distributedsystems are expected to support mobile computations executed over a computer network of fixed and mobile hosts. The authors examine the requirements for structuring such mobile computations that access sha...
详细信息
Availability of high-speed networks and increasingly powerful workstations has spawned new opportunities for research in distributedsystems. distributed collaborative systems using multiple media exploit high-speed, ...
详细信息
Mobile computing is a rapidly emerging trend in distributed computing. The new mobile computing environment presents many challenges due to the mobile nature of the hosts. The authors present some fault-tolerant data ...
详细信息
In parallel and distributed simulations, it is sometimes desirable that the applications time-stamped events and/or the simulator's time-management control messages be exchanged over a combination of reliable and ...
详细信息
ISBN:
(纸本)0769511058
In parallel and distributed simulations, it is sometimes desirable that the applications time-stamped events and/or the simulator's time-management control messages be exchanged over a combination of reliable and unreliable network channels. A challenge in developing infrastructure for such simulations is to correctly compute simulation time advances despite the loss of some simulation events and/or control messages. Presented here are algorithms for synchronization in distributed simulations performed directly over best-effort network transport. The algorithms are presented in a sequence of progressive refinement, starting with all reliable transport and finishing with combinations of relable and unreliable transports for both time-stamped events and time management messages. performance results from a preliminary implementation of these algorithms are also presented. To our knowledge, this is the first work to solve asynchronous time synchronization performed directly over unreliable network transport.
The authors present a new token based distributed mutual exclusion algorithm for a distributed computer system of N sites. The proposed algorithm is based on timestamps and the theory of finite projective planes. It a...
详细信息
作者：
Wirz, B.Nett, E.
Schloß Birlinghoven St. Augustin53757 Germany
Logs are an important facility for fault-tolerant distributedsystems since they allow to reliably store information that is needed to provide a global consistent system state also in the presence of failures. The aut...
详细信息
To efficiently perform collective communications in current high-performance computing systems is a time-consuming task. With future exascale systems, this communication time will be increased further. However, global...
详细信息
ISBN:
(纸本)9781728101767
To efficiently perform collective communications in current high-performance computing systems is a time-consuming task. With future exascale systems, this communication time will be increased further. However, global information is frequently required in various physical models. By exploiting domain knowledge of the model behaviors globally needed information can be distributed more efficiently, using only peer-to-peer communication which spread the information to all processes asynchronous during multiple communication steps. In this article, we introduce a multi-hop based Manhattan Street Network (MSN) for global information exchange and show the conditions under which a local neighbor exchange is sufficient for exchanging distributed information. Besides the MSN, in various models, global information is only needed in a spatially limited region inside the simulation domain. Therefore, a second network is introduced, the local exchange network, to exploit this spatial assumption. Both non-collective global exchange networks are implemented in the massively parallel NAStJA framework. Based on two models, a phase-field model for droplet simulations and the cellular Potts model for biological tissue simulations, we exemplary demonstrate the wide applicability of these networks. Scaling tests of the networks demonstrate a nearly ideal scaling behavior with an efficiency of over 90%. Theoretical prediction of the communication time on future exascale systems shows an enormous advantage of the presented exchange methods of O(1) by exploiting the domain knowledge.
The authors describe a new class of name and resource management facility - a trading service - which allows users of a heterogeneous large distributed system to share resources and services (resources) which are not ...
详细信息
The authors introduce an efficient indexing organization which provides for parallel retrievals. Given a p-processor massively parallel computer, the organization allows a maximum of p retrievals to be performed concu...
parallel database systems are suitable for use in applications with high capacity and high performance and availability requirements. The trend in such systems is to provide efficient online capability for performing ...
详细信息
暂无评论