No abstract

This paper proposes multidestination message passing on wormhole k-ary n-cube networks using a new base-routing-conformed-path (BRCP) model. This model allows both unicast (single-destination) and multidestination messages to co-exist in a given network without leading to deadlock. The model is illustrated with several common routing schemes (deterministic, as well as adaptive), and the associated deadlock-freedom properties are analyzed. Using this model a set of new algorithms for popular collective communication operations, broadcast and multicast, are proposed and evaluated. It is shown that the proposed algorithms can considerably reduce the latency of these operations compared to the Umesh (unicast-based multicast) 22] and the Hamiltonian-Path-based 21] schemes. A very interesting result that is presented shows that a multicast can be implemented with reduced or near-constant latency as the number of processors participating in the multicast increases beyond a certain number. It is also shown that the BRCP model can take advantage of adaptivity in routing schemes to further reduce the latency of these operations. The multidestination mechanism and the BRCP model establish a new foundation to provide fast and scalable collective communication support on wormhole-routed systems.

IntroductionTraditionally, operations with memory on other nodes (remote memory) in cluster environments interconnected with technologies like Gigabit Ethernet have been expensive with latencies several magnitudes slower than local memory accesses. Modern RDMA capable networks such as InfiniBand and Quadrics provide low latency of a few microseconds and high bandwidth of up to 10 Gbps. This has significantly reduced the latency gap between access to local memory and remote memory in modern clusters. Remote idle memory can be exploited to reduce the memory pressure on individual nodes. This is akin to adding an additional level in the memory hierarchy between local memory and the disk, with potentially dramatic performance improvements especially for memory intensive applications. In this paper, we take on the challenge to design a remote paging system for remote memory utilization in InfiniBand clusters. We present the design and implementation of a high performance networking block device (HPBD) over InfiniBandfabric, which serves as a swap device ofkernel Virtual Memory (VM) systemfor efficientpage transfer to/from remote memory servers. Our experiments show that using HPBD, quick sort performs only 1.45 times slower than local memory system, and up to 21 times faster than local disk. And our design is completely transparent to user applications. To the best ofour knowledge, it is thefirst work of a remote pager design using InfiniBandfor remote memory utilization.

A reliable multicast algorithm, called RMA, for mobile ad hoc networks is presented that is based on a new cost criterion, called link lifetime, for determining the optimal path between a pair of nodes. The algorithm has the characteristics of using an undirected graph for its routing operations rather than a fixed structure like a tree or a mesh. Previously proposed routing metrics for mobile ad hoc networks were designed for use in wired environments, where link stability is not a concern. We propose a new metric, called the lifetime, which is more appropriate for mobile ad hoc networks. The lifetime metric is dependent on the predicted future life of the link under consideration. We developed a simulator for the mobile ad hoc networks, which is portable and scalable to a large number of nodes. Using the simulator, we carried out a simulation study to analyze the effectiveness of the routing metrics and the performance of the proposed reliable multicast algorithm. The simulation results show that the lifetime metric helps achieve better performance in mobile ad hoc environments than the hop count metric.

No abstract