| HORIE, T., ISHIHATA, H., AND IKESAKA, M. Design and implementation of an interconnection network for the AP1000. In Proc. IFIP World Computer Congress (1992), vol. I, Information Processing, pp. 555--561. |
....approach for many of today s distributed memory machines. Specifically, the computation agent explicitly waits for all of the input queues to be NotInMessage before proceeding to the next phase and explicitly forwards the input queues for the next phase. In several other distributed memory systems[17, 15, 1, 14, 20], the communication and computation agents are not as tightly coupled, so the computation agent cannot directly observe and control the input queues. Changing a traditional wormhole router as described in Section 2.2.1 to support the synchronized switch requires a small change to the switching ....
HORIE, T., ISHIHATA, H., AND IKESAKA, M. Design and implementation of an interconnection network for the AP1000. In Proc. IFIP World Computer Congress (1992), vol. I, Information Processing, pp. 555--561.
....distributed memory machines. Specifically, the computation agent explicitly waits for all of the input queues to be NotInMessage before proceeding to the next phase and explicitly forwards the input queues for the next phase. In several other distributed memory systems[LAD 92, Int91, Ada93, HII92, Sni93] the communication and computation agents are not as tightly coupled, so the computation agent cannot directly observe and control the input queues. Changing a traditional wormhole router as described in Section 2.2.1 to support the synchronized switch requires a small change to the ....
T. Horie, H. Ishihata, and M. Ikesaka. Design and implementation of an interconnection network for the AP1000. In Proc. IFIP World Computer Congress, volume I, pages 555--561. Information Processing, 1992.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC