Guest editors' introduction: challenges in designing fault-tolerant routing in networks

Abstract

+ I INRE arc several means nf intercomiecting the compo-T nents of a computer system. Components can be connccted through direct or indirect connections. In a direct network, channels directly connect each processor. Neighboring nodes communicate with each other by scnding messages across the channcls. In an indirect network, processors are cnnnccted through switches and/or buses. Depending on how components are connected, we have bus-based networks, router-based networks, switcli-bascd networks, and cluster-based (hybrid of the other thrce networks) nctworks. Nctworks can also be classified bascd on topologies, which cau be either r e p l n r or irregulnr. Anothcr classification is based on applications-there are local arca networks (LANs), networks of workstations (NOWs), multiprocessors, and mobilc networks.In a computer network, the efficiency of a routing process is critical to the system performance. A routing process deals with moving data between processors in a given network. When a routing process involves oiic source and onc destination, i t is called unicnst or poirit-lopoint communication. When such a proccss involves more than one source and/or one destination, it is called a collective comrnlrnicntion process. Examples of collective communication include iirulticnst (one-to-many, one source and many destinations), brundcnst (one-to-all), and gossip (all-to-all).Morc rcceutly, thc lower cost NOWs have bccome accepted as a replacement for expensive supcrcomputers. Routing in NOWs is done primarily using a large LAN. Thus, routing in such networks has bccome incrcasingly critical for high performance computing applications. Thc router design and routing techniques are altogether different for NOWs and LANs, as cnmparcd to conventional multiprocessors. Support for PVM-type messagc communication mechanisms and their fffcctive use has becomc increasingly important. Such schemcs differ greatly in mobile networks, because the signals have to pass through the noisy atmosphcrc. Also, the conversion from medium access control in the air to ATM-type protocol for the backbone network has bccome a nccessity in the communication world.As thc iiumber of processors in a network incrcases, the probability of proccssor failurc also incrcases. Thcrefore, it is important to design a routing process with fault Tor iiJiwirotion oii obliii,iiiig repriiih o j this nrficlr, pi ipnsBcurnpiifer-.ory, nwd rc/Cre,icc 1EEECS Log Niiinbrr