[GHC] #8158: Replace IO manager's IntMap with a mutable hash table

GHC ghc-devs at haskell.org
Tue Sep 3 21:53:26 CEST 2013


#8158: Replace IO manager's IntMap with a mutable hash table
-------------------------------+-------------------------------------------
        Reporter:  bos         |            Owner:  AndreasVoellmy
            Type:  feature     |           Status:  patch
  request                      |        Milestone:  7.8.1
        Priority:  high        |          Version:  7.7
       Component:              |         Keywords:
  libraries/base               |     Architecture:  Unknown/Multiple
      Resolution:              |       Difficulty:  Easy (less than 1 hour)
Operating System:              |       Blocked By:
  Unknown/Multiple             |  Related Tickets:
 Type of failure:  Other       |
       Test Case:              |
        Blocking:              |
-------------------------------+-------------------------------------------

Comment (by AndreasVoellmy):

 OK, I re-did the measurement of acme-http (pong) scaling with the same
 setup as I used in the Mio paper (same server, same acme parameters
 (-A32m), same test machine and software (weighttp), etc). I've attached
 the graph.  On my server the patch does make a small, but noticeable
 improvement of about 5-10% improvement for values of -N parameter from
 1-4. At higher cores I can't see any difference. @bos: does my measurement
 seem reasonable to you? Are you seeing the same thing?

-- 
Ticket URL: <http://ghc.haskell.org/trac/ghc/ticket/8158#comment:19>
GHC <http://www.haskell.org/ghc/>
The Glasgow Haskell Compiler




More information about the ghc-tickets mailing list