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 105 papers for “MPICH2”

Rehm: An MPICH2 Channel Device Implementation over

René Grabner, Frank Mietke, Wolfgang Rehm · 2004

Abstract — MPICH2, the successor of one of the most popular open source message passing implementations, aims to fully support the MPI-2 standard. Due to a complete redesign, MPICH2 is also cleaner, more flexible, and faster. The InfiniBand network technology is an open industry standard and provides high bandwidth and low latency, as well as reliability, availability, serviceability (RAS) features. It is currently spreading its influence on the market of costeffective cluster computing. We expect for the near future that upcoming requirements in many cluster environments can only be satisfied by the functionality of MPICH2 and the performance of InfiniBand. Hence, there is the need for an effective support of the InfiniBand interconnect technology by MPICH2. In this paper we present our experience that has been gained during the implementation of our MPICH2 Device for InfiniBand. Further, a performance overview is given, as well as ideas for future developments. The device is implemented in terms of the Channel Interface (CH3) and uses both the channel semantics (Send/Receive) and memory semantics (RDMA) provided by Mellanox ’ Verbs implementation VAPI. With this combined approach a significant performance gain can be achieved. The design decisions discussed may also be of interest beyond the scope of this paper.

Cited by 0

Portable MPICH2 Clusters with Android Devices

DigiNole (Florida State University) · 2015

Recent surveys have shown that over 6.8 billion mobile phones are in use today where roughly 50% are running the Android Operating System. This large population of devices can be utilized for message-passing interface (MPI) computing. There have been few examples of this in practice due to the complexity of installation and lack of functionality. However, I propose a new framework which permits client devices to dynamically join and leave an MPI-based network without drastically affecting its performance. It uses the MPICH2 library to execute processes and pass messages between the devices. Unlike the previous implementations, this framework uses Java (in the form of an Android application) to perform the initialization steps and execution of MPICH. As a result, the purpose of this paper is to demonstrate the feasibility of implementing the framework as an independent library.

Cited by 0

Evaluation of WIEN2K Performance on MPICH2 vs. MPICH1

Hadi Khalilieh, Nidal Kafri · 2014

The emerging multi-core computer architecture attracts the researchers to utilize this architecture as an adequate and inexpensive solution to achieve high performance computation for many problems. Where, the multi-core architecture enables us to implement shared memory and/or message passing parallel processing paradigms. Therefore, we need appropriate standard software libraries in order to utilize the resources efficiently for a given computational problem. In this work we evaluate the performance of two versions of the well known massage passing interface (MPI) library: MPICH1 vs. MPICH2. In our experiments we used two benchmarks. The first one is the WIEN2K application which is based on Density Function Theory, and the second is a Matrix multiplication. The results show that we achieve better performance when MPICH2 is used than MPICH1.

Cited by 0

PARALLEL FUZZY LOGIC CONTROLLER IMPLEMENTATION USING MPICH2

Bakir A. R. Al-Hashemy, AboTalib H. Mahfoodh · Journal of Engineering · 2010

In this work FLC program is implemented using C++ codes. Two implementations are presented one with the rules stored inside the program, the other with rules in a rulebase file. The execution times of these two implementations, along with MATLAB FLC implementation, are compared using different simulated FLCs. Furthermore, to reduce the rulebase searching time, a parallel FLC is implemented using C++ and MPI (Message Passing Interface). The MPICH2 package is used to run the parallel FLC. A cluster of four computers is used as the parallel environment. The execution time of this FLC program is evaluated using servomotor, Anti Skid System, and other simulated applications. The speedup and efficiency are studied using different number of computers. The results show that decomposing the rulebase searching operation to more than a computer reduce the execution time significantly.

Cited by 0Open access

Shared Memory Support for InfiniBand MPICH2-Device

Frank Mietke, Marco Steiger, Torsten Mehlan et al. · 2004

With the Top500 list from June 2004, cluster systems exceeded not only the 50 % threshold in number of systems the first time, but also in total performance. Due to the good price-performance ratio of cluster systems, these rates will further increase in the next lists. Multiprocessor nodes – typically 2 or 4 way systems – can improve that price-performance ratio in clusters. This paper presents our MPICH2 device called SHIBA (Shared-Memory + InfiniBand) which is the logical successor of our InfiniBand-only device to follow the above mentioned trend. We aim to provide a fast connection for intra-node communication and therewith to disburden the PCI connection to the InfiniBand Host Channel Adapter (HCA) for inter-node communication. 1

Cited by 0

ES-MPICH2: A Message Passing Interface with enhanced security

Xiaojun Ruan, Qing Yang, I. Alghamdi Mohammed et al. · International Performance, Computing, and Communications Conference · 2010

