Skip to content
Free planNo credit card requiredStart writing
AI for Research home
AI for Research home
Dashboard
Sign inGet started

Search 250 million research papers

Read the abstract, see who wrote it and who cites it, and get the citation in APA, MLA, Chicago or Harvard.

About 10,000 papers for “distributed memory systems”

Dynamical distributed memory systems

Kazuhiro Kojima, Kenichiro Ito · 2003

We propose an autonomous dynamical pattern recognition and learning system. It is demonstrated that when the embedded pattern, i.e., known pattern, is given to the network, the firing pattern of the network immediately goes to the relevant embedded pattern and the network state reduces to the oscillatory state at once. Second, when no embedded pattern, i.e., unknown pattern, is given to the network, the network state oscillates chaotically. It is considered as "I don't know" state proposed by Freeman and coworkers. Finally, when Hebb rule is applied to the network under the external stimuli that are unknown patterns, the internal state of the network is inversely bifurcated from the chaotic state to the periodic state according to the progress of learning. By using this phase transition as an index of the progress of learning, the network can learn new patterns without any external observers.

Cited by 0

Finite Memory Distributed Systems

Dorofeenko Victor, Shorish Jamsheed · RePEc: Research Papers in Economics

A distributed system model is studied, where individual agents play repeatedly against each other and change their strategies based upon previous play. It is shown how to model this environment in terms of continuous population densities of agent types. A complication arises because the population densities of different strategies depend upon each other not only through game payoffs, but also through the strategy distributions themselves. In spite of this, it is shown that when an agent imitates the strategy of his previous opponent at a sufficiently high rate, the system of equations which governs the dynamical evolution of agent populations can be reduced to one equation for the total population. In a sense, the dynamics 'collapse' to the dynamics of the entire system taken as a whole, which describes the behavior of all types of agents. We explore the implications of this model, and present both analytical and simulation results.

Cited by 0

Finite Memory Distributed Systems

Victor Dorofeenko, Jamsheed Shorish · Contributions in Theoretical Economics · 2006

A distributed system model is studied, where individual agents play repeatedly against each other and change their strategies based upon previous play. It is shown how to model this environment in terms of continuous population densities of agent types. A complication arises because the population densities of different strategies depend upon each other not only through game payoffs, but also through the strategy distributions themselves. In spite of this, it is shown that when an agent imitates the strategy of his previous opponent at a sufficiently high rate, the system of equations which governs the dynamical evolution of agent populations can be reduced to one equation for the total population. In a sense, the dynamics 'collapse' to the dynamics of the entire system taken as a whole, which describes the behavior of all types of agents. We explore the implications of this model, and present both analytical and simulation results.

Cited by 3Open access

Aggregate Distributed Objects for Distributed Memory Parallel Systems

W. L. Bain · 2005

This paper describes a class of distributed objects, called a g g r e g a t e objects, that can be partitioned across the processing nodes of a parallel system. Element objects are accessed according to the logical structure of the aggregate object. This simplifies the design of applications that access multiple, logically related objects. It also provides opportunities to exploit the large memory and processing power of parallel systems, and it simplifies the load balancing of these systems. The paper also describes how aggregate objects can be created and accessed using the Interwork IIm Concurrent Programming Toolkit on the Intel iPSC Parallel System. This toolkit allows the programmer to define new aggregate object types which extend Interwork 11's semantics for simple objects. It also enables efficient, fully parallel access to these objects from multiple processing nodes. Using this toolkit, an aggregate object defined as an n-dimensional array of simple objects was defined. This object type was used to build a large discrete event simulation.

Cited by 4

Compiling shared-memory applications for distributed-memory systems

Basumallik, Ayon · Purdue e-Pubs (Purdue University System) · 2007

OpenMP has established itself as the de facto standard for parallel programming on shared-memory platforms. OpenMP provides a simpler approach to parallel programming, allowing programs to be parallelized incrementally by the insertion of directives. In contrast, parallel programming on distributed-memory systems using message-passing is effort intensive. It requires the programmer to parallelize programs as a whole and to explicitly manage data transfers. This dissertation aims to extend the ease of parallel programming in OpenMP to distributed-memory systems as well. To that end, we propose two approaches. In the first approach, we use an underlying layer of Software Distributed Shared Memory (SDSM). However, SDSM systems have some inherent performance limitations. Therefore, this dissertation explores a second approach - direct translation of OpenMP applications to MPI. We present the basic translation scheme and optimizations for both regular and irregular OpenMP applications. Experiments with seven OpenMP benchmarks indicate that OpenMP applications translated directly to MPI using our techniques achieve average scalability within 12% of their hand-coded MPI counterparts, a 30% higher average scalability than corresponding SDSM applications and between 12% to 89% higher scalability than corresponding applications written in High Performance Fortran (HPF). A combined compile-time/runtime scheme, evaluated using three representative irregular OpenMP applications, achieves performance to within 10% of its MPI counterpart in one case and outperforms its hand-coded MPI counterparts in the other two cases.

