With massively parallel hardware, concurrency is the biggest challenge facing programmers today. It's a terrible programming paradigm. And your Non-blocking algorithms only add complexity to an already difficult problem. For example, I've seen NBAs lead to a priority inversion bug.
Did you even watch his talk? I didn't see anyone arguing against Bret's scaling concerns with threads in yesterday's HN comments.
In my experience, the worse bugs I have encountered are deadlocks and race conditions. Saying just use a “non-blocking algorithm” is no silver bullet to the difficultly of working with threads.
If we don’t come up with a new system, I can’t imagine how hard it will be to efficiently program CPUs when they start coming out with 128, 256, and 512 cores.
Comments
After Bret Victor's inspiring talk yesterday, it's fitting to see a new CPU architecture make it to the top of HN.
Keep it up! Maybe tomorrow we'll see an idea to replace threads and locks.
You can replace threads and locks, if you use non-blocking programming techniques.
http://en.wikipedia.org/wiki/Non-blocking_algorithm
With massively parallel hardware, concurrency is the biggest challenge facing programmers today. It's a terrible programming paradigm. And your Non-blocking algorithms only add complexity to an already difficult problem. For example, I've seen NBAs lead to a priority inversion bug.
Did you even watch his talk? I didn't see anyone arguing against Bret's scaling concerns with threads in yesterday's HN comments.
In my experience, the worse bugs I have encountered are deadlocks and race conditions. Saying just use a “non-blocking algorithm” is no silver bullet to the difficultly of working with threads.
If we don’t come up with a new system, I can’t imagine how hard it will be to efficiently program CPUs when they start coming out with 128, 256, and 512 cores.