In largely distributed clusters, computing nodes are geographically deployed in various computing sites. Information processed in a distributed cluster is shared among a group of distributed processes or users by virtue of messages passing protocols (e.g. message passing interface - MPI) running on the Internet. Because of the open accessible nature of the Internet, data encryption for these large-scale distributed clusters becomes a non-trivial and challenging problem. To address this issue, we enhanced the security of the MPI (Message Passing Interface) protocol by encrypting and decrypting messages sent and received among computing nodes. In this study we focused on MPI rather than other protocols because MPI is one of the most popular communication protocols for cluster computing environments. From among a variety of MPI implementations, we picked MPICH2 developed by the Argonne National Laboratory. The design goal of MPICH2 - a widely used MPI implementation - is to combine portability with high performance. We integrated encryption algorithms into the MPICH2 library so that data confidentiality of MPI applications could be readily preserved without a need to change the source codes of the MPI applications. since we provide a security enhanced MPI-library with the standard MPI interfact, data communications of a conventional MPI program can be secured without converting the program into the corresponding secure version. We used Sandia Micro Benchmark and Intel MPI Benchmarks to evaluate and compared the performance of original MPICH2 and Enhanced Security MPICH2. According to the performance evaluation, ES-MPICH2 provides secured Message Passing Interface by sacrificing reasonable system performance.

Cited by 8

ES-MPICH2: Message passing interface with enhanced key agreement

J. Sudha, Vishnu Vardhan Paranthaman · 2013

In the concept of MPI chatting and file transmission are implemented, the decryption part will be done automatically. Here we use three methods in the keys; they are public, private and secret key. Keys are displayed to the destination only if they accept the request or else displaying of key is not possible in the destination side and also it won't give or establish the Connection. In largely spread clusters, computing nodes are naturally deployed in a variety of computing sites. The Information processed in a spread cluster is communal among a group of distributed processes or client by high-quality of messages passing protocols (e.g. message passing interface - MPI) running on the Internet. Because of the open available nature of the Internet, data encryption for these large-scale distributed clusters becomes a non-trivial and challenging problem. We improved the security of the MPI (Message Passing Interface) protocol by encrypting and decrypting messages sent and received among computing nodes. We are listening carefully on MPI rather than more protocols because MPI is one of the most accepted communication protocols for cluster computing environments. From among a multiple of MPI implementations, we selected MPICH2 developed by the Argonne National Laboratory. Design goal of MPICH2 - a commonly use MPI implementation - is to join portability with high presentation. we gives a security enhanced MPI-library with the standard MPI interface, data communications of a conservative MPI program can be secured without converting the program into the corresponding secure report. We included encryption algorithms into the MPICH2 library so that data in secret of MPI applications could be readily preserved without require modifying the source codes of the MPI applications.

Cited by 1

Configurando um cluster no Fedora Core 4 com MPICH2

Simões Junior, Fernando Jaques Ruiz · Biblioteca Digital da Memória Científica do INPE (National Institute for Space Research) · 2008

A utilização de conjuntos de computadores (clusters) tornou-se uma poderosa ferramenta para solução de problemas complexos. Com a popularização da informática, os computadores domésticos (PCs) tornaram-se muito baratos, de forma que a utilização de vários PCs trabalhando em paralelo surgiu como uma alternativa para utilização de processamento de alto desempenho. Neste guia vamos realizar detalhadamente todos os passos para configurar uma conjunto de micros de modo à obter-se um cluster para processamento paralelo. A idéia deste trabalho é oferecer os passos operacionais do processo de configuração, portanto não vamos entrar em discussões profundas dos detalhes de cada procedimento.

Cited by 0Open access

A Secure Mechanism for MPI in Parllel Computing

Uday Kishore Pusuluri · 2013

An increasing number of commodity clusters are connected to each other by public networks, which have become a potential threat to security sensitive parallel applications running on the clusters. To address Interface (MPI) implementation to preserve confidentiality of messages communicated among nodes of clusters in an unsecured network. We focus on MPI rather than other protocols, because MPI is one of the most popular communication protocol for parallel computing on clusters. Our MPI implementation—called ES-MPICH2 was built based on MPICH2 developed by the Argonne National Laboratory. Like MPICH2, ES-MPICH2 aims at supporting a large variety of computation and communication platforms like commodity clusters and high-speed networks. We integrated encryption and decryption algorithms into the MPICH2library with the standard MPI interface and; thus, data confidentiality of MPI applications can be readily preserved without a need to change the source codes of the MPI applications. MPI-application programmers can fully configure any confidentiality services inMPICHI2, because a secured configuration file in ES-MPICH2 offers the programmers flexibility in choosing any cryptographic schemes and keys seamlessly incorporated in ES-MPICH2. We used the Sandia Micro Benchmark and Intel MPI Benchmark suites to evaluate and compare the performance of ES-MPICH2 with the original MPICH2 version. Our experiments show that overhead incurred by the confidentiality services in ES-MPICH2 is marginal for small messages. The security overhead in ES-MPICH2 becomes more pronounced with larger messages. Our results also show that security overhead can be significantly reduced in ES-MPICH2 by high-performance clusters. The executable binaries and source code of the ES-MPICH2 implementation are freely available at The executable binaries and source code of the ES-MPICH2 implementation are freely available at http://www.eng.auburn. edu/~xqin/software/es-mpich2/

