Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
On the model of computation: counterpoint: parallel programming wall and multicore software spiral: denial hence crisis
Vishkin U. Communications of the ACM65 (9):32-34,2022.Type:Article
Date Reviewed: Jan 24 2023

Vishkin’s counterpoint to Dally [1] talks about what the model of computation should now be in light of multicore processors (and their programming difficulties) and parallel processing. His view recommends a new model of computation, other than RAM and PRAM, that considers the communication time to communicate with global memory in addition to the processing time in the central processing unit (CPU).

This counterpoint still favors RAM as the best model--it is simple; it considers all computations sequential; it is mathematically sound as it is based on mathematical induction. Furthermore, multicore programming has a dearth of programmers because programming such systems is simply too difficult. The current exploitation of parallelism supported by the new model for general-purpose application performance falls far behind the single-core CPU that supports the RAM model, which played a key role in easy programming for serial computing and naturally paved the path to PRAM models. By allowing us to reason about the true costs of computation, the parallel explicit communication model (PECM) will allow us to design more efficient computations. However, when PECM is compared with PRAM, the cost-effectiveness appeal for programmers will again tilt the scales to PRAM; the PECM reliance on specialized efficiencies will make the systems even more brittle.

Another fact: not much software is explicitly parallel processing on multicores; even software written on earlier multicores needs to be retuned to run on new multicores. The original rationale of PRAM was to execute an unlimited number of serial operations of RAM in a single step, but remain as simple as RAM.

Overall, the counterpoint view makes a strong case. It favors the RAM and PRAM models against any newer models--at least for the time being, until some universal model appears that can take care of all the energy complexity types.

Reviewer:  K R Chowdhary Review #: CR147541 (2303-0031)
1) Dally, W. On the model of computation: point: we must extend our model of computation to account for cost and location. Communications of the ACM 65(9), Sept. 2022, 30-32.
Bookmark and Share
  Featured Reviewer  
 
General (D.1.0 )
 
Would you recommend this review?
yes
no
Other reviews under "General": Date
Problems in programming
Vitek A., Tvrdy I., Reinhardt R., Mohar B. (ed), Martinec M., Dolenc T., Batagelj V. (ed), John Wiley & Sons, Inc., New York, NY, 1991. Type: Book (9780471930174)
Aug 1 1992
KNOs: KNowledge acquisition, dissemination, and manipulation Objects
Tsichritzis D., Fiume E., Gibbs S., Nierstrasz O. ACM Transactions on Information Systems 5(1): 96-112, 1987. Type: Article
Nov 1 1987
Programmer perceptions of productivity and programming tools
Hanson S. (ed), Rosinski R. Communications of the ACM 28(2): 180-189, 1985. Type: Article
Jul 1 1985
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy