate the design of ATP. the semantics of ATP matching is defined to parallel the well-known notion of regular expression and Brzozowski's derivatives but extended to handle an infinite alphabet, arbitrary predicates, elapsing time and inhibitory conditions. this approach is compared to those developed in log auditing and for the runtime verification of realtime logics. ATP are implemented by translation into a core subset of the Antescofo domain-specific language. this compilation has proven efficient enough to avoid the extension of the real-time runtime of the language and has been validated with composers in actual pieces.

关键词： timed regular expressions event-driven programming score following timed and reactive system domain-specific language computer music Antescofo

来源：评论

学校读者我要写书评

暂无评论

NUMA-aware graph-structured analytics 2015

NUMA-aware graph-structured analytics

引用

20th acm sigplan symposium on principles and practice of parallel programming, PPoPP 2015

作者： Zhang, Kaiyuan Chen, Rong Chen, Haibo Shanghai Key Laboratory of Scalable Computing and Systems Institute of Parallel and Distributed Systems Shanghai Jiao Tong University China

ISBN: (纸本)9781450332057

Graph-structured analytics has been widely adopted in a number of big data applications such as social computation, web-search and recommendation systems. though much prior research focuses on scaling graph-analytics on distributed environments, the strong desire on performance per core, dollar and joule has generated considerable interests of processing large-scale graphs on a single server-class machine, which may have several terabytes of RAM and 80 or more cores. However, prior graph-analytics systems are largely neutral to NUMA characteristics and thus have suboptimal performance. this paper presents a detailed study of NUMA characteristics and their impact on the efficiency of graph-analytics. Our study uncovers two insights: 1) either random or interleaved allocation of graph data will significantly hamper data locality and parallelism;2) sequential inter-node (i.e., remote) memory accesses have much higher bandwidth than both intra- and inter-node random ones. Based on them, this paper describes Polymer, a NUMA-aware graph-analytics system on multicore with two key design decisions. First, Polymer differentially allocates and places topology data, application-defined data and mutable runtime states of a graph system according to their access patterns to minimize remote accesses. Second, for some remaining random accesses, Polymer carefully converts random remote accesses into sequential remote accesses, by using lightweight replication of vertices across NUMA nodes. To improve load balance and vertex convergence, Polymer is further built with a hierarchical barrier to boost parallelism and locality, an edge-oriented balanced partitioning for skewed graphs, and adaptive data structures according to the proportion of active vertices. A detailed evaluation on an 80-core machine shows that Polymer often outperforms the state-of-the-art single-machine graph-analytics systems, including Ligra, X-Stream and Galois, for a set of popular real-world and synthetic grap

关键词： Random access storage

来源：评论

学校读者我要写书评

暂无评论

Structure-Preserving Compilation: Efficient Integration of Functional DSLs into Legacy Systems 14

Structure-Preserving Compilation: Efficient Integration of F...

引用

16th International symposium on principles and practice of Declarative programming (PPDP)

作者： Kranz, Julian Simon, Axel Tech Univ Munich Inst Informat 2 Garching Germany

ISBN: (纸本)9781450329477

We present a translation scheme from a pure functional domain-specific language (DSL) to C. the over-arching idea of this scheme is to preserve the structure of the input program as much as possible. this includes, among other things, to refrain from inlining user-written functions and to retain variable names as much as possible. We apply this translation scheme to GDSL, a DSL used for the specification of decoders for machine instructions. GDSL offers non-trivial language features such as monadic actions that our translation scheme maps one-to-one to C statements, resulting in code that closely resembles hand-written C code. Indeed, it is feasible to debug and profile the program at the C level and to interface the generated code with existing C code without marshaling data. Our translation scheme is therefore an attractive starting point for a light-weight DSL since no other language-specific tools besides the compiler are necessary. Moreover, the generated code is amenable to compiler optimizations found in off-the-shelf C compilers. this is illustrated by the performance achieved by a decoder for x86 machine instructions implemented in GDSL which is as fast as a production-quality decoding library shipped by Intel.

关键词： Domain specific languages translation scheme interfacing between programming languages code optimization functional languages type inference instruction decoding

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：