Elisa Heymann

Adaptive Scheduling for Master-Worker Applications on the Computational Grid

We address the problem of how many workers should be allocated for executing a distributed application that follows the master-worker paradigm, and how to assign tasks to workers in order to maximize resource efficiency and minimize application execution time. We propose a simple but effective scheduling strategy that dynamically measures the execution times of tasks and uses this information to dynamically adjust the number of workers to achieve a desirable efficiency, minimizing the impact in loss of speedup. The scheduling strategy has been implemented using an extended version of MW, a runtime library that allows quick and easy development of master-worker computations on a computational grid. We report on an initial set of experiments that we have conducted on a Condor pool using our extended version of MW to evaluate the effectiveness of the scheduling strategy.

Analysis of Dynamic Heuristics for Workflow Scheduling on Grid Systems

Scheduling is an important factor for the efficient execution of computational workflows on grid environments. A large number of static scheduling heuristics has been presented in the literature. These algorithms allocate tasks before job execution starts and assume a precise knowledge of timing information, which may be difficult to obtain in general. To overcome this limitation of static strategies, dynamic scheduling strategies may be needed for a changing environment such as the grid. While they incur runtime overheads, they may better adapt to timing changes during job execution. In this work, we analyse five well-known heuristics (min-min, max-min, sufferage, HEFT and random) when used as static and dynamic scheduling strategies in a grid environment in which computing resources exhibit congruent performance differences. The analysis shows that non-list based heuristics are more sensitive than list-based heuristics to inaccuracies in timing information. Static list-based heuristics perform well in the presence of low or moderate inaccuracies. Dynamic versions of these heuristics may be needed only in environments where high inaccuracies are observed. Our analysis also shows that list-based heuristics significantly outperform non-list based heuristics in all cases and, therefore, constitute the most suitable strategies by which to schedule workflows either statically or dynamically

First Prototype of the CrossGrid Testbed

The CrossGrid project is developing new grid middleware components, tools and applications with a special focus on parallel and interactive computing. In order to support the development effort and provide a test infrastructure, an international grid testbed has been deployed across 9 countries. Through the deployment of the testbed and its supporting services, CrossGrid is also contributing to another important project objective, the expansion of the grid coverage in Europe. This paper describes the status of the CrossGrid testbed.

Resource Management for Interactive Jobs in a Grid Environment

Most recent grid middleware technologies have been aimed at the execution of sequential batch jobs. However, some users require interactive access when running jobs on grid sites. Execution of these applications on a grid environment is a challenging problem that requires the cooperation of several middleware tools and services. Additional problems arise when this interactive support is intended for parallel applications, which may run remotely across several sites. We provide transparent and reliable support for such applications. Our solution, based on the notion of split execution and interposition agents, allows running applications on a remote machine while some I/O operations are sent back to a home machine. The paper describes how we have applied interposition agents transparently to interactive applications and also describes a simple multiprogramming mechanism that is used to start interactive applications as fast as possible even under heavy occupancy of resources. We provide a performance evaluation of the key elements involved in the execution of interactive jobs

Evaluation of an Adaptive Scheduling Strategy for Master-Worker Applications on Clusters of Workstations

We investigate the problem arising in scheduling parallel applications that follow a master-worker paradigm in order to maximize both the resource efficiency and the application performance. We propose a simple scheduling strategy that dynamically measures application execution time and uses these measurements to automatically adjust the number of allocated processors to achieve the desirable efficiency, minimizing the impact in loss of speedup. The effectiveness of the proposed strategy has been assessed by means of simulation experiments in which several scheduling policies were compared. We have observed that our strategy obtains similar results to other strategies that use a priori information about the application, and we have derived a set of empirical rules that can be used to dynamically adjust the number of processors allocated to the application.

First principles vulnerability assessment

Clouds and Grids offer significant challenges to providing secure infrastructure software. As part of a our effort to secure such middleware, we present First Principles Vulnerability Assessment (FPVA), a new analyst-centric (manual) technique that aims to focus the analysts attention on the parts of the software system and its resources that are most likely to contain vulnerabilities that would provide access to high-value assets. FPVA finds new threats to a system and is not dependent on a list of known threats. Manual assessment is labor-intensive, making the use of automated assessment tools quite attractive. We compared the results of FPVA to those of the top commercial tools, providing the first significant evaluation of these tools against a real-world known collection of serious vulnerabilities. While these tools can find common problems in a programs source code, they miss a significant number of serious vulnerabilities found by FPVA. We are now using the results of this comparison study to guide our future research into improving automated software assessment.

The EU-CrossGrid Approach for Grid Application Scheduling

This paper presents the approach being followed to implement scheduling components that are integrated as part of the EU CrossGrid project. The purpose of these components is to provide a basis for supporting the efficient execution of distributed interactive applications on Grid environments. When a user submits a job, the scheduling services search for the most suitable resources to run the application and take subsequent steps to ensure a reliable launching of the application. All these actions are carried out according to user-defined preferences.

Managing MPI Applications in Grid Environments

One of the goals of the EU CrossGrid project is to provide a basis for supporting the efficient execution of parallel and interactive applications on Grid environments. CrossGrid jobs typically consist of computationally intensive simulations that are often programmed using a parallel programming model and a parallel programming library (MPI). This paper describes the key components that we have included in our resource management system in order to provide effective and reliable execution of parallel applications on a Grid environment. The general architecture of our resource management system is briefly introduced first and we focus afterwards on the description of the main components of our system. We provide support for executing parallel applications written in MPI either in a single cluster or over multiple clusters.

Supporting efficient execution of MPI applications across multiple sites

One of the main goals of the CrossGrid Project [1] is to provide explicit support to parallel and interactive compute- and data- intensive applications. The CrossBroker job manager provides services as part of the CrossGrid middleware and allows execution of parallel MPI applications on Grid resources in a transparent and automatic way. This document describes the design and implementation of the key components responsible for an efficient and reliable execution of MPI jobs splitted over multiple Grid sites, executed either in an on-line or batch manner. We also provide details on the overheads introduced by our system, as well as an experimental study showing that our system is well-suited for embarrassingly parallel applications.

Experience with the international testbed in the crossgrid project

The International Testbed of the CrossGrid Project has been in operation for the last three years, including 16 sites in 9 countries across Europe. The main achievements in installation and operation are described, and also the substantial experience gained on providing support to application and middleware developers in the project. Results are presented showing the availability of a realistic Grid framework to execute distributed interactive and parallel jobs.