Cited by 0

Towards distributed software transactional memory systems

Paolo Romano, Nuno Carvalho, Luı́s Rodrigues · 2008

The recent architectural trend that has lead to the widespread adoption of multi-core CPUs has fostered a remarkable research interest in Software Transactional Memory (STM). As STMs are starting to face the high availability and scalability requirements of real-world production environments, it is natural to foresee the need for replication solutions specifically tailored for STMs. Since databases and STMs share the same key abstraction of atomic transaction, one could wonder whether the mechanisms originally designed for database replication could be successfully and seamlessly exploited also to support replication of STM systems.

Cited by 42

Distributed shared memory: concepts and systems

Jelica Protić, Milo Tomašević, Veljko Milutinović · IEEE Parallel & Distributed Technology Systems & Applications · 1996

In surveying current approaches to distributed shared memory computing, the authors find that the reduced cost of parallel software development will help make the DSM paradigm a viable solution to large scale, high performance computing.

Cited by 286

ANALYSIS ON DISTRIBUTED MEMORY PARALLEL SYSTEMS

Chao Yang, Padma Raghavan, Lloyd Arrowood et al. · 2008

A parallel computational scheme for analyzing large-scale molecular vibration on distributed memory computing platforms is presented in this paper. This method combines the implicitly restarted Lanczos algorithm with a state-of-art parallel sparse direct solver to compute a set of low frequency vibrational modes for molecular systems containing tens of thousands of atoms. Although the original motivation for developing such a scheme was to overcome memory limitations on traditional sequential and shared memory machines, our computational experiments show that with a careful parallel design and data partitioning scheme one can achieve scalable performance on lightly coupled distributed memory parallel systems. In particular, we demonstrate performance enhancement achieved by using the latency tolerant “selective inversion" scheme in the sparse triangular substitution phase of the computation.

Cited by 0

Gang Scheduling for Distributed Memory Systems

William Leinberger, George Karypis, Vipin Kumar · University of Minnesota Digital Conservancy (University of Minnesota) · 2000

Time-sliced gang scheduling improves the average response time of the jobs in a typical job stream. Recent research has shown that time-slicing is most effective when the jobs admitted for execution fit entirely into physical memory. We investigate two techniques for improving the performance of gang scheduling in the presence of memory pressure: 1) a novel backfill approach which improves memory utilization, and 2) an adaptive multi-programming level which balances processor/memory utilization with job response time performance. Our simulations show that these techniques reduce the slow-down performance metrics over naive FCFS methods on a distributed memory parallel system.

Cited by 2Open access

Monitoring and Debugging Distributed Memory Systems

Ys Te Ms, Jos'e C. Cunha, Henryk Krawczyk et al. · 2007

In this paper we present the objectives and design issues for two components of a parallel programming environment: a monitoring layer and a testing and debugging tool. These components are integral part of an environment that is being developed within the context of the SEPP (Software Engineering for Parallel Processing) COPERNICUS Project, investigating issues in the design and implementation of visual programming components for graph-based design of parallel programs, mapping components for specific hardware target distributed-memory architectures, simulation support tools, real parallel execution with monitoring and load-balancing support, and testing and debugging tools.

Cited by 0

Transactions in distributed shared memory systems

Peter Bodorik, Forrest Smith, D.J. Lewis · 2003

The authors propose a distributed shared memory model based on a paged segmented two-level address space and an extended set of memory operations. In addition to the traditional read and write operations, the memory model includes operations which support mapping between local and global address spaces and mapping of processes to transactions. An architecture and associated algorithm are outlined for a virtual memory management unit to provide concurrency control for transactions. Although the traditional concept of the transaction is assumed, only the aspects of concurrency control and coherence are addressed.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Cited by 9

Optimizing irregular shared-memory applications for distributed-memory systems

Ayon Basumallik, Rudolf Eigenmann · 2006

In prior work, we have proposed techniques to extend the ease of shared-memory parallel programming to distributed-memory platforms by automatic translation of OpenMP programs to MPI. In the case of irregular applications, the performance of this translation scheme is limited by the fact that accesses to shared-data cannot be accurately resolved at compile-time. Additionally, irregular applications with high communication to computation ratios pose challenges even for direct implementation on message passing systems. In this paper, we present combined compile-time/run-time techniques for optimizing irregular shared-memory applications on message passing systems in the context of automatic translation from OpenMP to MPI. Our transformations enable computation-communication overlap by restructuring irregular parallel loops. The compiler creates inspectors to analyze actual data access patterns for irregular accesses at runtime. This analysis is combined with the compile-time analysis of regular data accesses to determine which iterations of irregular loops access non-local data. The iterations are then reordered to enable computation-communication overlap. In the case where the irregular access occurs inside nested loops, the loop nest is restructured. We evaluate our techniques by translating OpenMP versions of three benchmarks from two important classes of irregular applications - sparse matrix computations and molecular dynamics. We find that for these applications, on sixteen nodes, versions employing computation-communication overlap are almost twice as fast as baseline OpenMP-to-MPI versions, almost 30% faster than inspector-only versions, almost 25% faster than hand-coded versions on two applications and about 9% slower on the third.

Cited by 67