distributed Object computing (DOC) is an increasingly important computing paradigm exploiting the benefits of object-oriented technology and using the internet and its communications infrastructure as a vehicle for th...
详细信息
ISBN:
(纸本)0769511651
distributed Object computing (DOC) is an increasingly important computing paradigm exploiting the benefits of object-oriented technology and using the internet and its communications infrastructure as a vehicle for the delivery of a wide range of sophisticated value-added distributed services. ne integration of the Common Object Request Broker Architecture (CORBA) and object-oriented programming languages facilitates the introduction and deployment of such applications in a platform-independent open distributed environment. In this paper, we introduce both CORBA and Java as tools to facilitate the development of distributed object computing-based applications, and we present a prototype electronic commerce application using the DOC paradigm.
Recent advances in miniaturization and low-cost, low-power design have led to active research in large-scale, highly distributedsystems of small, wireless, low-power unattended sensors and actuators. We explore the u...
详细信息
ISBN:
(纸本)0769510779
Recent advances in miniaturization and low-cost, low-power design have led to active research in large-scale, highly distributedsystems of small, wireless, low-power unattended sensors and actuators. We explore the use of Random, Ephemeral TRansaction Identifiers (RETRI) in such systems, and contrast it with the typical design philosophy of using static identifiers in roles such as node addressing or efficient data naming. Instead of using statically assigned identifiers that are guaranteed to be unique, nodes randomly select probabilistically, unique identifiers for each new transaction. We show how this randomized scheme can significantly, improve the system's energy efficiency in contexts where that efficiency is paramount, such as energy-constrained wireless sensor networks. Benefits are realized if the typical data size is small compared to the size of an identifier and the number of transactions seen by an individual node is small compared to the number of nodes that exist in the entire system. Our scheme is designed to scale well: identifier sizes grow with a system's density, not its overall size. We quantify these benefits using an analytic model that predicts our scheme's efficiency. We also describe an implementation as applied to packet fragmentation and an experiment that validates our model.
Debugging and testing is a larger part of the effort spent in a software development cycle. Debugging a program is time consuming and is a continuous cycle of code modification and testing. The task of debugging depen...
详细信息
ISBN:
(纸本)0769511651
Debugging and testing is a larger part of the effort spent in a software development cycle. Debugging a program is time consuming and is a continuous cycle of code modification and testing. The task of debugging depends on the environment, the language, the operating system;the problem;and more interestingly even the individual program. In this paper, we present a detailed implementation of software architecture for debugging distributed programs. The debugger based on the proposed architecture supports debugging based on a mechanism for controlling multiple processes, presenting the process states, grouping of processes for ease in controlling the distributed programs developed on PVM based computing paradigm.
Partial-order data structures used in distributed-system observation tools typically use vector timestamps to efficiently determine event precedence. Unfortunately, all current dynamic vector-timestamp algorithms eith...
详细信息
ISBN:
(纸本)0769510779
Partial-order data structures used in distributed-system observation tools typically use vector timestamps to efficiently determine event precedence. Unfortunately, all current dynamic vector-timestamp algorithms either require a vector of size equal to the number of processes in the computation or require a graph search operation to determine event precedence. This fundamentally limits the scalability of such observation systems. In this paper we present an algorithm for hierarchical, clustered vector timestamps. We present results for a variety of computation environments that demonstrate such timestamps can reduce space consumption by more than an order-of-magnitude over Fidge/Mattern timestamps while still providing acceptable time bounds for computing timestamps and determining event precedence.
In distributedsystems, a real-time task has several subtasks which need to be executed at different nodes. Some of these subtasks can be executed in parallel on different nodes without violating their precedence rela...
详细信息
ISBN:
(纸本)0769513816
In distributedsystems, a real-time task has several subtasks which need to be executed at different nodes. Some of these subtasks can be executed in parallel on different nodes without violating their precedence relationships, if any, among them. To better exploit the parallelism, it becomes necessary to assign separate deadlines to subtasks and schedule them independently. We use three subtask deadline assignment policies which we have introduced earlier [4], [7] to develop a bidding-based fault-tolerant scheduling algorithm for distributed real-time systems. A local scheduler which resides on each node, tries to determine a schedule for each subtask according to the primary-backup approach. In this paper we discuss the algorithm and present the results of simulation studies conducted to establish the efficacy of our algorithm.
Monitoring of global predicates is a fundamental problem in asynchronous distributedsystems. This problem arises in various contexts such as design, testing and debugging, and fault-tolerance of distributed programs....
