Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Home Topics Titles Quotes Blog Featured Help
Search
 
Gabriel Mateescu
NCSA
Urbana, Illinois
 

Gabriel Mateescu works as a systems and applications analyst in the Research Computing Support Group at the National Research Council Canada (NRC), Ottawa. His responsibilities include building computational grids, developing middleware for managing access to multiple grid resources, and enabling scientific applications to effectively use grid resources. Gabriel has grid-enabled several pleasantly parallel applications, providing features such as transparent file staging and load balancing across multiple resource managers. Gabriel has been a member of the team that in November 2002 built Canada's largest virtual supercomputer, the Canadian Internetworked Scientific Supercomputer. He is also porting parallel programs to emerging Linux cluster platforms.

Gabriel has given presentations on grid resource management at prestigious venues, such as the Global Grid Forum and the Supercomputing 2002 Conference.

Before coming to NRC, Gabriel worked at Platform Computing in Toronto, VTLS Inc. in Blacksburg, VA, and Los Alamos National Laboratory (LANL). At Platform, he worked on the LSF distributed resource management system. At LANL, Gabriel developed parallel software and algorithms for solving very large systems of linear equations that arise from the discretization of partial differential equations. His work at VTLS, a leading library solutions company, involved developing the serial publications component of the server tier of the Virtua integrated library system.

Gabriel is the author of peer-reviewed papers on parallel iterative linear solvers, grid computing, parallel sorting, and broadcast scheduling for parallel programs using message passing communication.

Gabriel holds a Ph.D. and an M.S. in Computer Science from Virginia Tech.


     

The quest for low storage latency changes everything
Rick Coulson. YouTube, 01:29:40, published on Mar 3, 2016, stanfordonline, https://www.youtube.com/watch?v=hXurTRtmfWc. Type: Video

This video presentation gives an introduction to 3D XPoint, Intel’s next-generation non-volatile memory (NVM) technology, developed by Intel and Micron. The presenter, Rick Coulson, senior fellow at Intel and leader of the In...

 

 Computing petaflops over terabytes of data: the case of genome-wide association studies
Fabregat-Traver D., Bientinesi P. ACM Transactions on Mathematical Software 40(4): 1-22, 2014.  Type: Article

With its catchy but somewhat misleading title, this paper proposes a method for efficiently solving multiple instances of the same problem when the instances are correlated....

 

A primal-dual randomized algorithm for weighted paging
Bansal N., Buchbinder N., Naor J. Journal of the ACM 59(4): 1-24, 2012.  Type: Article

Paging is a memory management technique where memory is divided into chunks of constant size called pages. A two-level memory system consists of a small fast memory (the cache) that can hold k pages and a large slow ...

 

Biased range trees
Dujmović V., Howat J., Morin P. Algorithmica 62(1-2): 21-37, 2012.  Type: Article

A two-sided orthogonal range counting query over a finite set S of n points in R2 asks, given a query point q = (qx, qy<...

 

SLA enabled CARE resource broker
Balakrishnan P., Somasundaram T. Future Generation Computer Systems 27(3): 265-279, 2011.  Type: Article

An essential function of cloud and grid computing is the management of resources (including provisioning, scheduling, and monitoring). Scheduling middleware that coordinates lower-level schedulers that control resources is called a met...

 
  more...

 
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy