43 Hits in 2.5 sec

Supporting efficient noncontiguous access in PVFS over Infiniband

Jiseheng Wu, Wyckoff, Panda
2003 Proceedings IEEE International Conference on Cluster Computing CLUSTR-03  
We have designed and incorporated this approach in a version of PVFS over InfiniBand.  ...  This characteristic imposes a requirement of native noncontiguous I/O access support in cluster file systems for high performance.  ...  Acknowledgments We would like to thank the PVFS team at Argonne National Laboratory and Clemson University for giving us access to the latest versions of PVFS and for providing us with crucial insights  ... 
doi:10.1109/clustr.2003.1253333 dblp:conf/cluster/WuWP03 fatcat:zesr2ucmyban7e3fmo6jegahqy

PVFS over InfiniBand: design and performance evaluation

J. Wu, P. Wyckoff, Dhabaleswar Panda
2003 2003 International Conference on Parallel Processing, 2003. Proceedings.  
To the best of our knowledge, this is the first design, implementation and evaluation of PVFS over InfiniBand.  ...  Compared to a PVFS implementation over standard TCP/IP on the same InfiniBand network, our implementation offers three times the bandwidth if workloads are not disk-bound and 40% improvement in bandwidth  ...  Acknowledgments: We would like to thank the PVFS team at Argonne National Laboratory and Clemson University for giving us the access to the latest version of PVFS implementation and for providing us with  ... 
doi:10.1109/icpp.2003.1240573 dblp:conf/icpp/WuWP03 fatcat:cdy2wvjzargfjeiuzln32vv37q

RXIO: Design and implementation of high performance RDMA-capable GridFTP

Yuan Tian, Weikuan Yu, Jeffrey S. Vetter
2012 Computers & electrical engineering  
For its low-latency, high bandwidth, and low CPU utilization, Remote Direct Memory Access (RDMA) has established itself as an effective data movement technology in many networking environments.  ...  In this study, we examine the architecture of GridFTP for the feasibility of enabling RDMA.  ...  We are very thankful for an InfiniBand equipment donation from HPC Advisor Council to Auburn University.  ... 
doi:10.1016/j.compeleceng.2011.11.008 fatcat:6f7z7seu7fdrnmlta4pwhqptgi

Noncontiguous locking techniques for parallel file systems

Avery Ching, Wei-keng Liao, Alok Choudhary, Robert Ross, Lee Ward
2007 Proceedings of the 2007 ACM/IEEE conference on Supercomputing - SC '07  
We implement our scalable distributed lock manager (DLM) in the PVFS parallel file system and show that these techniques improve locking throughput over a naive noncontiguous locking approach by several  ...  Atomic I/O in current parallel file systems is often slow when multiple processes simultaneously access interleaved, shared files.  ...  This work was supported in part by DOE's  ... 
doi:10.1145/1362622.1362658 dblp:conf/sc/ChingLCRW07 fatcat:6ccnageaffgexi5zsi7pvcnek4

RADAR: Runtime Asymmetric Data-Access Driven Scientific Data Replication [chapter]

John Jenkins, Xiaocheng Zou, Houjun Tang, Dries Kimpe, Robert Ross, Nagiza F. Samatova
2014 Lecture Notes in Computer Science  
Our system can produce up to manyfold improvements in commonly used subvolume decomposition access patterns.  ...  We capture datatype-and collective-aware I/O access patterns (indicating logical access) via MPI-IO tracing and use a combination of coarse-grained and fine-grained performance modeling to evaluate and  ...  Acknowledgments This work was supported by the U.S. Department of Energy, Office of Science, under Contract No. DE-AC02-06CH11357.  ... 
doi:10.1007/978-3-319-07518-1_19 fatcat:j6h4wsxfsbfybfxt2p5w5mbvsq

Scalable I/O forwarding framework for high-performance computing systems

Nawab Ali, Philip Carns, Kamil Iskra, Dries Kimpe, Samuel Lang, Robert Latham, Robert Ross, Lee Ward, P. Sadayappan
2009 2009 IEEE International Conference on Cluster Computing and Workshops  
POSIX requires extensions to enable efficient noncontiguous I/O.  ...  Whereas ZOIDFS can describe noncontiguous I/O in a single function call, POSIX-based ROMIO drivers have to take less efficient approaches, such as data sieving [28] .  ... 
doi:10.1109/clustr.2009.5289188 dblp:conf/cluster/AliCIKLLRWS09 fatcat:imcs7pkqhza7dpfsckol446fsu

High Performance Block I/O for Global File System (GFS) with InfiniBand RDMA

Shuang Liang, Weikuan Yu, Dhabaleswar K. Panda
2006 2006 International Conference on Parallel Processing (ICPP'06)  
We evaluate this new scheme in comparison with our copy based scheme and TCP over the same InfiniBand hardware.  ...  performance over 10Gbps InfiniBand network.  ...  Several levels of Quality of Service (QoS) are supported in InfiniBand. The Reliable Connection (RC) service guarantees reliable transport and supports RDMA in hardware. Figure 2.  ... 
doi:10.1109/icpp.2006.47 dblp:conf/icpp/LiangYP06 fatcat:nyxne57ptbdvfi7jlt4dmhqq4y

AHPIOS: An MPI-Based Ad Hoc Parallel I/O System

Florin Isaila, Javier Garcia Blas, Jesus Carretero, Wei-keng Liao, Alok Choudhary
2008 2008 14th IEEE International Conference on Parallel and Distributed Systems  
PVFS [15] is an open source parallel file system that targets the efficient access to large data sets. AHPIOS can be used alternatively to these parallel file systems.  ...  AH-PIOS virtualizes on-demand available distributed storage resources and allows the files to be striped over several storage devices.  ...  The communication inside the file systems and in the MPICH2 was done with TCP/IP sockets over Infiniband.  ... 
doi:10.1109/icpads.2008.50 dblp:conf/icpads/IsailaBCLC08 fatcat:2jnwgugypvdp7hvoy6242ahrqm

