Another Word For It Patrick Durusau on Topic Maps and Semantic Diversity

January 5, 2011

Is Parallel Programming Hard, And, If So, What Can You Do About It?

Filed under: Computation,Parallelism,Software — Patrick Durusau @ 12:36 pm

Is Parallel Programming Hard, And, If So, What Can You Do About It? Editor Paul E. McKenney

Kirk Lowery forwarded this link to my attention.

Just skimming the first couple of chapters, I have to say it has some of the most amusing graphics I have seen in any CS book.

Performance, productivity and generality are all goals of parallel programming, as cited by this book.

I have to wonder though if subject recognition tasks, analogous to computer vision, that are inherently parallel.

Doing them in parallel does not make them easier but not doing them in parallel certainly makes them harder.

For example, consider the last time you failed to recognize someone who wasn’t in the location or context where you normally see them.

Do you recognize the context in addition or in parallel to your recognition of the person’s face?

Questions:

  1. What benefits/drawbacks do you see in parallel processing of TMDM instances? (3-5 pages, citations)
  2. How would you design subject identifications for processing in a parallel environment? (3-5 pages, citations)
  3. How would you evaluate the need for parallel processing of subject identifications? (3-5 pages, citations)

No Comments

No comments yet.

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

Powered by WordPress