A comparative study of Java and C performance in two large-scale parallel applications

In the 1990s the Message Passing Interface Forum defined MPI bindings for Fortran, C, and C++. With the success of MPI these relatively conservative languages have continued to dominate in the parallel computing community. There are compelling arguments in favour of more modern languages like Java. These include portability, better runtime error checking, modularity, and multi-threading. But these arguments have not converted many HPC programmers, perhaps due to the scarcity of full-scale scientific Java codes, and the lack of evidence for performance competitive with C or Fortran. This paper tries to redress this situation by porting two scientific applications to Java. Both of these applications are parallelized using our thread-safe Java messaging system—MPJ Express. The first application is the Gadget-2 code, which is a massively parallel structure formation code for cosmological simulations. The second application uses the finite-domain time-difference method for simulations in the area of computational electromagnetics. We evaluate and compare the performance of the Java and C versions of these two scientific applications, and demonstrate that the Java codes can achieve performance comparable with legacy applications written in conventional HPC languages. Copyright © 2009 John Wiley & Sons, Ltd.

Towards self-ware via swarm-array computing

The work reported in this paper proposes Swarm-Array computing, a novel technique inspired by swarm robotics, and built on the foundations of autonomic and parallel computing. The approach aims to apply autonomic computing constructs to parallel computing systems and in effect achieve the self-ware objectives that describe self-managing systems. The constitution of swarm-array computing comprising four constituents, namely the computing system, the problem/task, the swarm and the landscape is considered. Approaches that bind these constituents together are proposed. Space applications employing FPGAs are identified as a potential area for applying swarm-array computing for building reliable systems. The feasibility of a proposed approach is validated on the SeSAm multi-agent simulator and landscapes are generated using the MATLAB toolkit.

Achieving intelligent agents and its feasibility in swarm-array computing

The work reported in this paper proposes ‘Intelligent Agents’, a Swarm-Array computing approach focused to apply autonomic computing concepts to parallel computing systems and build reliable systems for space applications. Swarm-array computing is a robotics a swarm robotics inspired novel computing approach considered as a path to achieve autonomy in parallel computing systems. In the intelligent agent approach, a task to be executed on parallel computing cores is considered as a swarm of autonomous agents. A task is carried to a computing core by carrier agents and can be seamlessly transferred between cores in the event of a predicted failure, thereby achieving self-* objectives of autonomic computing. The approach is validated on a multi-agent simulator.

Exploring carrier agents in swarm-array computing

How can a bridge be built between autonomic computing approaches and parallel computing systems? The work reported in this paper is motivated towards bridging this gap by proposing a swarm-array computing approach based on ‘Intelligent Agents’ to achieve autonomy for distributed parallel computing systems. In the proposed approach, a task to be executed on parallel computing cores is carried onto a computing core by carrier agents that can seamlessly transfer between processing cores in the event of a predicted failure. The cognitive capabilities of the carrier agents on a parallel processing core serves in achieving the self-ware objectives of autonomic computing, hence applying autonomic computing concepts for the benefit of parallel computing systems. The feasibility of the proposed approach is validated by simulation studies using a multi-agent simulator on an FPGA (Field-Programmable Gate Array) and experimental studies using MPI (Message Passing Interface) on a computer cluster. Preliminary results confirm that applying autonomic computing principles to parallel computing systems is beneficial.

Scaling up classification rule induction through parallel processing

The fast increase in the size and number of databases demands data mining approaches that are scalable to large amounts of data. This has led to the exploration of parallel computing technologies in order to perform data mining tasks concurrently using several processors. Parallelization seems to be a natural and cost-effective way to scale up data mining technologies. One of the most important of these data mining technologies is the classification of newly recorded data. This paper surveys advances in parallelization in the field of classification rule induction.

Non-uniform data distribution for communication-efficient parallel clustering

Global communicationrequirements andloadimbalanceof someparalleldataminingalgorithms arethe major obstacles to exploitthe computational power of large-scale systems. This work investigates how non-uniform data distributions can be exploited to remove the global communication requirement and to reduce the communication costin parallel data mining algorithms and, in particular, in the k-means algorithm for cluster analysis. In the straightforward parallel formulation of the k-means algorithm, data and computation loads are uniformly distributed over the processing nodes. This approach has excellent load balancing characteristics that may suggest it could scale up to large and extreme-scale parallel computing systems. However, at each iteration step the algorithm requires a global reduction operationwhichhinders thescalabilityoftheapproach.Thisworkstudiesadifferentparallelformulation of the algorithm where the requirement of global communication is removed, while maintaining the same deterministic nature ofthe centralised algorithm. The proposed approach exploits a non-uniform data distribution which can be either found in real-world distributed applications or can be induced by means ofmulti-dimensional binary searchtrees. The approachcanalso be extended to accommodate an approximation error which allows a further reduction ofthe communication costs. The effectiveness of the exact and approximate methods has been tested in a parallel computing system with 64 processors and in simulations with 1024 processing element

Dynamic group communication for large-scale parallel data mining

Exascale systems are the next frontier in high-performance computing and are expected to deliver a performance of the order of 10^18 operations per second using massive multicore processors. Very large- and extreme-scale parallel systems pose critical algorithmic challenges, especially related to concurrency, locality and the need to avoid global communication patterns. This work investigates a novel protocol for dynamic group communication that can be used to remove the global communication requirement and to reduce the communication cost in parallel formulations of iterative data mining algorithms. The protocol is used to provide a communication-efficient parallel formulation of the k-means algorithm for cluster analysis. The approach is based on a collective communication operation for dynamic groups of processes and exploits non-uniform data distributions. Non-uniform data distributions can be either found in real-world distributed applications or induced by means of multidimensional binary search trees. The analysis of the proposed dynamic group communication protocol has shown that it does not introduce significant communication overhead. The parallel clustering algorithm has also been extended to accommodate an approximation error, which allows a further reduction of the communication costs. The effectiveness of the exact and approximate methods has been tested in a parallel computing system with 64 processors and in simulations with 1024 processing elements.

