Efficient I/O for parallel visualization
Thomas Fogal,Jens Krüger +1 more
- 10 Apr 2011
- pp 81-90
TL;DR: This work details the experiences using parallel file systems, details performance using current systems and software, and suggests a new API which has greater potential for increased scalability.
read more
Abstract: While additional cores and newer architectures, such as those provided by GPU clusters, steadily increase available compute power, memory and disk access has not kept pace, and most believe this trend will continue. It is therefore of critical importance that we design systems and algorithms which make effective use of off-processor storage. This work details our experiences using parallel file systems, details performance using current systems and software, and suggests a new API which has greater potential for increased scalability.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
An analysis of scalable GPU-based ray-guided volume rendering
Thomas Fogal,Alexander Schiewe,Jens Krüger +2 more
- 02 Dec 2013
TL;DR: This work demonstrates that the major limitation of most volume rendering approaches is their inability to switch the data sampling rate (and thus data size) quickly, and uses a volume renderer inspired by recent work to investigate design decisions typically swept under the rug in volume rendering literature.
References
NetCDF: an interface for scientific data access
TL;DR: The discussion covers NetCDF data abstraction and interface; dimensions, variables, and attributes; direct access and hyperslab access, the netCDF library; the data format; ncdump and ncgen utilities; experience, usability, and performance; limitations of NetC DF; and future plans.
861
Parallel netCDF: A High-Performance Scientific I/O Interface
Jianwei Li,Wei-keng Liao,Alok Choudhary,Robert Ross,Rajeev Thakur,William Gropp,Robert Latham,Andrew R. Siegel,Brad Gallagher,Michael Zingale +9 more
- 15 Nov 2003
TL;DR: This work presents a new parallel interface for writing and reading netCDF datasets that defines semantics for parallel access and is tailored for high performance, and compares the implementation strategies and performance with HDF5.
Disk-directed I/O for MIMD multiprocessors
TL;DR: In this article, the authors proposed a disk-directed I/O technique, which allows the disk servers to determine the flow of data for maximum performance, and demonstrated that this technique provided consistent high performance that was largely independent of data distribution.
347
Server-Directed Collective I/O in Panda
Kent E. Seamons,Y. Chen,P. Jones,J. Jozwiak,Marianne Winslett +4 more
- 08 Dec 1995
TL;DR: Panda achieves remarkable performance levels on the IBM SP2, showing excellent scalability as data size increases and as the number of nodes increases, and provides throughputs close to the full capacity of the AIX file system on the SP2 the authors used.
232
Performance characterization and optimization of parallel I/O on the Cray XT
Weikuan Yu,Jeffrey S. Vetter,H.S. Oral +2 more
- 14 Apr 2008
TL;DR: It is demonstrated that the I/O performance of the S3D combustion application can be improved at large scale by tuning the I-O system to avoid a bandwidth degradation of 49% with 8192 processes when compared to 4096 processes.
98