Message passing in pvm, Computer Networking

Message Passing

The PVM communication model gives asynchronous blocking receive, asynchronous blocking send and nonblocking receive functions. In our methodology, a blocking send returns as early as the send buffer is free for use again, and an asynchronous send does not depend on the receiver calling a matching receive before the send can return. There are options in PVM 3 that request that data be transmitted directly from task to task. In this case, if the message is big, the sender may block until the receiver has called a matching receive.

A nonblocking receive instantaneously returns with both the data and a flag that the data has not arrived, whereas a blocking receive returns only when the data is in the receive buffer. In adding to these point-to-point communication functions, the model supports the multicast to a place of tasks and the broadcast to a user-defined group of tasks. There are also functions to present global max, global sum, etc. across a user-defined set of tasks. Wildcards can be specified in the receive for the label and the source, allowing either or both of these contexts to be ignored. A routine can be called to return the information about the received messages.

The PVM model assuring that the message order is conserved. If task 1 sends message A to task 2, then after task 1 sends message B to task 2, message A will arrive at task 2 earlier than message B. Furthermore, if both the messages arrive earlier than task 2 does a receive, then a wildcard receive will the entire time return message A.

  • int bufid = pvm_mkbuf( int encoding )

Generates a new message buffer. encoding states the buffer's encoding scheme.

Posted Date: 3/2/2013 7:23:29 AM | Location : United States

Related Discussions:- Message passing in pvm, Assignment Help, Ask Question on Message passing in pvm, Get Answer, Expert's Help, Message passing in pvm Discussions

Write discussion on Message passing in pvm
Your posts are moderated
Related Questions
What is use of Carrier sense multiple access Carrier sense multiple access with collision detection (CSMA/CD) is a form of medium access control in which a station listens to

Selective Repeat ARQ - Processing at the receiver additionally complex - The window size is reduced to 2(m-1) (2m/2 at most) - Both the transmitter as well as the receiv

FDDI has four specifications: 1. Media Access Control it defines how the medium is accessed 2. Physical Layer Protocol it defines data encoding/decoding procedures 3. Phy

- EIA/ TIA specifies and RJ-45 connector for unshielded twisted-pair (UTP) cable. The letter RJ stand for registered jack and the number 45 refers to a specific wiring sequence.

What is the difference among  flow control and error control?    Ans) Flow control: adjust and confirm data flow rate for successful transmission. Error Control: a method to

What is the difference between a hub and a switch? A hub acts as a multiport repeater. Though, as more and more devices connect to it, it would not be able to efficiently manag

identify different network topologies.

Dimensionality of Interconnection Network Dimensionality refers the arrangement of processing elements or nodes in an interconnection network. In linear network or single dimen

Parallel Programming Environment Characteristics  The parallel programming environment consists of an debugger, a editor,  performance evaluator and programme visualizer for i