Hey,
I think you can consider Hadoop as a data processing system which uses
the Map-Reduce paradigm to process data. There is a project called
"Hama" (I think) which supports some linear algebra operations on
Hadoop. However, it will only be effective for a subset of linear
algebra operations.
Hadoop is not a general purpose clustering software, and it cannot run
linpack. If you want to do large-scale linear algebra operations
(without having to write FORTRAN-style code in C/C++), I would
recommend investigating PETSc.
Brian
On Dec 18, 2008, at 6:47 AM, Leeau wrote:
Dear,
I am thinking that how to know the performance of hadoop. As a kind of
cluster, can we run linpack in hadoop? and how can we do that?
thanks!
