PBLAS Routine Naming Conventions

Developer Reference for Intel® oneAPI Math Kernel Library for Fortran

Download PDF

ID 766686

Date 3/22/2024

Version

Public

Visible to Intel only — GUID: GUID-E936F8B0-9489-464B-A638-62998627FBEA

View Details

PBLAS Routine Naming Conventions

The naming convention for PBLAS routines is similar to that used for BLAS routines (see Routine Naming Conventions). A general rule is that each routine name in PBLAS, which has a BLAS equivalent, is simply the BLAS name prefixed by initial letter p that stands for "parallel".

The Intel® oneAPI Math Kernel Library (oneMKL) PBLAS routine names have the following structure:

    p <character> <name> <mod> ( )

The <character> field indicates the Fortran data type:

s: real, single precision
c: complex, single precision
d: real, double precision
z: complex, double precision
i: integer

Some routines and functions can have combined character codes, such as sc or dz.

For example, the function pscasum uses a complex input array and returns a real value.

The <name> field, in PBLAS level 1, indicates the operation type. For example, the PBLAS level 1 routines p?dot, p?swap, p?copy compute a vector dot product, vector swap, and a copy vector, respectively.

In PBLAS level 2 and 3, <name> reflects the matrix argument type:

ge: general matrix
sy: symmetric matrix
he: Hermitian matrix
tr: triangular matrix

In PBLAS level 3, the <name>=tran indicates the transposition of the matrix.

The <mod> field, if present, provides additional details of the operation. The PBLAS level 1 names can have the following characters in the <mod> field:

c: conjugated vector
u: unconjugated vector

The PBLAS level 2 names can have the following additional characters in the <mod> field:

mv: matrix-vector product
sv: solving a system of linear equations with matrix-vector operations
r: rank-1 update of a matrix
r2: rank-2 update of a matrix.

The PBLAS level 3 names can have the following additional characters in the <mod> field:

mm: matrix-matrix product
sm: solving a system of linear equations with matrix-matrix operations
rk: rank-k update of a matrix
r2k: rank-2k update of a matrix.

The examples below show how to interpret PBLAS routine names:

`pddot`	`<p> <d> <dot>`: double-precision real distributed vector-vector dot product
`pcdotc`	`<p> <c> <dot> <c>`: complex distributed vector-vector dot product, conjugated
`pscasum`	`<p> <sc> <asum>`: sum of magnitudes of distributed vector elements, single precision real output and single precision complex input
`pcdotu`	`<p> <c> <dot> <u>`: distributed vector-vector dot product, unconjugated, complex
`psgemv`	`<p> <s> <ge> <mv>`: distributed matrix-vector product, general matrix, single precision
`pztrmm`	`<p> <z> <tr> <mm>`: distributed matrix-matrix product, triangular matrix, double-precision complex.

Product and Performance Information
Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex. Notice revision #20201201

Product and Performance Information

Performance varies by use, configuration and other factors. Learn more at www.Intel.com/PerformanceIndex.

Notice revision #20201201

Parent topic: PBLAS Routines

Select Your Language

Using Intel.com Search

Quick Links

Recent Searches

Advanced Search

Only search in

Developer Reference for Intel® oneAPI Math Kernel Library for Fortran

PBLAS Routine Naming Conventions