Scalable distributed algorithms for multi-robot near-optimal motion planning
6
Citation
57
Reference
10
Related Paper
Citation Trend
Future supercomputing demands that large scale parallel algorithms and applications have good scalability. Previous scalability studies lay stress on the studies of the algorithms scalability,but few on that of the application programs. They couldn't give users the information about how to adjust programs to improve its performance. The numerical scalability and parallel scalability are provided to describe whether the parallel system maintains its numerical attributes and parallel attributes. Furthermore, a suit of scalability evaluation criterion is provided to help the user to find the reason causing the bad scalability and to modify programs. This criterion and the near optimal scalability method are used to analyze the scalability of a large scale application program, namely two\|dimensional electromagnetic plasma with particle in cell method. Results show that the criteria help to locate the reason why the scalability is bad, and that the near optimal scalability method provides an approach to predict how many processors are to be used by a larger problem to get a reasonable utility, where its time is near to the shortest time to run and its efficiency is much improved.
Cite
Citations (0)
Cite
Citations (5)
Multi-agent systems are one of the significant areas in the field of distributed systems. Scalability is one of the issues which effect successful deployment of a particular multi-agent system. Performance and scalability of a system cannot be decoupled because, within a limited number of resources, performance of a multi-agent system varies as the load is increased. Message transport service (MTS) provides the means for agents to communicate across the platform. For high performance, MTS must be dynamically scalable. Existing architectures, which are claimed to be scalable, have some shortcomings. They are centralized, and use component replication or distribution, which has some drawbacks. Priority based queues are one way to achieve dynamic scalability at component level. Peer communication channels provide a mechanism to distribute the load while communication is taking place among different machines. Moreover, peer-to-peer architecture is a hybrid approach to achieving scalability, as it is a blend of the merits of component distribution and replication.
Component (thermodynamics)
Replication
Message queue
Cite
Citations (5)
Cite
Citations (23)
Problem localization is a critical part of providing crucial system management capabilities to modern distributed environments. One key open challenge is for problem localization solutions to scale for systems containing hundreds or even thousands of nodes, whilst still remaining fast enough to resp
Cite
Citations (1)
Conducting experiments in large-scale distributed systems is usually time-consuming and labor-intensive. Uncontrolled external load variation prevents to reproduce experiments and such systems are often not available to the purpose of research experiments, e.g. production or yet to deploy systems. Hence, many researchers in the area of distributed computing rely on simulation to perform their studies. However, the simulation of large-scale computing systems raises several scalability issues, in terms of speed and memory. Indeed, such systems now comprise millions of hosts interconnected through a complex network and run billions of processes. Most simulators thus trade accuracy for speed and rely on very simple and easy to implement models. However, the assumptions underlying these models are often questionable, especially when it comes to network modeling. In this paper, we show that, despite a widespread belief in the community, achieving high scalability does not necessarily require to resort to overly simple models and ignore important phenomena. We show that relying on a modular and hierarchical platform representation, while taking advantage of regularity when possible, allows us to model systems such as data and computing centers, peer-to-peer networks, grids, or clouds in a scalable way. This approach has been integrated into the open-source SimGrid simulation toolkit. We show that our solution allows us to model such systems much more accurately than other state-of-the-art simulators without trading for simulation speed. SimGrid is even sometimes orders of magnitude faster.
Representation
Cite
Citations (21)
The management of the large scale heterogeneous resources is a critical issue in grid computing. The resource management system (RMS) is an essential component of grids. To ensure the QoS of the upper layer service, it raises high requirement for the performance, fault-tolerance and scalability of RMS. In this paper, we study three typical structures of RMS, including centralized, hierarchical and peer-to-peer structures, and make a comprehensive analysis of performance, fault tolerance and scalability. We put forward the performance, fault tolerance and scalability evaluation metrics of the RMS, and give the mathematical expressions and detailed calculation processes. Besides, we make further discussions on the interactions of the performance, fault-tolerance and scalability, and make a comparison of the RMSs with the three typical structures. We believe that the results of this work will help system architects make informed choices for building the RMS.
Resource Management
Fault management
Cite
Citations (8)
Problem localization is a critical part of providing crucial system management capabilities to modern distributed environments. One key open challenge is for problem localization solutions to scale for systems containing hundreds or even thousands of nodes, whilst still remaining fast enough to respond to rapid environment changes and sufficiently cost-effective to avoid overloading any management or application component. This paper meets the challenge by introducing two scalable frameworks applicable to a wide range of existing problem localization solutions: one based on a summary-driven, narrow-down procedure, the other through decomposing and decentralizing the problem localization process. Both frameworks, at their best, are able to achieve O(logN) problem localization time and O(1) per node communication load. The contrasting natures of both frameworks provide them with complimentary strengths that make them suitable for different scenarios in practice. We demonstrate our approaches in simulation settings and two real-world environments and show promising scalability benefits that can make a difference in system management operations.
Component (thermodynamics)
Cite
Citations (1)
This paper explores the challenges associated with distributed application management in large-scale computing environments. In particular, we investigate several techniques for extending Plush, an existing distributed application management framework, to provide improved scalability and fault tolerance without sacrificing performance. One of the main limitations of Plush is the structure of the underlying communication fabric. We explain how we incorporated the use of an overlay tree provided by Mace, a toolkit that simplifies the implementation of overlay networks, in place of the existing communication subsystem in Plush to improve robustness and scalability.
Overlay network
Robustness
Cite
Citations (4)
Presents a flexible load-sharing algorithm which achieves scalability by partitioning a system into domains. Each node dynamically and adaptively selects other nodes to be included in its domain. FLS applies load sharing within each domain, independently of how it is applied within other domains.< >
Load sharing
Cite
Citations (36)