Balancing networks, originally introduced by Aspnes et al. (Proc. of the 23rd Annual ACM Symposium on Theory of Computing, pp. 348-358, May 1991), represent a new class of distributed, low-contention data structures suitable for solving many fundamental multi-processor coordination problems that can be expressed as balancing problems. In this work, we present a mathematical study of the combinatorial structure of balancing networks, andavariety of its applications. Our study identies important combinatorial transfer parameters of balancing networks. In turn, necessary and sucient combinatorial conditions are established, expressed in terms of transfer parameters, which precisely characterize many important and well studied classes of balancing networks suchascounting networks and smoothing networks.We propose these combinatorial conditions to be \balancing analogs" of the well known Zero-One principle holding for sorting networks.

Distributed systems are composed of several loosely-coupled computers communicating over a high-bandwidth network. To achieve an even distribution of the workload in a distributed system, either preemptive or non-preemptive load distribution strategies are used. Preemptive load distribution involves process migration, while non-preemptive strategies are based on initial placement of processes on the machines. Process migration is a mechanism where a process on one machine is moved to another machine in a distributed system. This paper discusses the design of process migration facilities in distributed systems with respect to key issues, such as the system models on which the mechanisms are implemented, the hardware platforms they run on, the methods used in moving a process from one machine to another, the load distribution policies adopted, network transparency, etc. 1 Introduction Enhancements to computer hardware have made "distributed computing systems" more and more available to ...

In this thesis, we present a new methodology for resource sharing algorithms in distributed systems. We propose that a distributed computing system should be composed of a decentralized community of microeconomic agents. We show that this approach decreases complexity and can substantially improve performance. We compare the performance, generality and complexity of our algorithms with non-economic algorithms. To validate the usefulness of our approach, we present economies that solve three distinct resource management problems encountered in large, distributed systems. The first economy performs CPU load balancing and demonstrates how our approach limits complexity and effectively allocates resources when compared to non-economic algorithms. We show that the economy achieves better performance than a representative non-economic algorithm. The load balancing economy spa...

The paper presents a family of distributed file structures, coined DiFS, for record structured, disk resident files with key based exact or interval match access. The file is organized into buckets that are spread among multiple servers, where a server may hold several buckets. Client requests are serviced by mapping keys onto buckets and looking up the corresponding server in an address table. Dynamic growth, in terms of file size and access load, is supported by bucket splits and bucket migrations onto the existing or newly created servers. The major problem that we are addressing is achieving scalability in the sense that both the file size and the client throughput can be scaled up by linearly increasing the number of servers and dynamically redistributing the data. Unlike previous work with similar objectives, our data redistribution considers explicitly the cost/performance ratio of the system by aiming to minimize the number of servers that are used to provide the required perfo...

The design of a process migration mechanism for the Amoeba distributed operating system is described. The primary motivation for this implementation is to carry out experimental and realistic studies of load balancing algorithms for a distributed operating system. Our aim has been the implementation of a mechanism which is general, efficient and fully transparent, and which is reliable in the presence of network and processor failures. 1

We investigate the problem of mapping dynamically generated tasks onto the processors of an MIMD-system. Our main concern is to construct an algorithm which can be integrated in distributed runtime systems like PVM or MPI. An algorithm which is suitable for this purpose has to be scalable for massively parallel systems and has to be adaptive to different architecture- and application-characteristics. Existing methods are often not adjustable to different architecture- and applicationdemands. Even if they are, the adjustment has to be done by hand via time-consuming experiments. A universally applicable strategy has to adjust its parameters automatically according to hardware- and application-characteristics. We concentrate on bidding-algorithms which check the load of K randomly selected processors before placing a task. The analysis of this method is based on a model which allows predictions of the behaviour of the scheduler. Especially for a large number n of processes it is possible...

This paper discusses a load balancing technique for distributed processing systems in which the load may vary over a wide range and at a high rate. Each processor performs a source or server algorithm for migrating processes when its load crosses some assigned upper or lower bound; these bounds are dynamically adjusted. Taking into account the speed at which loads vary and the latency of the underlying network, we specify conditions under which the algorithm is stable and responds satisfactorily to fast load changes. Simulation confirms the validity of these conditions.

The COOL system has been designed to satisfy two goals; to provide an efficient object oriented support layer built directly on the Chorus Micro-Kernel and to co-exist with UNIX by offering access paths, and by using Unix functionality. In this paper we discuss how the COOL system has been designed to exploit features of the Chorus operating system model, to provide abstractions suited to object oriented systems. A significant feature of the COOL project is that we have been able to exploit the design methodology of Chorus to allow COOL to co-exist with the Chorus Unix implementation (MiX) [Armand89] yet still build our abstractions directly onto the Chorus Nucleus, exploiting its efficiency and functionality. To demonstrate the feasibility of this approach and to gain experience of our abstractions, we have built a C++ environment on the COOL sub-system that enables programmers to develop applications consisting of standard C++ objects that can then be used in a transparent manner...

A runtime support is necessary for parallel computations with irregular and dynamic structures. One important component in the support system is the runtime scheduler which balances the working load in the system. We present a new algorithm, Symmetrical Hopping, for dynamic scheduling of ultra-lightweight processes. It is a dynamic, distributed, adaptive, and scalable scheduling algorithm. This algorithm is described and compared to four other algorithms that have been proposed in this context, namely the randomized allocation, the sender-initiated scheduling, the receiver-initiated scheduling, and the gradient model. The performance of these algorithms on Intel Touchstone Delta is presented. The experimental results show that the Symmetrical Hopping algorithm achieves much better performance due to its adaptiveness. 1. Introduction Large distributed memory parallel machines are becoming increasingly available. To efficiently use such large machines to solve an application problem, th...

This paper looks at the design, implemntation and performance of RHODOS' process migration facility which reflects RHODOS' aspiration to support both load balancing and parallel execution on a distributed system. Following the presentation of the design issues, which reflect the requirements of both load balancing and parallel execution on a distributed system, research into an initial implementation and testing of RHODOS' process migration facility will be presented. Performance measurements of this initial implementation have also been taken and the results are reported in this paper. A comparison with several modern distributed operating systems show that already our migration facility compares favourably. 1. This work was partly supported by the Australian Research Council Grant A49232429 and the Deakin University Research Grant 0504010151 2 1 Introduction In a distributed system there are many workstations which are either idle or lightly loaded, while others are heavily load...