Landscape of intelligent cores: An autonomic multi-agent approach for space application

Space applications are challenged by the reliability of parallel computing systems (FPGAs) employed in space crafts due to Single-Event Upsets. The work reported in this paper aims to achieve self-managing systems which are reliable for space applications by applying autonomic computing constructs to parallel computing systems. A novel technique, 'Swarm-Array Computing' inspired by swarm robotics, and built on the foundations of autonomic and parallel computing is proposed as a path to achieve autonomy. The constitution of swarm-array computing comprising for constituents, namely the computing system, the problem / task, the swarm and the landscape is considered. Three approaches that bind these constituents together are proposed. The feasibility of one among the three proposed approaches is validated on the SeSAm multi-agent simulator and landscapes representing the computing space and problem are generated using the MATLAB.

Swarms of computational resources and swarms of tasks for building reliable systems

Space applications demand the need for building reliable systems. Autonomic computing defines such reliable systems as self-managing systems. The work reported in this paper combines agent based and swarm robotic approaches leading to swarm-array computing, a novel technique to achieve autonomy for distributed parallel computing systems. Two swarm-array computing approaches based on swarms of computational resources and swarms of tasks are explored. FPGA is considered as the computing system. The feasibility of the two proposed approaches that binds the computing system and the task together is simulated on the SeSAm multi-agent simulator.

Locally twisted cubes are 4-pancyclic

The locally twisted cube is a newly introduced interconnection network for parallel computing. Ring embedding is an important issue for evaluating the performance of an interconnection network. In this paper, we investigate the problem of embedding rings into a locally twisted cube. Our main contribution is to find that, for each integer l is an element of (4,5,...,2(n)}, a ring of length I can be embedded into an n-dimensional locally twisted cube so that both the dilation and the load factor are one. As a result, a locally twisted cube is Hamiltonian. We conclude that a locally twisted cube is superior to a hypercube in terms of ring embedding capability. (C) 2004 Elsevier Ltd. All rights reserved.

Pancyclicity of Mobius cubes with faulty nodes

An interconnection network with n nodes is four-pancyclic if it contains a cycle of length l for each integer l with 4 <= l <= n. An interconnection network is fault-tolerant four-pancyclic if the surviving network is four-pancyclic in the presence of faults. The fault-tolerant four-pancyclicity of interconnection networks is a desired property because many classical parallel algorithms can be mapped onto such networks in a communication-efficient fashion, even in the presence of failing nodes or edges. Due to some attractive properties as compared with its hypercube counterpart of the same size, the Mobius cube has been proposed as a promising candidate for interconnection topology. Hsieh and Chen [S.Y. Hsieh, C.H. Chen, Pancyclicity on Mobius cubes with maximal edge faults, Parallel Computing, 30(3) (2004) 407-421.] showed that an n-dimensional Mobius cube is four-pancyclic in the presence of up to n-2 faulty edges. In this paper, we show that an n-dimensional Mobius cube is four-pancyclic in the presence of up to n-2 faulty nodes. The obtained result is optimal in that, if n-1 nodes are removed, the surviving network may not be four-pancyclic. (C) 2005 Elsevier B.V. All rights reserved.

Implementing intelligent cores using processor virtualization for fault tolerance

Processor virtualization for process migration in distributed parallel computing systems has formed a significant component of research on load balancing. In contrast, the potential of processor virtualization for fault tolerance has been addressed minimally. The work reported in this paper is motivated towards extending concepts of processor virtualization towards ‘intelligent cores’ as a means to achieve fault tolerance in distributed parallel computing systems. Intelligent cores are an abstraction of the hardware processing cores, with the incorporation of cognitive capabilities, on which parallel tasks can be executed and migrated. When a processing core executing a task is predicted to fail the task being executed is proactively transferred onto another core. A parallel reduction algorithm incorporating concepts of intelligent cores is implemented on a computer cluster using Adaptive MPI and Charm ++. Preliminary results confirm the feasibility of the approach.

An MPI-based implementation of intelligent agents on clusters

Recent research in multi-agent systems incorporate fault tolerance concepts, but does not explore the extension and implementation of such ideas for large scale parallel computing systems. The work reported in this paper investigates a swarm array computing approach, namely 'Intelligent Agents'. A task to be executed on a parallel computing system is decomposed to sub-tasks and mapped onto agents that traverse an abstracted hardware layer. The agents intercommunicate across processors to share information during the event of a predicted core/processor failure and for successfully completing the task. The feasibility of the approach is validated by implementation of a parallel reduction algorithm on a computer cluster using the Message Passing Interface.

Intelligent agents for fault tolerance: from multi-agent simulation to cluster-based implementation

Recent research in multi-agent systems incorporate fault tolerance concepts, but does not explore the extension and implementation of such ideas for large scale parallel computing systems. The work reported in this paper investigates a swarm array computing approach, namely 'Intelligent Agents'. A task to be executed on a parallel computing system is decomposed to sub-tasks and mapped onto agents that traverse an abstracted hardware layer. The agents intercommunicate across processors to share information during the event of a predicted core/processor failure and for successfully completing the task. The feasibility of the approach is validated by simulations on an FPGA using a multi-agent simulator, and implementation of a parallel reduction algorithm on a computer cluster using the Message Passing Interface.