The Quadrics network (QsNet): high-performance clustering technology
نویسندگان
چکیده
The Quadrics interconnection network (QsNet) contributes two novel innovations to the field of highperformance interconnects: (1) integration of the virtualaddress spaces of individual nodes into a single, global, virtual-address space and (2) network fault tolerance via link-level and end-to-end protocols that can detect faults and automatically re-transmit packets. QsNet achieves these feats by extending the native operating system in the nodes with a network operating system and specialized hardware support in the network interface. As these and other important features of QsNet can be found in the InfiniBand specification, QsNet can be viewed as a precursor to InfiniBand. In this paper, we present an initial performance evaluation of QsNet. We first describe the main hardware and software features of QsNet, followed by the results of benchmarks that we ran on our experimental, Intel-based, Linux cluster built around QsNet. Our initial analysis indicates that QsNet performs remarkably well, e.g., user-level latency under 2 s and bandwidth over 300 MB/s.
منابع مشابه
Network Performance in High Performance Linux Clusters
Linux-based clusters have become more prevalent as a foundation for High Performance Computing (HPC) systems. With a better understanding of network performance in these environments, we can optimize configurations and develop better management and administration policies to improve operations. To assist in this process, we developed a network measurement tool to measure UDP, TCP and MPI commun...
متن کاملA New MPI Implementation for Cray SHMEM
Previous implementations of MPICH using the Cray SHMEM interface existed for the Cray T3 series of machines, but these implementations were abandoned after the T3 series was discontinued. However, support for the Cray SHMEM programming interface has continued on other platforms, including commodity clusters built using the Quadrics QsNet network. In this paper, we describe a design for MPI that...
متن کاملThe Quadrics Network Extends the Native Operating System in Processing Nodes with a Network Operating System and Specialized Hardware Support in the Network
The interconnection network and its associated software libraries are critical components for high-performance cluster computers and supercomputers, Web-server farms, and network-attached storage. Such components will greatly impact the design, architecture, and use of future systems. Key solutions in high-speed interconnects include Gigabit Ethernet, GigaNet, the Scalable Coherent Interface (S...
متن کاملDesign and Implementation of Open MPI over QsNet/Elan4
Open MPI is a project recently initiated to provide a fault-tolerant, multi-network capable, and productionquality implementation of MPI-2 [20] interface based on experiences gained from FT-MPI [8], LA-MPI [10], LAM/MPI [28], and MVAPICH [23] projects. Its initial communication architecture is layered on top of TCP/IP. In this paper, we have designed and implemented Open MPI point-to-point laye...
متن کاملEfficient RDMA-based Multi-port Collectives on Multi-rail QsNet Clusters
Many scientific applications use MPI collective communications intensively. Therefore, efficient and scalable implementation of collective operations is critical to the performance of such applications running on clusters. Quadrics QsNet is a high-performance interconnect for clusters that implements some collectives at the Elan level. These collectives are directly used by their corresponding ...
متن کامل