This paper studies wavelength assignments in a class of all-optical networks without wavelength converters topologized as star-ring: some sub-rings connected by a backbone ring. In the static model, we give an upper b...
详细信息
Although the networking world is still dominated by IP over Ethernet, these protocol suites were not originally designed to work on modern Gigabit or Gigabyte networks. The major problem facing IP at 1 Gb/s and 10 Gb/...
详细信息
high-performance networks require sophisticated management systems to identify sources of bottlenecks and detect faults. At the same time, the impact of network queries on the latency and bandwidth available to the ap...
详细信息
One of the most important problem in computational biology is the tree editing problem which is to determine the edit distance between two rooted labeled trees. It has been shown to have significant applications in bo...
详细信息
Gossip protocols provide a means by which failures can be detected in large, distributed systems in an asynchronous manner without the limits associated with reliable multicasting for group communications. However, in...
Gossip protocols provide a means by which failures can be detected in large, distributed systems in an asynchronous manner without the limits associated with reliable multicasting for group communications. However, in order to be effective with application recovery and reconfiguration, these protocols require mechanisms by which failures can be detected with system-wide consensus in a scalable fashion. This paper presents three new gossip-style protocols supported by a novel algorithm to achieve consensus in scalable, heterogeneous clusters. The round-robin protocol improves on basic randomized gossiping by distributing gossip messages in a deterministic order that optimizes bandwidth consumption. Redundant gossiping is completely eliminated in the binary round-robin protocol, and the round-robin with sequence check protocol is a useful extension that yields efficient detection times without the need for system-specific optimization. The distributed consensus algorithm works with these gossip protocols to achieve agreement among the operable nodes in the cluster on the state of the system featuring either a flat or a layered design. The various protocols are simulated and evaluated in terms of consensus time and scalability using a high-fidelity, fault-injection model for distributed systems comprised of clusters of workstations connected by high-performance networks.
This paper addresses the problem of multicasting and broadcasting in undirected wavelength-division multiplexing (WDM) networks. Given an undirected network G=(VE) with a source nodes and a set of destination nodes D,...
详细信息
This paper addresses the problem of multicasting and broadcasting in undirected wavelength-division multiplexing (WDM) networks. Given an undirected network G=(VE) with a source nodes and a set of destination nodes D, /spl Lambda/ is the set of wavelength that can be used in the network. Associated with every edge e, there is a set of available wavelengths on it. The multicast problem is to find a tree rooted at s including all nodes in D such that the cost of the tree is minimum in terms of the cost of wavelength conversion at nodes and the cost of using wavelength on edges. This paper proves that the multicast problem is NP-complete and can not be approximated within a constant factor, unless P=NP. Then we construct an auxiliary graph for the original WDM networks and reduce the multicast problem to a group Steiner tree problem on the auxiliary graph. Employing the known algorithm for the group Steiner tree problem, we derive an algorithm for the problem, which delivers a solution within O(log/sup 2/ (nk)loglog(nk)logp) times the optimum.
This paper studies wavelength assignments in a class of all-optical networks without wavelength converters topologized as star-ring: some sub-rings connected by a backbone ring. In the static model, we give an upper b...
详细信息
ISBN:
(纸本)0780370104
This paper studies wavelength assignments in a class of all-optical networks without wavelength converters topologized as star-ring: some sub-rings connected by a backbone ring. In the static model, we give an upper bound 5L/2 and a lower bound 9L/4 of the necessary wavelengths to accommodate any requests with load L. These become (3L/2)/spl lceil/log/sub 2/ (N+(/spl Sigma//sub i=1//sup n/)m/sub i/)/spl rceil/+max/sub 1/spl les/i/spl les/n/ (L/spl lceil/log/sub 2/ m/sub i//spl rceil/)+2L and 0.5L(/spl lceil/log/sub 2/ (N+max/sub 1/spl les/i/spl les/n/ (m/sub i/))/spl rceil/+1) respectively in the dynamic model, where N is the number of nodes in the backbone ring and m/sub i/ is the number of nodes in the ith sub-ring (1/spl les/i/spl les/n).
This paper studies wavelength assignments in a class of all-optical networks without wavelength converters topologized as star-ring:some sub-rings connected by a backbone *** the static model,we give an upper bound 5L...
详细信息
This paper studies wavelength assignments in a class of all-optical networks without wavelength converters topologized as star-ring:some sub-rings connected by a backbone *** the static model,we give an upper bound 5L/2 and a lower bound 9L/4 of the necessary wavelengths to accommodate any requests with load *** become 3L/2[log(N+∑i=1nm]+max+2L 0.5L([log(N + max（m）]+1) respectively in the dynamic model,where N is the number of nodes in the backbone ring and mt is the number of nodes in the ith sub-ring .
Problem solving environments are an attractive approach to the integration of calculation and management tools for various scientific and engineering applications. These applications often require highperformance com...
详细信息
ISBN:
(纸本)0769509819
Problem solving environments are an attractive approach to the integration of calculation and management tools for various scientific and engineering applications. These applications often require highperformancecomputing components in order to be computationally feasible. It is therefore a challenge to construct integration technology, suitable for problem solving environments, that allows both flexibility as well as the embedding of parallel and highperformancecomputing systems. Our DISCWorld system is designed to meet these needs and provides a Java-based middleware to integrate component applications across wide-area networks. Key features of our design are the abilities to: access remotely stored data; compose complex processing requests either graphically or through a scripting language; execute components on heterogeneous and remote platforms; reconfigure task subgraphs to run across multiple servers. Operators in task graphs can be slow (but portable) pure Java implementations or wrappers to fast (platform specific) supercomputer implementations.
暂无评论