Eliminating receive livelock in an interrupt-driven kernel
ACM Transactions on Computer Systems (TOCS)
HIP: hybrid interrupt-polling for the network interface
ACM SIGOPS Operating Systems Review
Experiences with VI communication for database storage
ISCA '02 Proceedings of the 29th annual international symposium on Computer architecture
The Virtual Interface Architecture
IEEE Micro
Performance Evaluation of the Quadrics Interconnection Network
Cluster Computing
PCI Express System Architecture
PCI Express System Architecture
High performance support of parallel virtual file system (PVFS2) over Quadrics
Proceedings of the 19th annual international conference on Supercomputing
Efficient remote block-level I/O over an RDMA-capable NIC
Proceedings of the 20th annual international conference on Supercomputing
A performance study of sequential I/O on windows NTTM4
WINSYM'98 Proceedings of the 2nd conference on USENIX Windows NT Symposium - Volume 2
Hi-index | 0.00 |
Building commodity networked storage systems is an important architectural trend; Commodity servers hosting a moderate number of consumer-grade disks and interconnected with a high-performance network are an attractive option for improving storage system scalability and cost-efficiency. However, such systems incur significant overheads and are not able to deliver to applications the available throughput. We examine in detail the sources of overheads in such systems, using a working prototype to quantify the overheads associated with various parts of the I/O protocol. We optimize our base protocol to deal with small requests by batching them at the network level and without any I/O-specific knowledge. We also redesign our protocol stack to allow for asynchronous event processing, in-line, during send-path request processing. These techniques improve performance for a 8-disk SATA RAID0 array from 200 to 290 MBytes/s (45% improvement). Using a ramdisk, peak performance improves from 320 to 474 MBytes/s (48% improvement), which is 72% of the maximum possible throughput in our experimental setup. We also analyze the remaining system bottlenecks, and find that although commodity storage systems have potential for building high-performance I/O subsystems, traditional network and I/O protocols are not fully capable of delivering this potential.