Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.programming.threads > #3889 > unrolled thread

My scalable Threadpool is coming soon..

Started byIntelli2 <intelli2@mama.com>
First post2017-10-08 18:39 -0400
Last post2017-10-08 18:39 -0400
Articles 1 — 1 participant

Back to article view | Back to comp.programming.threads


Contents

  My scalable Threadpool is coming soon.. Intelli2 <intelli2@mama.com> - 2017-10-08 18:39 -0400

#3889 — My scalable Threadpool is coming soon..

FromIntelli2 <intelli2@mama.com>
Date2017-10-08 18:39 -0400
SubjectMy scalable Threadpool is coming soon..
Message-ID<ore9eq$gtt$2@dont-email.me>
Hello..


My scalable Threadpool is coming soon..

I have found the following paper about Scalable Concurrent Counting:

http://people.csail.mit.edu/shanir/publications/HLS.pdf

Notice on the graph that the Counting network on shared-memory is truly 
scalable, i am currently implementing counting network in
Delphi and FreePascal, and i will soon incorporate it  in my
following more scalable Threadpool:

https://sites.google.com/site/aminer68/an-efficient-threadpool-engine-that-scales-well

So this counting network it will give a scalable Threadpool that will
be really interesting, and work-stealing of the Threadpool will use a 
counting network and the producer side of the Threadpool will use also a 
counting network and my Threadpool uses a technic called lock=striping, 
so it will become a scalable Threadpool, so stay tuned i am actually 
finishing it.


Also, I am currently finishing a second software project that is an MPMC 
FIFO queue that is wait-free on the producer side and lockfree on the 
consumer side and that uses Hazard Pointers Safe Memory Reclamation on 
the consumer side of the FIFO queue. This is a more professional and 
good quality piece of work that will be finished soon.

My other project that is coming soon is:

My next Parallel compression library is coming, it will work under 
Windows and Linux and it will be more optimized, there will be only two 
threads that do the IO (and they are not contending) so that it reduces 
at best the contention, and it will be more optimized for NUMA and it 
will support processor groups on windows etc. it will support also 
Parallel LZMA and Parallel Zstd and Parallel LZ4 compression and 
decompression, and it will support parallel AES encryption with 256-bit 
key etc. This is a more professional and good quality piece of work that 
will be finished soon.

My Parallel archiver is already optimized for NUMA and it supports 
processor groups on windows and it has only two threads that do the IO 
(and they are not contending) so that it reduces at best the contention 
etc. please look at my new Parallel archiver.

You can download my software projects from:

https://sites.google.com/site/aminer68/

Thank you,
Amine Moulay Ramdane.

[toc] | [standalone]


Back to top | Article view | comp.programming.threads


csiph-web