Abstract-One measure of usefulness of a general-purpose distrib-uted computing system is the system’s ability to provide a level of per-formance commensurate to the degree of multiplicity of resources pres-ent in the system. Many different approaches and metrics of performance have been proposed in an attempt to achieve this goal in existing systems. In addition, analogous problem formulations exist in other fields such as control theory, operations research, and produc-tion management. However, due to the wide variety of approaches to this problem, it is difficult to meaningfully compare different systems since there is no uniform means for qualitatively or quantitatively eval-uating them. It is difficult to successfully build upon existing work or identify areas worthy of additional effort without some understanding of the relationships between past efforts. In this paper, a taxonomy of approaches to the resource management problem is presented in an attempt to provide a common terminology and classification mecha-nism necessary in addressing this problem. The taxonomy, while pre-sented and discussed in terms of distributed scheduling, is also appli-cable to most types of resource management. As an illustration of the usefulness of the taxonomy an annotated bibliography is given which classifies a large number of distributed scheduling approaches accord-ing to the taxonomy. Index Terms-Distributed operating systems, distributed resource management, general-purpose distributed computing systems, sched-uling, task allocation, taxonomy. T I.

A process is an operating system abstraction representing an instance of a running computer program. Process migration is the act of transferring a process between two machines during its execution. Several implementations

The maximum flow algorithm for minimizing energy functions of binary variables has become a standard tool in computer vision. In many cases, unary costs of the energy depend linearly on parameter λ. In this paper we study vision applications for which it is important to solve the maxflow problem for different λ’s. An example is a weighting between data and regularization terms in image segmentation or stereo: it is desirable to vary it both during training (to learn λ from ground truth data) and testing (to select best λ using high-knowledge constraints, e.g. user input). We review algorithmic aspects of this parametric maximum flow problem previously unknown in vision, such as the ability to compute all breakpoints of λ and corresponding optimal configurations in finite time. These results allow, in particular, to minimize the ratio of some geometric functionals, such as flux of a vector field over length (or area). Previously, such functionals were tackled with shortest path techniques applicable only in 2D. We give theoretical improvements for “PDE cuts ” [5]. We present experimental results for image segmentation, 3D reconstruction, and the cosegmentation problem. 1.

In this paper, we consider the complexity of a number of combinatorial problems; namely, Intervalizing Colored Graphs (DNA physical mapping), Triangulating Colored Graphs (perfect phylogeny), (Directed) (Modified) Colored Cutwidth, Feasible Register Assignment and Module Allocation for graphs of bounded pathwidth. Each of these problems has as a characteristic a uniform upper bound on the tree or path width of the graphs in "yes"-instances. For all of these problems with the exceptions of Feasible Register Assignment and Module Allocation, a vertex or edge coloring is given as part of the input. Our main results are that the parameterized variant of each of the considered problems is hard for the complexity classes W [t] for all t 2 N. We also show that Intervalizing Colored Graphs, Triangulating Colored Graphs, and Colored Cutwidth are NP-Complete. 1 Introduction This paper focuses on a number of graph decision problems which share the characteristic that all have a uniform upper bo...

We investigate how delayed communication affects the dynamic formation of groups in distributed systems, where all decision-making agents join the same group because each expects to improve its own performance. For example, distributed job schedulers may form a group to utilize the idle resources of other members within the group. Forming a group is a search problem and we examine agents which use the feedback mechanism of stochastic learning automata to carry out this search. Although a group formation may have the potential for synergy, the agents must successfully coordinate their actions within the group relevant to the application. For example, job schedulers who form a group must still balance the load among the shared resources; that is, the collective actions of the schedulers need to be coordinated and greedy schedulers who all pick the same processor may not be successful. Agents may find that working alone is more desirable since their actions need not be coordinated and the r...

Each job scheduling agent in large decentralized load balancing systems generally has a set of remote hosts (i.e. its scope of control) to consider for ofloading when the local load is too high. Typically, each agent’s scope of control includes all the hosts in the system. We investigate the potential performance benefits of limiting the size of each agent’s scope of control. The larger the scope of control, the less often state informa-tion can be communicated, and old state information can lower the quality of load balancing decisions. The smaller the scope of control, the less opportunity an agent has forfinding a lightly loaded host for ofload-ing. Agents adaptively modify their scope of control over time based on feedback regarding the success of load balancing decisions and act as a self-organizing system to eficiently share available processing power. 1

Abstract—Load balancing is the process of improving the performance of a parallel and distributed system through a redistribution of load among the processors [1] [5]. In this paper we present the performance analysis of various load balancing algorithms based on different parameters, considering two typical load balancing approaches static and dynamic. The analysis indicates that static and dynamic both types of algorithm can have advancements as well as weaknesses over each other. Deciding type of algorithm to be implemented will be based on type of parallel applications to solve. The main purpose of this paper is to help in design of new algorithms in future by studying the behavior of various existing algorithms. Keywords—Load balancing (LB), workload, distributed systems, Static Load balancing, Dynamic Load Balancing I.

Abstract not found

A bipartite graph G = (U, V, E) is a chain graph [9] if there is a bijection π: {1,..., |U|} → U such that Γ (π (1)) ⊇ Γ (π (2)) ⊇... ⊇ Γ (π (|U|)), where Γ is a function that maps a node to its neighbors. We give approximation algorithms for two variants of the Minimum Chain Completion problem, where we are given a bipartite graph G(U, V, E), and the goal is find the minimum set of edges F that need to be added to G such that the bipartite graph G ′ = (U, V, E ′ ) (E ′ = E ∪ F) is a chain graph. 1

Abstract—Decrease in hardware costs and advances in computer networking technologies have led to increased interest in the use of large-scale parallel and distributed computing systems. One of the biggest issues in such systems is the development of effective techniques/algorithms for the distribution of the processes/load of a parallel program on multiple hosts to achieve goal(s) such as minimizing execution time, minimizing communication delays, maximizing resource utilization and maximizing throughput. Substantive research using queuing analysis and assuming job arrivals following a Poisson pattern, have shown that in a multi-host system the probability of one of the hosts being idle while other host has multiple jobs queued up can be very high. Such imbalances in system load suggest that performance can be improved by either transferring jobs from the currently heavily loaded hosts to the lightly loaded ones or distributing load evenly/fairly among the hosts.The algorithms known as load balancing algorithms, helps to achieve the above said goal(s). These algorithms come into two basic categories-static and dynamic. Whereas static load balancing algorithms (SLB) take decisions regarding assignment of tasks to processors based on the average estimated values of process execution times and communication delays at compile time, Dynamic load balancing algorithms (DLB) are adaptive to changing situations and take decisions at run time. The objective of this paper work is to identify qualitative parameters for the comparison of above said algorithms. In future this work can be extended to develop an experimental environment to study these Load balancing algorithms based on comparative parameters quantitatively.