Changes between Version 73 and Version 74 of DataParallel/BenchmarkStatus

Timestamp:

12/02/10 20:25:26 (2 years ago)

Author:

benl

Comment:

--

DataParallel/BenchmarkStatus

@@ -181 +181 @@
 These programs also use user defined algebraic data types. Vectorization of these programs is still a work in progress.
 
- [http://darcs.haskell.org/libraries/dph/dph-examples/real/NBody/ BarnesHut] '''(BROKEN)'''::
+ [http://darcs.haskell.org/libraries/dph/dph-examples/real/NBody/ BarnesHut] '''(SLOWLORIS)'''::
   This benchmark implements the Barnes-Hut algorithm to solve the ''n''-body problem in two dimensions. There is a naive O(n^2^) version in the same package.
 
-  '''Status''': Core-lint error due to bug in GHC's rule matcher. If we compile without -dcore-lint it segfaults when run.
+  || '''name''' || '''runtime''' || '''speedup''' || '''efficiency''' || '''notes''' ||
+  || dph.nbody.vector.seq.N4 || 100ms || 1 || || A ||
+  || dph.nbody.vectorised.seq.N4 || 4681ms || ~50x slower   ||  || ||
+  || dph.nbody.vectorised.par.N1 || 2381ms ||  ~25x slower || || ||
+  A : Time stated is end-to-end, not just for the kernel.
+
+  '''Status''':  Compiles, but fusion hasn't worked so is very slow. [[br]]
+  '''!ToDo''': Make the vectorised way give the same output as the vector way. The benchmark setup is a bit different. Fixing this won't cause a 50x speed difference though.
+
 
 [[br]]