Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 20 of 43 for “"InfiniBand"”.

  1. Measurement of PVFS2 performance on InfiniBand

    <p>InfiniBand is becoming increasingly popular as a fast interconnect technology between servers and storage. It has far better price/performance ratio compared to both Gigabit Ethernet and 10 Gigabit Ethernet, and hence is being increasingly used for high-performance computing applications. PVFS2, …

    iastate Repository record for Measurement of PVFS2 performance on InfiniBand (opens in a new tab)

  2. High Performance Computing using Infiniband-based clusters

    L'abstract è presente nell'allegato / the abstract is in the attachment

    poli-torino Repository record for High Performance Computing using Infiniband-based clusters (opens in a new tab)

  3. Overlapping Computation and Communication through Offloading in MPI over InfiniBand

    … using CORE-Direct offloading support in Mellanox InfiniBand adapters. We present an evaluation of the designs using microbenchmarks and an application kernel that shows that offloaded non-blocking communication operations can provide latency that is comparable to that of their blocking …

    queens Repository record for Overlapping Computation and Communication through Offloading in MPI over InfiniBand (opens in a new tab)

  4. Sistema Aldeia : programação paralela e distribuída em Java sobre Infiniband e DECK

    … uma tentativa de padronização a nova tecnologia Infiniband. Ela proporciona uma baixa latência de comunicação e uma alta vazão de dados, além de uma série de vantagens implementadas diretamente no hardware. É neste contexto que se desenvolve o presente trabalho de dissertação de mestrado. O seu …

    brazil-ufrgs Repository record for Sistema Aldeia : programação paralela e distribuída em Java sobre Infiniband e DECK (opens in a new tab)

  5. Performance analysis and improvement of InfiniBand networks. Modelling and effective Quality-of-Service mechanisms for interconnection networks in cluster computing systems.

    The InfiniBand Architecture (IBA) network has been proposed as a new industrial standard with high-bandwidth and low-latency suitable for constructing high-performance interconnected cluster computing systems. This architecture replaces the traditional bus-based interconnection with a switch-based …

    bradford Repository record for Performance analysis and improvement of InfiniBand networks. Modelling and effective Quality-of-Service mechanisms for interconnection networks in cluster computing systems. (opens in a new tab)

  6. Design and Evaluation of Efficient Collective Communications on Modern Interconnects and Multi-core Clusters

    … that utilize the modern features of Quadrics and InfiniBand interconnects as well as the availability of multiple cores on emerging clusters. To overcome bandwidth limitations and to enhance fault tolerance, using multiple independent networks known as multi-rail networks is very promising. …

    queens Repository record for Design and Evaluation of Efficient Collective Communications on Modern Interconnects and Multi-core Clusters (opens in a new tab)

  7. Region-based techniques for modeling and enhancing cluster OpenMP performance

    … on two clusters connected with Ethernet and InfiniBand interconnects. The performance data is quantitatively analyzed and modeled. MCBENCH is used to evaluate the impact of ReP techniques on memory consistency cost. The evaluation results demonstrate that, on average, CLOMP spends 75% and 55% …

    anu Repository record for Region-based techniques for modeling and enhancing cluster OpenMP performance (opens in a new tab)

  8. Improving Message-Passing Performance and Scalability in High-Performance Clusters

    … MPI libraries over modern cluster interconnects: InfiniBand, iWARP Ethernet, and Myrinet. Using novel techniques, I assess characteristics such as overlap and communication progress ability, buffer reuse effect on latency, and multiple-connection scalability. The outcome highlights some of the …

    queens Repository record for Improving Message-Passing Performance and Scalability in High-Performance Clusters (opens in a new tab)

  9. Improving High Performance Networking Technologies for Data Center Clusters

    … the SDP level. Datagram offloading for IP over InfiniBand (IPoIB) is found to work well. In the first work of its kind, hybrid high-speed/Ethernet networks are shown to resolve the issues of SDP underperformance and demonstrate the potential for hybrid high-speed networking local area Remote …

    queens Repository record for Improving High Performance Networking Technologies for Data Center Clusters (opens in a new tab)

  10. Virtualized resource management in high performance fabric clusters

    … on emerging fabrics like 40+Gbps Ethernet and InfiniBand/RoCE with support for RDMA, VMM-bypass and hardware-level virtualization (SR-IOV). However, the benefits provided by these technology advances are offset by several management constraints: (i) the inability of the hypervisor to monitor …

    gatech Repository record for Virtualized resource management in high performance fabric clusters (opens in a new tab)

  11. Analysis and performance of a UPC implementation of a parallel longest common subsequence algorithm

    … in Unified Parallel C that runs on an Infiniband cluster. This performance is partly facilitated by exploiting the software cache of the MuPC runtime system. In addition, this implementation achieved speedup without bulk memory copy operations and the associated programming complexity …

    mtu Repository record for Analysis and performance of a UPC implementation of a parallel longest common subsequence algorithm (opens in a new tab)

  12. Enhancing HPC on Virtual Systems in Clouds through Optimizing Virtual Overlay Networks

    … of the overlay network on 10 Gbps Ethernet and InfiniBand interconnects, resulting in near-native performance for a wide range of microbenchmarks and MPI application benchmarks.

    unm Repository record for Enhancing HPC on Virtual Systems in Clouds through Optimizing Virtual Overlay Networks (opens in a new tab)

  13. Evaluation of mpi4py for Natural Language Processing Scenarios

    … with native C-based MPI benchmarks using an InfiniBand and a Gigabit Ethernet network interconnect. The results show that in many instances communication performance of the Python benchmarks was on par with their C-based counterparts. In the second part of the thesis, a few application …

    houston Repository record for Evaluation of mpi4py for Natural Language Processing Scenarios (opens in a new tab)

  14. Influence of Runtime Parameters on the Performance of Parallel Applications

    … this work were point-to-point parameters of the InfiniBand network, shared memory commmunication parameters, and algorithms used for collective communication operations. Two NAS Parallel Benchmarks were used to evaluate the overall performance gain in each scenario. The results indicate that by …

    houston Repository record for Influence of Runtime Parameters on the Performance of Parallel Applications (opens in a new tab)

  15. Hybrid design of MPI over SCTP

    … Direct Memory Access (RDMA) devices such as InfiniBand and the recently standardized Internet Wide Area RDMA Protocol (iWARP). The API itself provides both the point-to-point and remote memory access (RMA) operations to the application. However, it is often implemented based on one kind of …

    ubc Repository record for Hybrid design of MPI over SCTP (opens in a new tab) Download PDF for Hybrid design of MPI over SCTP (opens in a new tab)

  16. Performance analysis of the uDAPL library using a unified benchmarking framework

    … Currently, uDAPL is widely used on InfiniBand (IB) and virtual interface adapter (VIA) interconnects. This research evaluates the performance of uDAPL library on a 4X IB interconnect, and identifies the overheads imposed by the library on the underlying hardware. Micro-benchmark …

    ttu Repository record for Performance analysis of the uDAPL library using a unified benchmarking framework (opens in a new tab)

  17. Automatic Energy Saving Schemes for Parallel Applications

    … Additionally, network interconnect, such as Infiniband, may be exploited to</p> <p>maximize energy savings while the application performance loss and frequency switching overheads must be carefully balanced.</p> <p>This work first studies two important collective communication operations, …

    iastate Repository record for Automatic Energy Saving Schemes for Parallel Applications (opens in a new tab)

Page 1 of 3