Content deleted Content added
No edit summary |
mNo edit summary |
||
Line 11:
The [[Programming complexity | complexity measure]] of the PEM model is the I/O complexity<ref name=":0">{{Cite journal|last=Arge|first=Lars|last2=Goodrich|first2=Michael T.|last3=Nelson|first3=Michael|last4=Sitchinava|first4=Nodari|date=2008|title=Fundamental parallel algorithms for private-cache chip multiprocessors|journal=Proceedings of the Twentieth Annual Symposium on Parallelism in Algorithms and Architectures - SPAA '08|pages=197|___location=New York, New York, USA|publisher=ACM Press|doi=10.1145/1378533.1378573|isbn=9781595939739}}</ref>, which determines the number of parallel blocks transfers between the main memory and the cache. During a parallel block transfer each processor can transfer a block. So if <math>P</math> processors load parallelly a data block of size <math>B</math> form the main memory into their caches, it is considered as an I/O complexity of <math>O(1)</math> not <math>O(P)</math>. A program in the PEM model should minimize the data transfer between main memory and caches and operate as much as possible on the data in the caches.
=== Read
In the PEM model, there is no [[Computer network | direct communication network]] between the P processors. The processors have to communicate indirectly over the main memory. If multiple processors try to access the same block in main memory concurrently read/write conflicts<ref name=":0">{{Cite journal|last=Arge|first=Lars|last2=Goodrich|first2=Michael T.|last3=Nelson|first3=Michael|last4=Sitchinava|first4=Nodari|date=2008|title=Fundamental parallel algorithms for private-cache chip multiprocessors|journal=Proceedings of the Twentieth Annual Symposium on Parallelism in Algorithms and Architectures - SPAA '08|pages=197|___location=New York, New York, USA|publisher=ACM Press|doi=10.1145/1378533.1378573|isbn=9781595939739}}</ref> occur. Like in the PRAM model, three different variations of this problem are considered:
*Concurrent Read Concurrent Write (CRCW): The same block in main memory can be read and written by multiple processors concurrently.
| |||