{"id":101,"date":"2013-07-03T16:26:51","date_gmt":"2013-07-03T16:26:51","guid":{"rendered":"http:\/\/press3.mcs.anl.gov\/romio\/?p=101"},"modified":"2013-07-03T16:26:51","modified_gmt":"2013-07-03T16:26:51","slug":"large-transfers-in-romio","status":"publish","type":"post","link":"https:\/\/wordpress.cels.anl.gov\/romio\/2013\/07\/03\/large-transfers-in-romio\/","title":{"rendered":"Large transfers in ROMIO"},"content":{"rendered":"

Let’s say you had a fat node with lots of memory, and you wanted to write 2 GiB of data out to a file with MPI-IO.\u00a0 How would you do that?\u00a0 You would naturally look at<\/p>\n

MPI_File_write()<\/pre>\n
or one of its variants. This is the prototype for MPI_File_write_all:<\/p>\n
int MPI_File_write_all(MPI_File fh, const void *buf, int count,\n                       MPI_Datatype datatype, MPI_Status *status)<\/pre>\n
Perfect! Even though C ‘int’ types are 32 bits on most platforms, 2^31 (or anything less than 2147483649) will fit into a signed int. But until recently, when we try to do this in ROMIO, we fail.
\nIt turns out we were, despite ROMIO being 15 years old, using POSIX system calls incorrectly. The write(2)<\/code> system call doesn’t actually have to write out all the data you asked it to. It’s perfectly legal to return success, but only write a few bytes of your request. For most transfers, though, a successful write and a “full write” were the same thing, and so we got by for years without testing for “short writes”. Until recently:<\/p>\n