[prev in list] [next in list] [prev in thread] [next in thread] 

List:       openjdk-lambda-dev
Subject:    Parallel decompositions, C/P/N/Q experiment, take 2
From:       aleksey.shipilev () oracle ! com (Aleksey Shipilev)
Date:       2012-09-27 13:03:58
Message-ID: 50644EBE.3090104 () oracle ! com
[Download RAW message or body]

On 09/24/2012 05:26 PM, Aleksey Shipilev wrote:
> This is the second take on the experiment I've did couple of months ago
> [1]. tl;dr version: hand-crafted generator for longs in
> (0; N], simple filter (with variable cost Q) to empty sink, called by C
> clients, stream operations services by (fj)pool of size P.
> 
> The question to answer: how would performance change with juggling
> C/P/N/Q, in both sequential and parallel modes?

If anyone here cares, the updated data for 2x8x2 Xeon E5-2680
(SandyBridge) running Solaris 11, and latest lambda nightly with -d64
-XX:-TieredCompilation -XX:+UseParallelOldGC -XX:+UseNUMA
-XX:-UseBiasedLocking -XX:+UseCondCardMark, is here:
  http://shipilev.net/pub/jdk/lambda/bulk-fuzzy-20120925-snb-sol11/

It is somewhat similar to what we see with old RHEL running on Nehalem.

-Aleksey.


[prev in list] [next in list] [prev in thread] [next in thread]