Cited by 0

Design and Implementation of MPICH2 over InfiniBand with RDMA Support

Jiuxing Liu, Weihang Jiang, Pete Wyckoff et al. · 2004

Summary form only given. For several years, MPI has been the de facto standard for writing parallel applications. One of the most popular MPI implementations is MPICH. Its successor, MPICH2, features a completely new design that provides more performance and flexibility. To ensure portability, it has a hierarchical structure based on which porting can be done at different levels. In this paper, we present our experiences in designing and implementing MPICH2 over InfiniBand. Because of its high performance and open standard, InfiniBand is gaining popularity in the area of high-performance computing. Our study focuses on optimizing the performance of MPl-1 functions in MPICH2. One of our objectives is to exploit remote direct memory access (RDMA) in InfiniBand to achieve high performance. We have based our design on the RDMA channel interface provided by MP1CH2, which encapsulates architecture-dependent communication functionalities into a very small set of functions. Starting with a basic design, we apply different optimizations and also propose a zero-copy-based design. We characterize the impact of our optimizations and designs using microbenchmarks. We have also performed an application-level evaluation using the NAS parallel benchmarks. Our optimized MPICH2 implementation achieves 7.6/spl mu/s latency and 857 MB/s bandwidth, which are close to the raw performance of the underlying InfiniBand layer. Our study shows that the RDMA channel interface in MPICH2 provides a simple, yet powerful, abstraction that enables implementations with high performance by exploiting RDMA operations in InfiniBand. To the best of our knowledge, this is the first high-performance design and implementation ofMPICH2 on InfiniBand using RDMA support.

Cited by 87

Hybrid Programming Fun: Making Bzip2 Parallel with MPICH2 &

2014

ABSTRACT: The author shares his programming experience and performance analysis in making a parallel file compression program for the Cray XD1. Discussion includes a practical example of combining MPI and pthreads in a single application. Issues with thread safety and MPI relative to the example program and the XD1 are covered. The author includes an analysis of MPICH2 using TCP/IP over RapidArray.

Cited by 0

Pemanfaatan Teknologi Cluster Berbasis MPICH2 untuk Perkalian Matriks Menggunakan Message Passing

Prasetyo, Heru · Universitas Kristen Satya Wacana Institutional Repository (Universitas Kristen Satya Wacana) · 2013

When seen from the development of parallel computing technology today, the cluster systems requires the provision of computing resources in large numbers. Due to the provision of these resources, it needed a system that can improve the performance of computing for data processing maximally. This study analyzing performance of the system that occurs in the cluster, so as to enhance the performance of parallel computing. The cluster systems that are used for parallel computing using message passing interface algorithms (MPI) to perform parallel matrix multiplication with speed-up and efficiency as the parameters. The method used is by experiments using 2 computer cluster systems for the implementation of inter-connected and using MPICH2 as middleware, and then do an analysis the performance of the system in handling the matrix multiplication test scenarios with different orders. The results shown by the matrix multiplication with stand alone computer and the cluster indicates that the cluster system is faster if compared with a stand alone computer. This indicates that the implementation of MPICH2 parallel algorithms for matrix multiplication application on a cluster can improve the performance of parallel computing performance.

Cited by 0Open access

Design and evaluation of Nemesis, a scalable, low-latency, message-passing communication subsystem

Darius Buntinas, Guillaume Mercier, William Gropp · HAL (Le Centre pour la Communication Scientifique Directe) · 2006

This paper presents a new low-level communication subsystem called Nemesis. Nemesis has been designed and implemented to be scalable and efficient both in the intranode communication context using shared-memory and in the internode communication case using high-performance networks and is natively multimethod-enabled. Nemesis has been integrated in MPICH2 as a CH3 channel and delivers better performance than other dedicated communication channels in MPICH2. Furthermore, the resulting MPICH2 architecture outperforms other MPI implementations in point-to-point benchmarks.

Cited by 109Open access

COMPUTER PERFORMANCE ANALYSIS OF CLUSTER AND NON CLUSTER

Aridhatullah · Electronic theses and dissertations (Syiah Kuala University) · 2015

This final project discusses about how to create a cluster computer by usingMPICH2 then be analyzed the speed ratio between the computing performancecomputer cluster and non cluster. The purpose of writing of this project is tocreate five PCs which are connected into network and will form a clustercomputer by using MPICH2. Later, the performance of computing MPICH2 incluster computer will be analyzed. The phase on creating this cluster computerwill be started with designed topology which is used, doing the installation andconfiguration of MPICH2 to the five PCs, then will perform computational testing by performing matrix multiplication as well as collecting data from its comparison. These computers accelerate computing performance and shorten thetime of wide scale of calculation.Key words :cluster, MPICH2, matrix

Cited by 0