| NSF Supports Development of Stork Data Scheduler for Next Generation Data Intensive Computing |
 |
 |
 |
| Written by Tyler O'Neal | |||
| Thursday, 03 September 2009 06:22 | |||
|
The National Science Foundation (NSF) recently awarded Louisiana State University professor Tevfik Kosar with a half-million dollar grant to support development of the Stork Data Scheduler. As applications and experiments in all areas of science are becoming increasingly complex and more demanding in terms of their computational and data requirements, some applications generate data volumes reaching hundreds of terabytes and even petabytes. Sharing, disseminating, and analyzing these petascale data sets becomes a big challenge especially when distributed resources are used. Even though many regional and national optical networking initiatives such as LONI, ESNet and Teragrid provide high speed network connectivity to their users, majority of the users still fail to obtain even a fraction of the theoretical speeds promised by these networks due to mismanaged end-to-end data placement. The traditional distributed computing systems closely couple data placement and computation. They consider data resources as second class entities, and access to data as a side effect of computation. This makes the remote access and retrieval of data the main bottleneck in the end-to-end performance, reliability and automation of large-scale data-intensive and dynamic data-driven applications. Kosar’s NSF project, funded through the foundations Strategic Technologies for Cyberinfrastrucutre (STCI) program, will further develop and enhance the Stork data scheduler to mitigate the end-to-end data handling bottleneck in petascale distributed computing systems and make it available for a wide range of user community as a production quality software. The Stork Data Scheduler makes a distinctive contribution to distributed computing community because it focuses on planning, scheduling, monitoring and management of data movement tasks and data resources. Unlike existing approaches, Stork treats data resources and the tasks related to data access and movement as first class entities just like computational resources and compute tasks, and not simply the side effect of computation. Kosar, who holds a joint faculty appointment with the LSU Center for Computation & Technology, or CCT, has also received an NSF CAREER Award for his project titled “Data-aware Distributed Computing for Enabling Large-scale Collaborative Science.” earlier this year. The NSF CAREER Award is the foundation’s most prestigious award for junior faculty members. It is part of NSF’s Faculty Early Career Development Program, which “recognizes and supports the early career-development activities of those teacher-scholars who are most likely to become the academic leaders of the 21st century.” CAREER Award recipients are selected on the basis of creative career-development plans that effectively integrate research and education within the context of the missions of their institutions. Kosar’s CAREER grant establishes the theoretical background for data scheduling via development of novel mathematical models and algorithms. The recent STCI grant takes these models and algorithms, implements them in a production quality scheduling software, and makes them available for a wide range of science community. Enhanced functionalities of the Stork scheduler will include: data aggregation and caching; peer-to-peer and streamed data management; early error detection, classification, and recovery; job delegation and distributed data scheduling; integration with workflow planning and management; scheduled storage management; optimal protocol tuning; and end-to-end performance prediction services. The Stork data scheduler is considered a highly transformative project due to its potential to dramatically change how domain scientists perform their research and rapidly facilitate sharing of experience, raw data, and results. Future applications will be able to rely on Stork to manage storage and data movement reliably and transparently over a variety of storage and transfer protocols, thus eliminating unnecessary failure of distributed tasks. In December 2008, Kosar led a team of researchers who unveiled the first prototype of the Stork Data Scheduler (v 1.0), and made it available for the use of science community (www.storkproject.org). Data storage and management is Kosar’s research specialty at the University. In 2006, he received a $1 million grant from NSF to create advanced data archival, processing and visualization capabilities across the state through the PetaShare project (www.petashare.org).
|
|||
| Last Updated on Wednesday, 16 September 2009 09:09 |
| |
| |