grid is 2 x 2 Sequential: 1.65733 secs (10.8608 GFlop/s) Root waited for 1.7025e-05 seconds Populate root node: 0.0192072 secs Spread from root: 0.0596325 secs [MC,* ] AllGather: 0.00312802 secs (491.045 MB/s) for1500 x 128 local matrix [* ,MR] AllGather: 0.00325231 secs (472.28 MB/s) for 128 x 1500 local matrix Local gemm: 0.0624608 secs (9.22178 GFlop/s) for 1500 x 1500 x 128 product [MC,* ] AllGather: 0.00277708 secs (553.1 MB/s) for1500 x 128 local matrix [* ,MR] AllGather: 0.00323079 secs (475.425 MB/s) for 128 x 1500 local matrix Local gemm: 0.054564 secs (10.5564 GFlop/s) for 1500 x 1500 x 128 product [MC,* ] AllGather: 0.00298858 secs (513.957 MB/s) for1500 x 128 local matrix [* ,MR] AllGather: 0.00341181 secs (450.201 MB/s) for 128 x 1500 local matrix Local gemm: 0.0580573 secs (9.92124 GFlop/s) for 1500 x 1500 x 128 product [MC,* ] AllGather: 0.00277618 secs (553.279 MB/s) for1500 x 128 local matrix [* ,MR] AllGather: 0.003313 secs (463.627 MB/s) for 128 x 1500 local matrix Local gemm: 0.0619631 secs (9.29586 GFlop/s) for 1500 x 1500 x 128 product [MC,* ] AllGather: 0.00275298 secs (557.941 MB/s) for1500 x 128 local matrix [* ,MR] AllGather: 0.00332535 secs (461.906 MB/s) for 128 x 1500 local matrix Local gemm: 0.0541908 secs (10.6291 GFlop/s) for 1500 x 1500 x 128 product [MC,* ] AllGather: 0.00276901 secs (554.711 MB/s) for1500 x 128 local matrix [* ,MR] AllGather: 0.00328497 secs (467.583 MB/s) for 128 x 1500 local matrix Local gemm: 0.0606767 secs (9.49294 GFlop/s) for 1500 x 1500 x 128 product [MC,* ] AllGather: 0.00285674 secs (537.676 MB/s) for1500 x 128 local matrix [* ,MR] AllGather: 0.00327521 secs (468.978 MB/s) for 128 x 1500 local matrix Local gemm: 0.060793 secs (9.47477 GFlop/s) for 1500 x 1500 x 128 product [MC,* ] AllGather: 0.0021043 secs (593.07 MB/s) for1500 x 104 local matrix [* ,MR] AllGather: 0.0026644 secs (468.398 MB/s) for 104 x 1500 local matrix Local gemm: 0.0446588 secs (10.4795 GFlop/s) for 1500 x 1500 x 104 product Distributed Gemm: 0.516419 secs Gathered to root: 0.0785039 secs