PBLAS

Parallel Basic Linear Algebra Subprograms (PBLAS) is an implementation of Level 2 and 3 BLAS intended for distributed memory architectures.^[1] It provides a computational backbone for ScaLAPACK, a parallel implementation of LAPACK. It depends on Level 1 sequential BLAS operations for local computation and BLACS for communication between nodes. ^{[2] [3]}

1. ^ Choi, Jaeyoung; Jack Dongarra, Susan Ostrouchov, Antoine Petitet, David Walker and R. Clinton Whaley (1996). "A proposal for a set of parallel basic linear algebra subprograms". APPLIED PARALLEL COMPUTING COMPUTATIONS IN PHYSICS, CHEMISTRY AND ENGINEERING SCIENCE 1041. doi:10.1007/3-540-60902-4_13. http://www.springerlink.com.proxy.uchicago.edu/content/f5984pg047627566/fulltext.pdf. 
2. ^ "PBLAS". http://www.netlib.org/utk/papers/scalapack/node9.html. 
3. ^ Choi, Jaeyoung; Dongarra, J.J, Walker, D.W. (May 1994). "PB-BLAS: a set of Parallel Block Basic Linear Algebra Subprograms". Scalable High-Performance Computing Conference: 534–541. http://www.netlib.org/utk/people/JackDongarra/journals/079_1996_pb-blas-a-set-of-parallel-block-basic-linear-algebra-subroutines.pdf.