Multi-Layer Lambda Grid With Exact Bandwidth Provisioning Over Converged IP and Optical Networks

This paper proposes for the first time a multi-layer Lambda Grid, which is a platform to establish dynamically a computing environment with a guaranteed level of throughput and bandwidth over converged IP and optical networks according to each client reservation request. It enables client-side visualization of a computing environment comprising converged IP and optical networks and geographically distributed computing domains. In addition, it dynamically reserves a packet switched path/tunnel (sub-lambda) with the exact requested bandwidth and computer resources according to each client reservation request. Moreover, it links the reserved path to the reserved computers, controls the reserved sub-lambda and the relevant lambda as needed, and routes only the reserved packet flow between the reserved computers to the reserved sub-lambda. In terms of cost performance and bandwidth flexibility, it can use network resources more effectively than existing Lambda Grids. Through the multi-layer Lambda Grid, clients can execute applications in an established computing environment as if the environment was a virtual private computing environment for clients. To achieve the multi-layer Lambda Grid, this paper also proposes a novel technology. This technology dynamically reserves a sub-lambda with the exact requested bandwidth according to each client reservation request. Moreover, it links the reserved sub-lambda to the reserved computers, and controls the reserved sub-lambda and the relevant lambda as needed. In addition, it routes only the reserved packet flow between reserved computers to the reserved sub-lambda. To evaluate the feasibility of the technology, a common computing environment with the multi-layer Lambda Grid for scientific calculation, file transfer, and high-definition video streaming services is presented in an actual field environment. Through this experiment, the total feasibility of the multi-layer Lambda Grid is shown.

[1]  Masahiko Jinno,et al.  G-lambda and EnLIGHTened: wrapped in middleware co-allocating compute and network resources across Japan and the US , 2007, GridNets '07.

[2]  Ian T. Foster,et al.  The anatomy of the grid: enabling scalable virtual organizations , 2001, Proceedings First IEEE/ACM International Symposium on Cluster Computing and the Grid.

[3]  Yoshio Tanaka,et al.  GridARS: An Advance Reservation-Based Grid Co-allocation Framework for Distributed Computing and Network Resources , 2007, JSSPP.

[4]  M. Jinno,et al.  Multi-layer lambda Grid properly using lambdas and sub-lambdas , 2008, 2008 34th European Conference on Optical Communication.

[5]  Hisao Uose.  GEMnet2: NTT's new network testbed for global R&D , 2005, First International Conference on Testbeds and Research Infrastructures for the DEvelopment of NeTworks and COMmunities.

[6]  Tetsuro Fujii,et al.  4K SHD Real-Time Video Streaming System With JPEG 2000 Parallel Codec , 2006, APCCAS 2006 - 2006 IEEE Asia Pacific Conference on Circuits and Systems.

[7]  Michiaki Hayashi,et al.  G-lambda: Coordination of a Grid Scheduler and Lambda Path Service over GMPLS , 2006, 2006 European Conference on Optical Communications.

[8]  Doug Johnson,et al.  Computing in the Clouds. , 2010 .

[9]  Priya Vashishta,et al.  Sustainable Adaptive Grid Supercomputing: Multiscale Simulation of Semiconductor Processing across the Pacific , 2006, ACM/IEEE SC 2006 Conference (SC'06).

[10]  Cees T. A. M. de Laat,et al.  TransLight: a global-scale LambdaGrid for e-science , 2003, CACM.

[11]  Masahiko Jinno,et al.  The first functional demonstration of optical virtual concatenation as a technique for achieving Terabit networking , 2006, Future generations computer systems.

[12]  H. Tanaka,et al.  Advance reservation-based network resource manager with adaptive path discovery scheme for SOA-based networking , 2007, OFC/NFOEC 2007 - 2007 Conference on Optical Fiber Communication and the National Fiber Optic Engineers Conference.