Head-to-TOE Evaluation of High-Performance Sockets over Protocol Offload Engines

P. Balaji, W. Feng, Q. Gao, R. Noronha, W. Yu, D. K. Panda
2005 Proceedings IEEE International Conference on Cluster Computing  
In addition to 10GigE's advantage with respect to compatibility to wide-area network infrastructures, e.g., in support of grids, our results show that 10GigE also delivers performance that is comparable  ...  to traditional high-speed network technologies such as IBA and Myrinet in a system-area network environment to support clusters and that 10GigE is particularly wellsuited for sockets-based applications  ...  PVFS supports a set of feature-rich interfaces, including support for both contiguous and noncontiguous accesses to both memory and files.  ... 
doi:10.1109/clustr.2005.347068 dblp:conf/cluster/BalajiFGNYP05 fatcat:jxxjo3t42rds3nvtktzrow6fei

Mercury: Enabling remote procedure call for high-performance computing

Jerome Soumagne, Dries Kimpe, Judicael Zounmevo, Mohamad Chaarawi, Quincey Koziol, Ahmad Afsahi, Robert Ross
2013 2013 IEEE International Conference on Cluster Computing (CLUSTER)  
In addition, existing RPC frameworks often do not support handling large data arguments, such as those found in read or write calls.  ...  Additionally, the network implementation is abstracted, allowing easy porting to future systems and efficient use of existing native transport mechanisms.  ...  ACKNOWLEDGMENTS The work presented in this paper was supported by the We gratefully acknowledge the computing resources provided on "Fusion", a 320-node computing cluster operated by the Laboratory Computing  ... 
doi:10.1109/cluster.2013.6702617 dblp:conf/cluster/SoumagneKZCKAR13 fatcat:gxokjqh45zfmjj4pgazzcdzi64

Bridging the Ethernet-Ethernot Performance Gap

P. Balaji, Wu-chun Feng, D.K. Panda
2006 IEEE Micro  
PVFS also supports a set of feature-rich interfaces, including support for both contiguous and noncontiguous accesses to both memory and files.  ...  This is likely due to the MPI-Tile-I/O benchmark's noncontiguous data-access pattern, which adds significant overhead.  ... 
doi:10.1109/mm.2006.48 fatcat:fepv7mvg4vfepdfgcu5tahjskm

Demotion-based exclusive caching through demote buffering

Jiesheng Wu, Pete Wyckoff, Dhabaleswar K. Panda
2003 Proceedings of the international workshop on Storage network architecture and parallel I/Os - SNAPI '03  
In this paper, we propose a DEMOTE buffering mechanism over storage networks to reduce the visible costs of DEMOTE operations and provides more flexibility for optimizations.  ...  Multi-level buffer cache architecture has been widely deployed in today's multiple-tier computing environments. However, caches in different levels are inclusive.  ...  This work was supported in part by Sandia National Laboratory's contract #30505, Department of Energy's Grant #DE-FC02-01ER25506, and National Science Foundation's grants #EIA-9986052 and #CCR-0204429.  ... 
doi:10.1145/1162618.1162627 fatcat:25cda2vtnnc3xmzhkqxa2573om

I/O performance challenges at leadership scale

Samuel Lang, Philip Carns, Robert Latham, Robert Ross, Kevin Harms, William Allcock
2009 Proceedings of the Conference on High Performance Computing Networking, Storage and Analysis - SC '09  
Listed in the top 5 fastest supercomputers of 2008, Intrepid runs computational science applications with intensive demands on the I/O system.  ...  In this paper we present a case study of the I/O challenges to performance and scalability on Intrepid, the IBM Blue Gene/P system at the Argonne Leadership Computing Facility.  ...  Future work will focus on characterizing system performance over an extended period of time.  ... 
doi:10.1145/1654059.1654100 dblp:conf/sc/LangCLRHA09 fatcat:ztzndu4sivgbvctctnbkcstnaa

Can MPI Be Used for Persistent Parallel Services? [chapter]

Robert Latham, Robert Ross, Rajeev Thakur
2006 Lecture Notes in Computer Science  
We also ran experiments to determine the gaps between what the MPI Standard enables and what MPI implementations currently support.  ...  The results of our study indicate that MPI can enable persistent parallel systems to be developed with less effort and can provide high performance, but MPI implementations will need to provide better support  ...  over Myrinet, the native InfiniBand protocol over InfiniBand.  ... 
doi:10.1007/11846802_40 fatcat:ejg6zllsizhjxczhl4ec5hadii

High performance support of parallel virtual file system (PVFS2) over Quadrics

Weikuan Yu, Shuang Liang, Dhabaleswar K. Panda
2005 Proceedings of the 19th annual international conference on Supercomputing - ICS '05  
In this paper, we explore the challenges of supporting parallel file system with modern features of Quadrics, including user-level communication and RDMA operations.  ...  To the best of our knowledge, this is the first work in the literature to report the design of a high performance parallel file system over Quadrics user-level communication protocols.  ...  Furthermore, We also would like to thank Drs Daniel Kidger and David Addison from Quadrics, Inc for their valuable technical support.  ... 
doi:10.1145/1088149.1088192 dblp:conf/ics/YuLP05 fatcat:mpv3r6hmjzd4ben2zioklh2zs4
« Previous Showing results 1 — 15 out of 43 results