Advances in communication for parallel programming have yielded one-sided messaging systems. The MPI bindings for Ruby have been augmented to include the remote memory access functions of MPI-2.
ISBN:
(纸本)0780321754
Advances in communication for parallel programming have yielded one-sided messaging systems. The MPI bindings for Ruby have been augmented to include the remote memory access functions of MPI-2.
We present an efficient connection caching scheme in a distributed system where we divide the system into several host groups possibly overlapping and the connection between the hosts in the same group is kept prior t...
详细信息
ISBN:
(纸本)0818675578
We present an efficient connection caching scheme in a distributed system where we divide the system into several host groups possibly overlapping and the connection between the hosts in the same group is kept prior to others. Every host group consists of hosts which have heavy intercommunication. We present performance evaluation of the proposed connection caching scheme in several aspects including group size, grouping and types of the group. Simulation results show that host grouping is effective in every performance criterion and that proper grouping of hosts enhances the performance. We also present a simple Markov process model of our scheme and give some analysis results which are consistent with the simulation results.
In this paper we present the concept and first prototyping results of a modular fault-tolerant distributed mass storage architecture for large Linux PC clusters as they are deployed by the upcoming particle physics ex...
详细信息
ISBN:
(纸本)0769519199
In this paper we present the concept and first prototyping results of a modular fault-tolerant distributed mass storage architecture for large Linux PC clusters as they are deployed by the upcoming particle physics experiments. The device masquerading technique using an Enhanced Network Block Device (ENBD) enables local RAID over remote disks as the key concept of the ClusterRAID system. The block level interface to remote files, partitions or disks provided by the ENBD makes it possible to use the standard Linux software RAID to add fault-tolerance to the system. Preliminary performance measurements indicate that the latency is comparable to a local hard drive. With four disks throughput rates of up to 55MB/s were achieved with first prototypes for a RAIDO setup, and about 40MB/s for a RAID5 setup.
This paper proposes a highperformance communication facility using multiple commodity network interface cards (NICs). Called PM/Ethernet-kRMA, it is NIC-hardware-independent and provides (k)ernel-level Remote Memory ...
详细信息
ISBN:
(纸本)0769519199
This paper proposes a highperformance communication facility using multiple commodity network interface cards (NICs). Called PM/Ethernet-kRMA, it is NIC-hardware-independent and provides (k)ernel-level Remote Memory Access (kRMA) on multiple NICs. The PM/Ethernet-kRMA communication protocol is processed on the host processor and the protocol handler accesses user data space directly from the kernel, and then transfers the data to network using existing network device drivers. This protocol provides one-copy communication between user memory spaces on kernel. The PM/Ethernet-kRMA is implemented using the PM/Ethernet, one of the communication facilities of the SCore Cluster system software on Linux. The PM/Ethernet uses the Network Trunking technique, which provides message communication using multiple NICs. Existing protocols, such as TCP/IP, can be used on the PM/Ethernet-kRMA as well as the PM/Ethernet. We have evaluated the PM/Ethernet-kRMA using 2-node single Xeon 2.4GHz processor machines with three Intel PRO/ 1000 XTs and one Broadcom 5701 based Gigabit Ethernet NICs on each node. Network Trunking provides 420 MB/s of communication bandwidth using four Gigabit Ethernet NICs. PM/Ethernet-kRMA using four Gigabit Ethernet NICs, in contrast, provides 487 MB/s of bandwidth which is 97.4% of hardware-level bandwidth (500 MB/s).
In this paper, three main parts of BAG real-time distributed operating system are introduced: task migration, load balancing and distributed file system. Task migration, based on EFSM programming model, is implemented...
详细信息
In this paper, three main parts of BAG real-time distributed operating system are introduced: task migration, load balancing and distributed file system. Task migration, based on EFSM programming model, is implemented as a means of load balancing mechanism. A file system supporting the task migration mechanism is also designed and developed.
This paper describes ViSMI, a software distributed shared memory system for cluster systems connected via InfiniBand. ViSMI implements a kind of home-based lazy release consistency protocol, which uses a multiple-writ...
详细信息
ISBN:
(纸本)0769522424
This paper describes ViSMI, a software distributed shared memory system for cluster systems connected via InfiniBand. ViSMI implements a kind of home-based lazy release consistency protocol, which uses a multiple-writer coherence scheme to alleviate the traffic introduced by false sharing. For further performance gain, InfiniBand features and optimized page invalidation mechanisms are applied in order to reduce synchronization overhead. First experimental results show that ViSMI introduces good performance comparable to similar software DSMs.
PARallel and distributedcomputing (PArdISC) is a parallel programming environment which provides the needed transparency as a scalable OpenFrame computing Model. It can be used to configure the network as a loosely c...
详细信息
PARallel and distributedcomputing (PArdISC) is a parallel programming environment which provides the needed transparency as a scalable OpenFrame computing Model. It can be used to configure the network as a loosely coupled parallel machine, processor pool architecture, and distributedcomputing environment with Logical Network Connectivity. The software architecture of PArdISC is made up of configuration servers, client processes, processor pool servers and process communication interface.
The proceedings contains 74 papers from the 3rdinternational Conference on highperformancecomputing. Topics discussed include: genetic algorithms, optical computers, parallel webservers, direct circuit simulation, ...
详细信息
The proceedings contains 74 papers from the 3rdinternational Conference on highperformancecomputing. Topics discussed include: genetic algorithms, optical computers, parallel webservers, direct circuit simulation, character recognition, multimedia servers, the RAID configuration tool, concurrent call/return communication, parallel processing systems, real-time distributed operating systems, database systems, mobile computers, binary trees, close wormhole networks, program analysis, data-flow analysis, reduced instruction set computing (RISC), and edge extraction.
The proceedings contain 6 papers. The topics discussed include: performance engineering: a must for petascale and beyond;comparison of lock thrashing avoidance methods and its performance implications for lock design;...
ISBN:
(纸本)9781450307031
The proceedings contain 6 papers. The topics discussed include: performance engineering: a must for petascale and beyond;comparison of lock thrashing avoidance methods and its performance implications for lock design;making a case for distributed file systems at exascale;visual analysis of I/O system behavior for high-end computing;multi-scale analysis of large distributedcomputing systems;and an analysis of social gaming networks in online and face to face bridge communities.
Researchers in the biological and health industries need powerful and stable systems for their work. These systems must be dependable, fault-tolerant, highly available and easy to use. To cope with these demands we pr...
详细信息
ISBN:
(纸本)0769519199
Researchers in the biological and health industries need powerful and stable systems for their work. These systems must be dependable, fault-tolerant, highly available and easy to use. To cope with these demands we propose the use of computational and data clusters in a fail-over configuration combined with the grid technology and job scheduling. Our infrastructure has been deployed successfully for running time-critical applications in commercial environments. We also present experimental results from this pilot implementation that demonstrate the viability of our approach.
暂无评论