Global States Monitoring Research Articles

Distributed Application Global States Monitoring in PEGASUS DA Applied to Parallel Graph Partitioning

SummaryThis paper presents how parallel advanced graph partitioning algorithms can be designed and improved with the use of global application states monitoring of distributed programs. The proposed algorithms have been implemented inside a novel distributed program design framework Program Execution Governed by Asynchronous Supervision of States in Distributed Applications (PEGASUS DA). This framework provides system support for the design of execution control in distributed applications based on automated global state monitoring. Two strategies for the control design of advanced parallel/distributed graph partitioning algorithms are presented and discussed. In the first one, the parallel algorithm control runs on top of the popular graph partitioning METIS tool. The second control strategy is based on genetic programming so that partitioning primitives and the overall algorithmic control can be freely designed by the user. Advanced partitioning methods have been conveniently embedded inside the global state monitoring driven PEGASUS DA framework which controls partitioning distributed actions at the level of processes and threads. The use of the framework allowed easy design and testing of different graph optimization strategies. The presented graph partitioning methods are illustrated by experiments with benchmark graphs. The experiments have comparatively assessed the obtained graph partitioning quality (visible improvement has been observed) and have identified benefits of the proposed approach for programmers.

Application global state monitoring in optimization of parallel event‐driven simulation

SummaryThe paper is concerned with discrete event‐driven simulation, which is a well‐known technique used for modeling and simulating complex parallel systems. The focus of the paper is on distributed simulation, in which multiple simulated event queues are processed in parallel according to the Time Warp approach. With this approach, parallel simulation based on event queues is allowed to progress in an optimistic way until event correlation errors appear in parallel simulation branches what results in simulation rollbacks. In the absence of synchronization between simulator parallel queues, massive processing rollbacks can strongly slow down simulation. The paper presents a distributed optimistic event‐driven simulation control based on simulator global state monitoring. A systematic control of the simulator global states prevents excessive rollbacks in the Time Warp simulation. Each simulation event queue reports its progress to a global synchronizer which monitors the global simulation state based on virtual timestamps of recently processed events. Based on the global state, the synchronizer checks the simulation progress and sends control signals which asynchronously slow down servicing too advanced queues. The paper describes the principles of the proposed approach and experimental results of its basic program implementation. Comparisons to existing parallel simulation methods are provided.

PEGASUS DA framework for distributed program execution control based on application global states monitoring

SummaryThe paper presents basic design decisions taken in the implementation of a novel distributed program design framework Program Execution Governed by Asynchronous SUpervision of States in Distributed Applications (PEGASUS DA). This framework supports application program execution control design based on evolved automated program global states monitoring. The framework provides to a programmer a ready‐to‐use infrastructure for defining and handling local and global application states to be used as the basis for program execution control decisions. The paper presents how the provided infrastructure can be used for automated construction of strongly consistent application global states. Also, the use of global states for graphically supported specification of distributed program execution control is covered for clusters of multicore processors based on multithreading and message passing. Both architecture and implementation solutions applied for PEGASUS DA are discussed. Especially, multivariant algorithms for construction of program strongly consistent global states and methods for their use in the design of distributed program global execution control are shown. The use of PEGASUS DA is illustrated with an example of a traveling salesman problem solved by the branch and bound method. Copyright © 2015 John Wiley & Sons, Ltd.

A Distributed Program Global Execution Control Environment Applied to Load balancing

The paper is concerned with a new distributed program design environment based on the global application states monitoring. The environment called PEGASUS (from Program Execution Governed by Asynchronous SUpervision of States) supplies to a programmer a ready to use control primitives to design distributed program execution control in which decisions for synchronous and asynchronous control actions are based on predicates evaluated on global application states. Such strongly consistent global application states are automatically constructed by the run-time system which additionally provides mechanisms for their analysis and organizing the respective program execution control in processes and threads of user programs executed in multicore processors. The PEGASUS control mechanisms are graphically supported in the respective program design framework. The paper first presents main general features of the PEGASUS environment. Next, it presents a method for load balancing inside distributed programs based on a set of parameters which are dynamically measured during program execution. Then, the paper presents how the described load balancing method can be implemented inside the PEGASUS environment taking as an example distributed programs for solving the Traveling Salesman Problem (TSP).

Global states monitoring algorithm for distributed system

AbstractIn distributed systems, there are not only synchronization errors which stop the system but also errors in which each process behaves according to its process definition, yet the system as a whole behaves differently from the requirements. Since many synchronization errors cannot be detected by locally monitoring the state of each process, monitoring system global states is required. Stable global states are the global states with all channels empty, and monitoring them is known to be important for detecting loss of synchronization in distributed systems, i.e., the presence of unintended global states or the absence of intended ones.This paper describes a method of monitoring stable global states during the operation of distributed systems. This paper first clarifies the lower bound of control information, which each process should record for detection of stable global states, under the assumption that all control information is piggybacked with ordinary system messages. A stable global state monitoring algorithm is then presented, whose control information is minimum and which has minimum‐order complexity.