详细信息
ISBN:
(纸本)0769510779
Monitoring of global predicates is a fundamental problem in asynchronous distributedsystems. This problem arises in various contexts such as design, testing and debugging, and fault-tolerance of distributed programs. In this paper, we establish that the problem of determining whether there exists a consistent cut of a computation that satisfies a predicate in k-CNF, k greater than or equal to 2, in which no two clauses contain variables from the same process is NP-complete in general. A polynomial-time algorithm to find the consistent cut, if it exists, that satisfies the predicate for special cases is provided. We also give algorithms albeit exponential that can be used to achieve an exponential reduction in time over existing techniques for solving the general version. Furthermore, we present an algorithm to determine whether there exists a consistent cut of a computation for which the sum x(1) + x(2) + . . . + x(n) exactly equals some constant k, where each x(i) is an integer variable on process p(i) such that it is incremented or decremented by at most one at each step. As a corollary, any symmetric global predicate on boolean variables such as absence of simple majority and exclusive-or of local predicates can now be detected. Additionally, the problem is proved to be NP-complete if each x(i) can be changed by an arbitrary amount at each step. Our results solve the previously open problems in predicate detection proposed in [7] and bridge the wide gap between the known tractability and intractability results that existed until now.
A compute farm is a pool of clustered workstations to provide high performance computing services for CPU-intensive, memory-intensive, and I/O active jobs in a batch merle. Existing loan sharing schemes with memory co...
详细信息
ISBN:
(纸本)0769510779
A compute farm is a pool of clustered workstations to provide high performance computing services for CPU-intensive, memory-intensive, and I/O active jobs in a batch merle. Existing loan sharing schemes with memory considerations assume jobs' memory demand sizes are known in advance or predictable based on users' hints. This assumption call greatly simplify the designs and implementations of load sharing schemes, but is not desirable in practice. In order to address this concern, we present three new results and contributions in this study (1) Conducting Linux kernel instrumentation, we have collected different types of workload execution traces to quantitatively characterize job interactions, and modeled page fault behavior as a function of the overloaded memory sizes and the amount of jobs' I/O activities. (2) Based on experimental results and collected dynamic system information, we have built a simulation model which accurately emulates the memory system operations and job migrations with virtual memory considerations. (3) We have proposed a memory-centric load sharing scheme and its variations to effectively process dynamic memory allocation demands, aiming at minimizing execution time of each individual job by dynamically migrating and remotely submitting jobs to eliminate or reduce page faults and to reduce the queuing rime for CPU services. Conducting trace-driven simulations, we have examined these load sharing policies to show their effectiveness.
Global File store (GFS) is a pervasive computing oriented file storage system. It puts forward an idea - put all files on the Internet. There are 3 features associated with GFS: massive storage, high performance and m...
详细信息
ISBN:
(纸本)0769513816
Global File store (GFS) is a pervasive computing oriented file storage system. It puts forward an idea - put all files on the Internet. There are 3 features associated with GFS: massive storage, high performance and mobile computing support. The approach to construct GFS is to implement a layer on top of existing heterogeneous file systems. The above features are realized by means of multiple namespace, extensive use of caching and replication, and highly scalable architecture. The system is in the stage of prototype development.
The number of agents and the execution rime are two significant performance factors in mobile agent planning. Fewer agents cause lower network traffic and consume less bandwidth. Regardless of the number of agents use...
详细信息
ISBN:
(纸本)0769510779
The number of agents and the execution rime are two significant performance factors in mobile agent planning. Fewer agents cause lower network traffic and consume less bandwidth. Regardless of the number of agents used, the execution time for a task must be kept minimal, which means that use of the minimal number of agents must not impact on the execution time unfavorably. As the population of the mobile agent application domain grows, the importance of these two factors also increases. After careful review of these two factors, we propose two heuristic algorithms for finding the minimal number of traveling agents for retrieving information from a distributedcomputing environment, while keeping the latency minimal. Although agent planning, specifically Mobile Agent Planning (MAP), is quire similar to the famous Traveling Salesman Problem (TSP), agent planning has a different objective function from that of TSP. TSP deals with optimal total routing cost, while MAP attempts to minimize the execution time to complete tasks of information retrieval. In this paper, we suggest the cost-effective MAP algorithms, BYKY1 and BYKY2, which can be used in distributed information retrieval systems to find the factors mentioned above. At the end of each algorithm, 2OPT, a well-known TSP algorithm, is called to optimize each agent's local routing path. Experimental results show that BYKY2 produces near optimal performance. These algorithms are more realistic and applicable directly to the problem domains than those of previous works.
With the increasing popularity of MBone conferences, the need arose to archive the multimedia streams of these conferences. Like all network applications, a conference recorder is subject to packet loss, which introdu...
详细信息
ISBN:
(纸本)0769510809
With the increasing popularity of MBone conferences, the need arose to archive the multimedia streams of these conferences. Like all network applications, a conference recorder is subject to packet loss, which introduces gaps into recorded media streams. Such gaps degrade the playback quality of an archived media stream or even disable the correct playback. In this paper we present a novel scheme for the loss-free recording of media streams. In contrast to existing systems that require the placement of recording caches within the network, our approach enables an end-to-end system design. We present a detailed protocol analysis that confirms the validity of our approach.
暂无评论