chap07_3 - Prof Thriveni T K Advanced Computer Architecture...

Info iconThis preview shows pages 1–10. Sign up to view the full content.

View Full Document Right Arrow Icon
Prof Thriveni T K Advanced Computer Architecture Hwang, Chapter 7 Multiprocessors and Multicomputers 7.4 Message Passing Mechanisms
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Message Passing in Multicomputers Multicomputers have no shared memory, and each “computer” consists of a single processor, cache, private memory, and I/O devices. Some “network” must be provided to allow the multiple computers to communicate. The communication between computers in a multicomputer is called “message passing.”
Background image of page 2
Message Formats Messages may be fixed or variable length. Messages are comprised of one or more packets. Packets are the basic units containing a destination address (e.g. processor number) for routing purposes. Different packets may arrive at the destination asynchronously, so they are sequence numbered to allow reassembly. Flits (flow control digits) are used in
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Store and Forward Routing Packets are the basic unit in the store and forward scheme. An intermediate node must receive a complete packet before it can be forwarded to the next node or the final destination, and only then if the output channel is free and the next node has available buffer space for the packet. The latency in store and format networks is directly related to the number of
Background image of page 4
Flits and Wormhole Routing Wormhole routing divides a packet into smaller fixed-sized pieces called flits (flow control digits). The first flit in the packet must contain (at least) the destination address. Thus the size of a flit must be at least log 2 N in an N - processor multicomputer. Each flit is transmitted as a separate entity, but all flits belonging to a single packet must be transmitted in sequence, one
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Store and Forward vs. Wormhole
Background image of page 6
Asynchronous Pipelining Each intermediate node in a wormhole network, and the source and destination, each have a buffer capable of storing a flit. Adjacent nodes communicate requests and acknowledgements using a one-bit ready/request (R/A) line. When a receiver is ready, it pulls the R/A line low. When the sender is ready, it raises the R/A line high and transmits the next flit; the line is left high. After the receiver deals with the flit (perhaps sending it on to another node), it lowers the R/A line to indicate it is ready to accept another flit. The cycle repeats for transmission of other flits.
Background image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Wormhole Node Handshaking
Background image of page 8
Asynchronous Pipeline Speeds An asynchronous pipeline can be very efficient, and use a clock speed higher than that used in a synchronous pipeline.
Background image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 10
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 40

chap07_3 - Prof Thriveni T K Advanced Computer Architecture...

This preview shows document pages 1 - 10. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online