Using virtual Lustre clients on the WAN for analysis of data from high energy physics experiments
Creators
- 1. University of Florida, Gainesville, FL (United States)
- 2. Pittsburgh Supercomputing Center, Pittsburgh, PA (United States)
- 3. Florida State University, Tallahassee, FL (United States)
Description
We describe the work on creating system images of Lustre virtual clients in the ExTENCI project (Extending Science Through Enhanced National Cyberlnfrastructure), using several virtual technologies (Xen, VMware, VirtualBox, KVM). These virtual machines can be built at several levels, from a basic Linux installation (we use Scientific Linux 5 as an example), adding a Lustre client with Kerberos authentication, and up to complete clients including local or distributed (based on CernVM-FS) installations of the full CERN and project specific software stack for typical LHC experiments. The level, and size, of the images are determined by the users on demand. Various sites and individual users can just download and use them out of the box on Linux/UNIX, Windows and Mac OS X based hosts. We compare the performance of virtual clients with that of real physical systems for typical high energy physics applications like Monte Carlo simulations or analysis of data stored in ROOT trees.
Availability note (English)
Available from http://dx.doi.org/10.1088/1742-6596/396/3/032013Additional details
Identifiers
Publishing Information
- Journal Title
- Journal of Physics. Conference Series (Online)
- Journal Volume
- 396
- Journal Issue
- 3
- Journal Page Range
- [7 p.]
- ISSN
- 1742-6596
Conference
- Title
- International conference on computing in high energy and nuclear physics 2012
- Acronym
- CHEP2012
- Dates
- 21-25 May 2012
- Place
- New York, NY (United States)
INIS
- Country of Publication
- United Kingdom
- Country of Input or Organization
- International Atomic Energy Agency (IAEA)
- INIS RN
- 44038019
- Subject category
- S46: INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND TECHNOLOGY; S97: MATHEMATICAL METHODS AND COMPUTING;
- Resource subtype / Literary indicator
- Conference
- Descriptors DEI
- CERN LHC; COMPARATIVE EVALUATIONS; COMPUTER CALCULATIONS; COMPUTER CODES; COMPUTER NETWORKS; COMPUTERIZED SIMULATION; DATA ANALYSIS; DATA TRANSMISSION; DISTRIBUTED DATA PROCESSING; MONTE CARLO METHOD; MULTIPARTICLE SPECTROMETERS; MULTIPLE PRODUCTION; PARTICLE IDENTIFICATION; PERFORMANCE
- Descriptors DEC
- ACCELERATORS; CALCULATION METHODS; COMMUNICATIONS; CYCLIC ACCELERATORS; DATA PROCESSING; EVALUATION; MEASURING INSTRUMENTS; PARTICLE PRODUCTION; PROCESSING; SIMULATION; SPECTROMETERS; STORAGE RINGS; SYNCHROTRONS