Title :
Protocols for wide-area data-intensive applications: Design and performance issues
Author :
Yufei Ren ; Tan Li ; Dantong Yu ; Shudong Jin ; Robertazzi, Thomas ; Tierney, B.L. ; Pouyoul, Eric
Author_Institution :
Dept. of Electr. & Comput. Eng., Stony Brook Univ., Stony Brook, NY, USA
Abstract :
Providing high-speed data transfer is vital to various data-intensive applications. While there have been remarkable technology advances to provide ultra-high-speed network bandwidth, existing protocols and applications may not be able to fully utilize the bare-metal bandwidth due to their inefficient design. We identify the same problem remains in the field of Remote Direct Memory Access (RDMA) networks. RDMA offloads TCP/IP protocols to hardware devices. However, its benefits have not been fully exploited due to the lack of efficient software and application protocols, in particular in wide-area networks. In this paper, we address the design choices to develop such protocols. We describe a protocol implemented as part of a communication middleware. The protocol has its flow control, connection management, and task synchronization. It maximizes the parallelism of RDMA operations. We demonstrate its performance benefit on various local and wide-area testbeds, including the DOE ANI testbed with RoCE links and InfiniBand links.
Keywords :
bandwidth allocation; broadband networks; computer network performance evaluation; electronic data interchange; file organisation; local area networks; middleware; telecommunication congestion control; telecommunication links; transport protocols; wide area networks; DOE ANI testbed; InfiniBand links; RDMA networks; RoCE links; TCP/IP protocols; bare-metal bandwidth; communication middleware; connection management; design issues; hardware devices; high-speed data transfer; local-area testbeds; performance issues; remote direct memory access networks; ultrahigh-speed network bandwidth; wide-area data-intensive application protocols; wide-area testbeds; Bandwidth; Computer architecture; Data transfer; IP networks; Middleware; Protocols; Semantics;
Conference_Titel :
High Performance Computing, Networking, Storage and Analysis (SC), 2012 International Conference for
Conference_Location :
Salt Lake City, UT
Print_ISBN :
978-1-4673-0805-2