1 * segfault with ./loba cluster1000.xml -N64 -a fairstrategy
3 this is a deadlock occuring when:
4 - a process is in the finalize stage;
5 - all processes but one are blocked on receive;
6 - the process that is still running owns all the remaining load,
7 and sends it all to the finalizing process, and then goes in
9 The finalizing process receives the load, and blocks again,
10 waiting for a close message.
11 All processes are then blocked, and non-one is able to see that
12 there is no more load in the system!
14 * verify bookkeeping version.
17 - flops (total, avg, stddev)
18 - message exchanges (number, volume)
20 * add options -j/-J : minimum number of iterations ?
22 * add a variant to (not) change neighbor load information at send.
24 * implement loba_* algorithms (start with some trivial one)
26 * for automatic process topology,
27 -> implement some random initial distribution of load
29 * add synchronized mode
31 * translate README file?