Efficient block device sharing over myrinet with memory bypass

Koukis, E; Koziris, N

dc.contributor.author	Koukis, E	en
dc.contributor.author	Koziris, N	en
dc.date.accessioned	2014-03-01T02:44:35Z
dc.date.available	2014-03-01T02:44:35Z
dc.date.issued	2007	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/31889
dc.subject	Experimental Evaluation	en
dc.subject	Interconnection Network	en
dc.subject	Virtual Memory	en
dc.subject	Client Server	en
dc.subject.other	Bandwidth	en
dc.subject.other	Data storage equipment	en
dc.subject.other	Data transfer	en
dc.subject.other	Optimization	en
dc.subject.other	Memory bus	en
dc.subject.other	Memory-intensive applications	en
dc.subject.other	Parallel filesystem	en
dc.subject.other	Interconnection networks	en
dc.title	Efficient block device sharing over myrinet with memory bypass	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1109/IPDPS.2007.370219	en
heal.identifier.secondary	4227947	en
heal.identifier.secondary	http://dx.doi.org/10.1109/IPDPS.2007.370219	en
heal.publicationDate	2007	en
heal.abstract	Efficient sharing of block devices over an interconnection network is an important step in deploying a shared-disk parallel filesystem on a cluster of SMPs. In this paper we present gmbock, a client/server system for network sharing of storage devices over Myrinet, which uses an optimized data path in order to transfer data directly from the storage medium to the NIC, bypassing the host CPU and main memory bus. Its design enhances existing programming abstractions, combining the user level networking characteristics of Myrinet with Linux's virtual memory infrastructure, in order to construct the datapath in a way that is independent of the type of block device used. Experimental evaluation of a prototype system shows that remote I/O bandwidth can improve up to 36.5%, compared to an RDMA-based implementation. Moreover, interference on the main memory bus of the host is minimized, leading to an up to 41% improvement in the execution time of memory-intensive applications. © 2007 IEEE.	en
heal.journalName	Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM	en
dc.identifier.doi	10.1109/IPDPS.2007.370219	en