OGSA-based grid workload monitoring

Proceedings of the 2nd International ICST Conference on Scalable Information Systems

Oliveira

Bivens

et al. 2007

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 summarydriven, 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.

Section: Discussionmentioning

confidence: 99%

“…It was repeatedly executed using a sliding window (e.g. one day) of data collected at short intervals on eDiaMoND (through [15]), each time using one data point. The service labeled the slowest by DPL the most times was isolated.…”

Section: Slow Response Time In Ediamondmentioning

confidence: 99%

Scalable Problem Localization for Distributed Systems: Principles and Practices

Proceedings of the 2nd International ICST Conference on Scalable Information Systems

Oliveira

Bivens

et al. 2007

“…The retrieved images are returned as part of the service responses to the radiologist for viewing and comparing. As is described in previous work [25], standard OGSAbased middleware, are instrumented with monitoring points, which measure the time elapsed at middleware components. A monitoring agent resides on each machine, listening to the monitoring points for data, possibly batching them before reporting them to the management server.…”

Section: Workload Monitoring For Service-oriented Systemsmentioning

confidence: 99%

Comparing the use of bayesian networks and neural networks in response time modeling for service-oriented systems

Proceedings of the 2007 Workshop on Service-Oriented Computing Performance: Aspects, Issues, and Approaches

Bivens

2007

The new paradigm of service-oriented computing facilitates easy construction of dynamic, complex distributed systems. Recent research has shown that machine learning methods can be a promising way to autonomously and accurately derive models to assist autonomic management software or humans in understanding system behaviors and making informed decisions. However, the efficacy of different machine learning techniques in describing various system behaviors and meeting distinct application needs has not been systematically understood. Such an understanding can prove crucial in management infrastructure design and implementation for service-oriented systems.This paper is an initial step to bridge the gap and specifically contrasts the applications of Bayesian networks (BN) and neural networks (NN) in modeling the response time of service-oriented systems. Relatively simple BN and NN models are designed and implemented as a base of the comparison study. As far as model performance is concerned, a wide range of simulations show that BNs offer better accuracy, are less sensitive to small data set size and are therefore more suited for environments that change rapidly and need frequent response time model reconstructions; whereas NNs can achieve faster model evaluation time and support management routines that demand intensive response time predictions. From a non-performance perspective, it is analytically concluded that BNs can be more easily understood by human and support multi-direction evaluation, while NNs provide more flexible response time representation.

“…al. [18], all eDiaMoND services and underlying components can be instrumented to trace requests, capturing workflow and measuring the end-to-end response time and time elapsed on each service. These data are the input to the problem localization approach described in the following section.…”

Section: The Ediamond Service-oriented Systemmentioning

confidence: 99%

Performance problem localization in self-healing, service-oriented systems using Bayesian networks

Proceedings of the 2007 ACM Symposium on Applied Computing

Moyle

McKeever

et al. 2007

In distributed, service-oriented environments, performance problem localization is required to provide self-healing capabilities and deliver the desired quality of service (QoS). This paper presents an automated approach to identifying system elements causing performance problems. Applying probabilistic inference to collected response time and elapsed time data, the approach 1) infers elapsed time for services where data is missing, 2) estimates the response time degradation caused by different services using the duration, abnormality and response time correlation of their elapsed times, and 3) identifies the services that are the most important causes of slow response time and yield the most benefit if recovered. The approach has been used to localize a performance problem on the test bed of a real-world serviceoriented Grid. Evaluation using simulations shows that the approach consistently achieves better accuracy than traditional techniques in various service-oriented settings.