Mon Jan 23 18:33:07 HKT 2017
Stealing thread - http://badamczewski.blogspot.com.au/2012/05/work-stealing.html
Intel Guide for Developing Multithreaded Applications - http://software.intel.com[..]or-developing-multithreaded-applications
Difference ways to stop a thread - http://www.ddj.com[..]ept_url=/hpc-high-performance-computing/
Interesting, I am not sure if I agree, but chrome ( which use fork ) are really cool in performance:
There’s another problem with Unix programming in Ruby that I’ll just touch on briefly: Java people and Windows people. They’re going to tell you that fork(2) is bad because they don’t have it on their platform, or it sucks on their platform, or whatever, but it’s cool, you know, because they have native threads, and threads are like, way better anyways.
Don’t ever let anyone tell you that fork(2) is bad. Thirty years from now, there will still be a fork(2) and a pipe(2) and a exec(2) and smart people will still be using them to solve hard problems reliably and predictably, just like they were thirty years ago.
MRI Ruby people need to accept, like Python (you have seen multiprocessing, yes?), that Unix processes are one of two techniques for achieving reliable concurrency and parallelism in server applications. Threads are out. You can use processes, or async/events, or both processes and async/events, but definitely not threads. Threads are out. http://tomayko.com/writings/unicorn-is-unix
1x1 win M*N - http://binkley.blogspot.com/2012/01/1-1-beats-n-m.html
Best number of threads:
N = number of CPUs
U = target CPU utilization (0 <= U <= 1)
W/C = ration of wait time to cpu time (measured through profiling)
threads = N * U * (1 + W/C) - http://www.ibm.com/developerworks/library/j-jtp0730.html http://stackoverflow.com[..]ratio-of-java-threads-to-cpus-on-solaris
Another post about tuning thread pool - http://www.javaadvent.com[..]ortance-of-tuning-your-thread-pools.html
Threads Basics - http://www.hpl.hp.com/techreports/2009/HPL-2009-259html.html http://www.hpl.hp.com/personal/Hans_Boehm/c++mm/threadsintro.html
The Dos and Don'ts of Multithreading - http://www.infoq.com/presentations/multithreading https://www.infoq.com/articles/engstrand-microservice-threading http://cbloomrants.blogspot.hk[..]-26-09-low-level-threading-table-of.html
Tue Nov 15 10:49:25 HKT 2016
How to write safer concurrency code - http://www.artima.com/forums/flat.jsp?forum=276&thread=178345
reentrant and thread safe functions - http://kevinrodrigues.com[..]/31/reentrant-and-thread-safe-functions/
Libraries / toolkits for multicore process - http://www.ddj.com[..]intableArticle.jhtml?articleID=212900103
Introduction - http://www.ddj.com[..]QQSNDLRSKHSCJUNN2JVN?articleID=212903586 http://www.ddj.com[..]CQSNDLRSKHSCJUNN2JVN?articleID=213001517
Collections of links - http://dobbscodetalk.com[..]rallel-Or-Get-Left-Behind.html&Itemid=29
Briefing of difference modeling of threading system - http://www.ddj.com[..]intableArticle.jhtml?articleID=215900465 http://software.intel.com[..]inners-guide-to-multithreaded-libraries/ http://natishalom.typepad.com[..]haloms_blog/2010/08/concurrency-101.html
Saving the Failwhale: The Art of Concurrency (Page last updated December 2012, Added 2012-12-26, Author Dhanji R. Prasanna, Publisher informit). Tips:
1) Contention is unavoidable - some resources are just slower, and you must wait for them. The secrets to good concurrency are 1) ensuring that these slower resources are rarely used, and 2) during such waiting periods, giving the faster tiers other work to do so that they continue to be utilized well.
2) Overuse of synchronization constructs such as locks and mutexes leads to systems that perform poorly under load.
3) ConcurrentHashMap is an efficient thread-safe map while HashMap is not thread-safe.
4) ConcurrentHashMap doesn't do away with locks, it still uses them but it uses more than the single global lock, so that threads gain some measure of concurrency. It uses separate locks for partitions, so that multiple threads writing to the map are likely to access different partitions, using separate locks and therefore process their data simultaneously. This technique is known as lock-striping. Efficient striping uses a number of locks proportional to the number of CPU cores in a system.
5) The asynchronous processing model smooths resource spikes by adding requests to a queue which is serviced by a pool of workers - spikes in requests make the queue grow rather than overloading the workers. (The ExecutorService is essentially a thread pool accompanied by a task queue.) http://www.informit.com/articles/article.aspx?p=1994789
Discussion of using difference model for concurrency - http://highscalability.com[..]cks-application-architecture-pros-a.html
Concurrency vs Parallelism - http://concurrencyfreaks.blogspot.hk[..]/2013/12/concurrency-vs-parallelism.html
Compare between Actors, CSP, Disruptor and raw Threads - http://java-is-the-new-c.blogspot.com.au[..]omparision-of-different-concurrency.html
Few coding tips that should be useful for most languages - http://www.javacodegeeks.com[..]erformance-scalability-and-liveness.html http://www.javacodegeeks.com[..]2015/09/java-concurrency-essentials.html
Do one thing, do it well
No shared operational state
Bound your queues
Name custom thread pools and register an UncaughtExceptionHandler
Prefer immutable data objects over mutable statehttp://highscalability.com[..]-to-25-billion-notifications-during.html
Thu Sep 13 02:01:25 HKT 2007
Someone say the basic idea behind transactional memory is that memory is shared by all the threads in a read-only manner. Threads modify data by "committing" objects to shared memory. If a thread commits on top of a dirty object then it has to "rollback" and retry.http://research.sun.com[..]007/2007-08-13_transactional_memory.html http://www.hpcwire.com/hpc/1196095.html