@conference {1323, title = {Fast Batched Matrix Multiplication for Small Sizes using Half Precision Arithmetic on GPUs}, booktitle = {33rd IEEE International Parallel and Distributed Processing Symposium (IPDPS)}, year = {2019}, month = {2019-05}, publisher = {IEEE}, organization = {IEEE}, address = {Rio de Janeiro, Brazil}, author = {Ahmad Abdelfattah and Stanimire Tomov and Jack Dongarra} }