Enabling high-speed asynchronous data extraction and transfer using DART

Concurrency and Computation: Practice and Experience

Volumn 22, Issue 9, 2010, Pages 1181-1204

  Docan, Ciprian ^a   Parashar, Manish ^a   Klasky, Scott ^b  

^a RUTGERS UNIVERSITY   (United States)

^b OAK RIDGE NATIONAL LABORATORY   (United States)

Author keywords

Asynchronous transfers; Low overhead; RDMA

Indexed keywords

COMPUTER SOFTWARE; DATA TRANSFER; EXTRACTION; MAGNETOPLASMA; PLASMA SIMULATION; PLASMA TURBULENCE; QUALITY OF SERVICE;

ASYNCHRONOUS TRANSFERS; COMMUNICATION LATENCY; COMPUTATIONAL PERFORMANCE; FUSION SIMULATION PROJECTS; LARGE AMOUNTS OF DATA; LOW OVERHEAD; OAK RIDGE NATIONAL LABORATORY; RDMA;

DATA MINING;

EID: 77953455120     PISSN: 15320626     EISSN: 15320634     Source Type: Journal    
DOI: 10.1002/cpe.1567     Document Type: Article

References (18)

1. Brightwell R, Hudson T, Pedretti K, Riesen R, Underwood K. Implementation and performance of Portals 3.3 on the Cray XT3. Cluster Computing, Boston, MA, U.S.A., 26-30 September 2005; 1-10.
2. Lin Z, Hahm TS, Lee WW, Tang WM, White RB. Turbulent transport reduction by zonal flows: Massive parallel simulations. Science 1998; 281(5384):1835-1837.
3. Klasky S, Ethier S, Lin Z, Martins K, McCune D, Samtaney R. Grid-based parallel data streaming implemented for the gyrokinetic toroidal code. Supercomputing Conference (SC'03), Pheonix, AZ, U.S.A., 15-21 November 2003; 24.
4. Chang C, Ku S, Weitzner H. Numerical study of neoclassical plasma pedestal in a tokamak geometry. Physics of Plasmas 2004; 11(5):2649-2667.
5. Available at: http://info.nccs.gov/resources/jaguar [December 2008].
6. Inc C. Cray XT3tm sytem overview. Technical Report S-2423-2513, November 2005.
7. Inc CFS. Lustre: A scalable, high performance file system. Available at: http://lustre.org/docs/whitepaper.pdf [November 2008].
8. Inc I. GPFS: A shared-disk file system for large computing clusters. Available at: http://www.almaden.ibm.com/StorageSystems/projects/gpfs/Fast02.pdf [September 2008].
9. Thakur R, Gropp W, Lusk E. Data sieving and collective I/O in ROMIO. Symposium on the Frontiers of Massively Parallel Computation, Annapolis, MD, U.S.A., 21-25 February 1999; 182-189.
10. Stone N, Balog D, Gill B, Johanson B, Marsteller J, Nowoczynski P, Porter D, Reddy R, Scott J, Simmel D et al. PDIO: High-performance remote file I/O for portals-enabled compute nodes. Parallel and Distributed Processing Techniques and Applications (PDPTA'06), Las Vegas, NV, U.S.A., 26-29 June 2006; 925-930.
11. Abbasi H, Wolf M, Schwan K. LIVE data workspace: A flexible, dynamic and extensible platform for petascale applications. Cluster Computing, Austin, TX, U.S.A., 17-20 Septemeber 2007; 341-348.
12. HDF-5. Available at: http://hdf.ncsa.uiuc.edu/products/hdf5/index.html [August 2008].
13. Parallel netCDF. Available at: http://trac.mcs.anl.gov/projects/parallel- netcdf [August 2008].
14. Thain D, Basney J, Son SC, Livny M. The Kangaroo approach to data movement on the Grid. International Symposium on High Performance Distributed Computing (HPDC'04), Honolulu, HI, U.S.A., 4-6 June 2004; 325-333.
15. Magoutis K, Addetia S, Fedorova A, Seltzer M, Chase JS, Gallatin AJ, Kisley R, Wickremesinghe RG, Gabber E. Structure and performance of the direct access file system. Proceedings of USENIX Annual Technical Conference, Monterey, CA, U.S.A., 10-15 June 2002; 1-14.
16. del Rosario JM, Choudhary AN. High-performance I/O for massively parallel computers: problems and prospects. Computer 1994; 27(3):59-68.
17. Seamons KE, Chen Y, Jones P, Jozwiak J, Winslett M. Server-directed collective I/O in Panda. Supercomputing Conference (SC'95), San Diego, CA, U.S.A., 4-8 December 1995; 57.
18. Kotz D. Disk-directed I/O for MIMD multiprocessors. ACM Transactions on Computer Systems (TOCS) 1997; 15(1): 41-74.