OPIOM: off-processor I/O with myrinet
Future Generation Computer Systems - Best papers from symp. on cluster computing and the grid (CCGRID 2001)
MPI-IO/GPFS, an optimized implementation of MPI-IO on top of GPFS
Proceedings of the 2001 ACM/IEEE conference on Supercomputing
GPFS: A Shared-Disk File System for Large Computing Clusters
FAST '02 Proceedings of the Conference on File and Storage Technologies
GNBD/VIA: A Network Block Device over Virtual Interface Architecture on Linux
IPDPS '02 Proceedings of the 16th International Parallel and Distributed Processing Symposium
Cache Fusion: Extending Shared-Disk Clusters with Shared Caches
Proceedings of the 27th International Conference on Very Large Data Bases
READ2: Put disks at network level
CCGRID '03 Proceedings of the 3st International Symposium on Cluster Computing and the Grid
Using Idle Disks in a Cluster as a High-Performance Storage System
CLUSTER '02 Proceedings of the IEEE International Conference on Cluster Computing
The Quadrics Network (QsNet): High-Performance Clustering Technology
HOTI '01 Proceedings of the The Ninth Symposium on High Performance Interconnects
Flexible Network Attached Storage Using Remote DMA
HOTI '01 Proceedings of the The Ninth Symposium on High Performance Interconnects
Memory and Network Bandwidth Aware Scheduling of Multiprogrammed Workloads on Clusters of SMPs
ICPADS '06 Proceedings of the 12th International Conference on Parallel and Distributed Systems - Volume 1
High Performance Block I/O for Global File System (GFS) with InfiniBand RDMA
ICPP '06 Proceedings of the 2006 International Conference on Parallel Processing
Efficient remote block-level I/O over an RDMA-capable NIC
Proceedings of the 20th annual international conference on Supercomputing
Optimization and bottleneck analysis of network block I/O in commodity storage systems
Proceedings of the 21st annual international conference on Supercomputing
Implementation and Evaluation of an MPI-IO Interface for GPFS in ROMIO
Proceedings of the 15th European PVM/MPI Users' Group Meeting on Recent Advances in Parallel Virtual Machine and Message Passing Interface
Euro-Par 2010 Proceedings of the 2010 conference on Parallel processing
Hi-index | 0.00 |
We present gmblock, a block-level storage sharing system over Myrinet which uses an optimized I/O path to transfer data directly between the storage medium and the network, bypassing the host CPU and main memory bus of the storage server. It is device driver independent and retains the protection and isolation features of the OS. We evaluate the performance of a prototype gmblock server and find that: (a) the proposed techniques eliminate memory and peripheral bus contention, increasing remote I/O bandwidth significantly, in the order of 20---200% compared to an RDMA-based approach, (b) the impact of remote I/O to local computation becomes negligible, (c) the performance characteristics of RAID storage combined with limited NIC resources reduce performance. We introduce synchronized send operations to improve the degree of disk to network I/O overlapping. We deploy the OCFS2 shared-disk filesystem over gmblock and show gains for various application benchmarks, provided I/O scheduling can eliminate the disk bottleneck due to concurrent access.