Intrinsic functions, Computer Networking

Intrinsic Functions

HPF introduces some new intrinsic functions in addition to those defined in F90. The two most regularly used in parallel programming are the system inquiry functions PROCESSORS_SHAPE and NUMBER_OF_PROCESSORS  . These functions gives the information about the number of physical processors on which the running program implements and processor configuration. General syntax of is

NUMBER_OF_PROCESSORS is

NUMBER_OF_PROCESSORS (dim)

where dim is an optional argument. It returns the number of processors in the underlying array or, if the optional argument is present, the size of this array along a shows dimension.

General syntax of  PROCESSORS_SHAPE is

PROCESSORS_SHAPE()

It returns an one-dimensional array, whose  ith element provides the size of the underlying processor array in its ith dimension.

 Example 10

Consider the call of the two intrinsic functions discussed above for a 32-Processor (4×8) Multicomputer:

The function call NUMBER_OF_PROCESORS () will return 32. The function call NUMBER_OF_PROCESORS (1) will return 4. The function call NUMBER_OF_PROCESORS (2) will return 8.

The function call PROCESSORS_SHAPE () will return an array with two elements 4 and 8.

We can use these intrinsic functions in tandem with array declarations and HPF directives, to give flexibility to the programmer to declare abstract processor arrays that match available physical resources. For example, the following statement! HPF$ PROCESSORS P(NUMBER_OF_PROCESSORS()) declares an abstract processor array P with size equivalent to the number of physical processors.

Posted Date: 3/2/2013 8:13:56 AM | Location : United States







Related Discussions:- Intrinsic functions, Assignment Help, Ask Question on Intrinsic functions, Get Answer, Expert's Help, Intrinsic functions Discussions

Write discussion on Intrinsic functions
Your posts are moderated
Related Questions
A modified NRZ code known as enhanced-NRZ (E-NRZ) operates on 7-bit words; inverting bits 2,3,6 and 7; and adding one parity bit to each word. The parity bit is chosen to make the

What is difference between Baseband and Broadband Transmission in ccna? Ans) In a baseband transmission, the entire bandwidth of the cable is consumed by a one signal. In bro

what's the difference among Symmetric and Asymmetric cryptosystem?


Design the user interface for mode switching and data entry any way that you like. For example, to switch an application from transmit mode to receive mode, you can give it special

Q. What are the Routing Techniques? Routing Routing Techniques Static versus Dynamic Routing Routing Table for classful Addressing Routing Table for Cl

Measure the performance (both latency and throughput) of  TCP and UDP protocol stack  running on a pair of PCs  or Workstations in the Lab, and compare the results of your software

On a lan where are ip datagram transported?

What is MAN? MAN - Metropolitan Area Networks. MAN is bigger than a LAN and as its name implies, covers the area of a single city. MANs rarely extend beyond 100 KM and frequ

Inside a device, every address mask is stored as a 32-bit number. When we submit a prefix and an address mask they use a changed form of dotted decimal addressing known CIDR addres