skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Automatic Testing Tool for OSCAR Using System-level Virtualization

Abstract

To ensure quality, software development has to include testing mechanisms. OSCAR today supports several Linux distributions and several architectures. In such a context, the release cycle suffers of a important overhead created by the testing and stabilization phase. To address this issue, an approach is to implement a tool for automatic testing. This paper presents such a tool which is based on the OSCAR command line interface. This tool, based on system-level virtualization techniques, creates a virtual cluster to perform the test. This approach has the benefit of not corrupting the system of the physical machine and guarantee that the environment used for testing has not been corrupted before testing.

Authors:
 [1];  [1];  [1];  [1]
  1. ORNL
Publication Date:
Research Org.:
Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
Sponsoring Org.:
USDOE Laboratory Directed Research and Development (LDRD) Program
OSTI Identifier:
991173
DOE Contract Number:
DE-AC05-00OR22725
Resource Type:
Conference
Resource Relation:
Conference: The 5th Annual OSCAR Symposium, Saskatoon, Canada, 20070513, 20070516
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICAL METHODS AND COMPUTING; STABILIZATION; TESTING; O CODES; AUTOMATION; COMPUTER GRAPHICS

Citation Formats

Vallee, Geoffroy R, Naughton, III, Thomas J, Bland, Wesley B, and Scott, Stephen L. Automatic Testing Tool for OSCAR Using System-level Virtualization. United States: N. p., 2007. Web.
Vallee, Geoffroy R, Naughton, III, Thomas J, Bland, Wesley B, & Scott, Stephen L. Automatic Testing Tool for OSCAR Using System-level Virtualization. United States.
Vallee, Geoffroy R, Naughton, III, Thomas J, Bland, Wesley B, and Scott, Stephen L. Mon . "Automatic Testing Tool for OSCAR Using System-level Virtualization". United States. doi:.
@article{osti_991173,
title = {Automatic Testing Tool for OSCAR Using System-level Virtualization},
author = {Vallee, Geoffroy R and Naughton, III, Thomas J and Bland, Wesley B and Scott, Stephen L},
abstractNote = {To ensure quality, software development has to include testing mechanisms. OSCAR today supports several Linux distributions and several architectures. In such a context, the release cycle suffers of a important overhead created by the testing and stabilization phase. To address this issue, an approach is to implement a tool for automatic testing. This paper presents such a tool which is based on the OSCAR command line interface. This tool, based on system-level virtualization techniques, creates a virtual cluster to perform the test. This approach has the benefit of not corrupting the system of the physical machine and guarantee that the environment used for testing has not been corrupted before testing.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Mon Jan 01 00:00:00 EST 2007},
month = {Mon Jan 01 00:00:00 EST 2007}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • New virtualization solutions such as Xen allow users to execute hundreds of virtual machines on a single physical machine. The interest of these solutions have been proven for system isolation and security features, especially for Internet Service Providers (ISPs), as well as for high performance computing. A natural question is to know if it is possible to use all these virtual machines at the same time, creating a virtual cluster. This might be an interesting solution for the development and the experimentation of cluster applications. This document presents an extension of OSCAR for the deployment and the management of Xenmore » virtual machines. We also analyze in this paper the interest of virtualization for the development, the testing and the experimentation of applications for clusters, in particular with the use of a fully virtualized cluster.« less
  • System-level virtualization has been a research topic since the 70's but regained popularity during the past few years because of the availability of efficient solution such as Xen and the implementation of hardware support in commodity processors (e.g. Intel-VT, AMD-V). However, a majority of system-level virtualization projects is guided by the server consolidation market. As a result, current virtualization solutions appear to not be suitable for high performance computing (HPC) which is typically based on large-scale systems. On another hand there is significant interest in exploiting virtual machines (VMs) within HPC for a number of other reasons. By virtualizing themore » machine, one is able to run a variety of operating systems and environments as needed by the applications. Virtualization allows users to isolate workloads, improving security and reliability. It is also possible to support non-native environments and/or legacy operating environments through virtualization. In addition, it is possible to balance work loads, use migration techniques to relocate applications from failing machines, and isolate fault systems for repair. This document presents the challenges for the implementation of a system-level virtualization solution for HPC. It also presents a brief survey of the different approaches and techniques to address these challenges.« less
  • No abstract prepared.
  • Laser damage threshold measurement is statistical in nature. For a commercial qualification or for a user, the threshold determined by the weakest point is a satisfactory characterization. When a new coating is designed, threshold mapping is very useful. It enables the technology to be improved and followed more accurately. Different statistical parameters such as the minimum, maximum, average, and standard deviation of the damage threshold as well as spatial parameters such as the threshold uniformity of the coating can be determined. Therefore, in order to achieve a mapping, all the tested sites should give data. This is the major interestmore » of the R-on-1 test in spite of the fact that the laser damage threshold obtained by this method may be different from the 1-on-1 test (smaller or greater). Moreover, on the damage laser test facility, the beam size is smaller (diameters of a few hundred micrometers) than the characteristic sizes of the components in use (diameters of several centimeters up to one meter). Hence, a laser damage threshold mapping appears very interesting, especially for applications linked to large optical components like the Megajoule project or the National Ignition Facility (N.I.F). On the test bench used, damage detection with a Nomarski microscope and scattered light measurement are almost equivalent. Therefore, it becomes possible to automatically detect on line the first defects induced by YAG irradiation. Scattered light mappings and laser damage threshold mappings can therefore be achieved using a X-Y automatic stage (where the test sample is located). The major difficulties due to the automatic capabilities are shown. These characterizations are illustrated at 355 nm. The numerous experiments performed show different kinds of scattering curves, which are discussed in relation with the damage mechanisms.« less