Data Structures

Joffre Heredia joffre at yogi.uab.es
Fri Jul 5 12:22:45 PDT 2002


 Hi all,

  I would like to know why well-known data structures such as linked
lists, binary trees, and other kinds of intelligent forms of storing data in main
memory aren't usually used in clusters or parallel programming.

 I'm telling this because I've been examining some MD programs in search
of such data structures and I can't find any of them. This could be
because there is no need for these structures, but I doubt it since these programs
are usually managing huge amounts of information like atoms, etc...

 I was reading somewhere that parallel programmers don't like pointers,
what is in fact esential for dynamic data structures, because they won't
be able to predict the use of memory, leading to a loss of eficiency.

 After reading some information about MPI programming, I actually think
that it could be possible to implement data structures for these parts of
the parallel calculation that are running localy in a node. Maybe I'm
wrong. Could anyone help ?

 Thank you in advance.

-------------------------------------------------------------
Joffre Heredia Rodrigo                  Tel: (34)-93-5813812
Laboratory of Computational Medicine    Fax: (34)-93-5812344
Biostatistic Dept.
UAB School of Medicine. Bellaterra      Joffre.Heredia at uab.es
08193-Barcelona (SPAIN)
-------------------------------------------------------------




More information about the Beowulf mailing list