Advances in networking and storage technologies have made it possible to build multimedia on-demand servers that provide service similar to those of neighborhood videotape rental stores. We propose a specialized high ...
详细信息
Advances in networking and storage technologies have made it possible to build multimedia on-demand servers that provide service similar to those of neighborhood videotape rental stores. We propose a specialized high performance programmable multiprocessor architecture which allows continuous playback of mediastreams to large number of clients, a critical factor in multimedia servers. this architecture supports dynamic phase shifts in media distribution which is particularly useful in a distributed environment. We introduce algorithms for maximum concurrency extraction and to avoid task migration in the presence of dynamically changing demand. We show that our solution performs an order of magnitude better than other known solutions.< >
We propose a service partition model as a design methodology of distributed applications, especially in interactive distributed applications. We developed a set of communication mechanisms for linking each module of i...
详细信息
ISBN:
(纸本)0780318625
We propose a service partition model as a design methodology of distributed applications, especially in interactive distributed applications. We developed a set of communication mechanisms for linking each module of interactive distributed applications. Mapping an interactive distributed application into service partition model is much easier than mapping it into a client server model. Implementing distributed applications withthe mechanisms we supported for the service partition programming model is relatively easy.< >
this paper presents a method of computingthe sensitivity of electromagnetic quantities (e.g. induced voltage, force, field uniformity) to perturbations in sources and material parameters. the method applies to system...
详细信息
this paper presents a method of computingthe sensitivity of electromagnetic quantities (e.g. induced voltage, force, field uniformity) to perturbations in sources and material parameters. the method applies to systems with linear materials and time harmonic source currents;static problems are included as a special case. Sensitivity is equivalent to the derivative with respect to the distributed system parameters. therefore, withthe expressions for the sensitivity presented in this paper, first order optimization methods can be applied to optimization problems in electromagnetics. Applications are in the area of inverse problems, including electromagnetic device design, device optimization, and imaging.
In the basic version of the distributed mutual exclusion problem, only one resource is under contention. If there are many types and instances of shared resources in a distributed system, a node may want to access the...
详细信息
In the basic version of the distributed mutual exclusion problem, only one resource is under contention. If there are many types and instances of shared resources in a distributed system, a node may want to access the first available resource from any given subset of these resources. In this paper, we present a static tree based solution to this problem. the number of messages generated per critical section execution can be from 0 to 2(N-1), where N is the number of nodes in the system. A significant advantage of the presented algorithm is that the upper bound on the number of messages generated per critical section execution, namely 2(N-1), is independent of the number of resources covered by the algorithm.< >
the parallelizing compiler divides a given program into several processes which can be executed simultaneously in a distributed computer environment. However, a program has global variables generally. therefore, the p...
详细信息
the parallelizing compiler divides a given program into several processes which can be executed simultaneously in a distributed computer environment. However, a program has global variables generally. therefore, the processes have to share global variables since they are common in the program. In order to manage global variables correctly, the processes need to possess the broadcasting facility for data sharing and the synchronization facility for execution order control among the processes. We propose execution order control methods among the processes for sharing global variables in the distributed computer environment.< >
this paper presents an adaptive dynamic file migration algorithm in distributed systems on the CSMA/CD local area network. In our approach, we employed learning automata with several internal states and a file is migr...
详细信息
ISBN:
(纸本)0780318625
this paper presents an adaptive dynamic file migration algorithm in distributed systems on the CSMA/CD local area network. In our approach, we employed learning automata with several internal states and a file is migrated to the destination processor when it is in the right boundary state; and file migration with low computational complexity is done in distributed manner. Analytical and simulation results show that our algorithm is very well-suited for local area network environments.< >
As part of a research project which is underway, this paper describes the setting for a distributed multimedia platform. We envision that the information systems of the future will be based upon distributed systems an...
详细信息
As part of a research project which is underway, this paper describes the setting for a distributed multimedia platform. We envision that the information systems of the future will be based upon distributed systems and improved ways to store, process, retrieve and transfer information by utilizing text, images, graphics, animation, speech, video and sound. In this paper, we describe the environment of interest, we discuss the overhead required by such a system as well as systems which are currently available. Finally, we provide the description of our effort in this setting.< >
We demonstrate that it is possible to have a file facility architecture where separate interfaces are provided for the basic file service and transaction oriented file service. thus, depending upon the requirement a u...
详细信息
We demonstrate that it is possible to have a file facility architecture where separate interfaces are provided for the basic file service and transaction oriented file service. thus, depending upon the requirement a user can perform I/O on a file either using the semantics of the basic file service or the transaction oriented file service. the concept of stable storage is exploited to make our design very reliable. To achieve the best possible performance we have successfully established a very fruitful and yet new relationship among the storage policy of file data and the methods to make changes in the intentions list permanent to commit a transaction.< >
the commercial hypercube-based machine is a good physical architecture for embedding various logical topologies because of its logarithmic diameter and high bandwidth. through such embedding, the algorithms originally...
详细信息
ISBN:
(纸本)0780318625
the commercial hypercube-based machine is a good physical architecture for embedding various logical topologies because of its logarithmic diameter and high bandwidth. through such embedding, the algorithms originally developed for logical topologies can be directly executed on hypercubes. We present a distributed scheme for embedding larger binary trees into smaller hypercubes. From a practical point of view, this is a very important issue when a hypercube of limited size is available. the desired embeddings can be constructed in parallel with time complexity of O (log N), where N is the total number of tree nodes. It provides a uniform many-to-one embedding with minimum overhead.< >
Computer-supported cooperative work (CSCW) has emerged as an important area of research, where both commercial and academic interest is growing. this field draws from developments in distributedcomputing, communicati...
详细信息
Computer-supported cooperative work (CSCW) has emerged as an important area of research, where both commercial and academic interest is growing. this field draws from developments in distributedcomputing, communications, man-machine interactions, artificial intelligence and sociology. In CSCW applications, the emphasis is on user-machine-user interaction rather than on user-machine interaction. In the recent past, a few concurrent software engineering tools and environments have been reported, but most of these have been sculpted from the ground up. In this paper, we propose a concurrency enabler architecture, using which a nonconcurrent application can be made concurrent. Using this approach, we have implemented a concurrent editor, a debugger, an inspection tool and a conferencing and execution environment. Our experiences in using these concurrent tools are also reported.< >
暂无评论