Matthias Boehm created SYSTEMML-2147:

             Summary: Performance read of short and wide sparse matrices
                 Key: SYSTEMML-2147
             Project: SystemML
          Issue Type: Bug
            Reporter: Matthias Boehm

Since for sparse matrix reads of many columns we synchronize over the first row 
of a row of blocks, there is unnecessary contention with many read threads and 
a small number of rows (e.g., 10K rows -> 10 blocks, but 24 threads reading)

This message was sent by Atlassian JIRA

Reply via email to