fdurastante/psblas3
1
0
Fork 0
mirror of https://github.com/sfilippone/psblas3.git synced 2026-10-16 10:52:08 +00:00
Code Issues Packages Projects Releases Wiki Activity

Docs updates.

Browse Source
This commit is contained in:
sfilippone
2025-12-23 15:07:22 +01:00
parent 5eb4f55406
commit 7e0347281e
34 changed files with 6974 additions and 6974 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/html/userhtmlse1.html
+27 -27
View File
@@ -10,7 +10,7 @@
<link rel="stylesheet" type="text/css" href="userhtml.css">
</head><body
>
<!--l. 57--><div class="crosslinks"><p class="noindent">[<a
<!--l. 58--><div class="crosslinks"><p class="noindent">[<a
href="userhtmlse2.html" >next</a>] [<a
href="userhtmlli2.html" >prev</a>] [<a
href="userhtmlli2.html#tailuserhtmlli2.html" >prev-tail</a>] [<a
@@ -18,7 +18,7 @@ href="#tailuserhtmlse1.html">tail</a>] [<a
href="userhtml.html#userhtmlse1.html" >up</a>] </p></div>
<h3 class="sectionHead"><span class="titlemark">1 </span> <a
id="x4-30001"></a>Introduction</h3>
<!--l. 59--><p class="noindent" >The PSBLAS library, developed with the aim to facilitate the parallelization of
<!--l. 60--><p class="noindent" >The PSBLAS library, developed with the aim to facilitate the parallelization of
computationally intensive scientific applications, is designed to address parallel
implementation of iterative solvers for sparse linear systems through the
distributed memory paradigm. It includes routines for multiplying sparse
@@ -27,30 +27,31 @@ diagonal entries, preprocessing sparse matrices, and contains additional
routines for dense matrix operations. The current implementation of PSBLAS
addresses a distributed memory execution model operating with message
passing.
<!--l. 70--><p class="indent" > The PSBLAS library version 3 is implemented in the Fortran&#x00A0;2008&#x00A0;<span class="cite">[<a
<!--l. 71--><p class="indent" > The PSBLAS library version 3 is implemented in the Fortran&#x00A0;2008&#x00A0;<span class="cite">[<a
href="userhtmlli3.html#Xmetcalf">17</a>]</span>
programming language, with reuse and/or adaptation of existing Fortran&#x00A0;77 and
Fortran&#x00A0;95 software, plus a handful of C routines.
<!--l. 75--><p class="indent" > The use of Fortran&#x00A0;2008 offers a number of advantages over Fortran&#x00A0;95, mostly
<!--l. 76--><p class="indent" > The use of Fortran&#x00A0;2008 offers a number of advantages over Fortran&#x00A0;95, mostly
in the handling of requirements for evolution and adaptation of the library to new
computing architectures and integration of new algorithms. For a detailed
discussion of our design see&#x00A0;<span class="cite">[<a
href="userhtmlli3.html#XSparse03">11</a>]</span>; other works discussing advanced programming in
Fortran&#x00A0;2008 include&#x00A0;<span class="cite">[<a
href="userhtmlli3.html#XDesPat:11">21</a>,&#x00A0;<a
href="userhtmlli3.html#XRouXiaXu:11">19</a>]</span>; sufficient support for Fortran&#x00A0;2008 is now available
from many compilers, including recent versions of the GNU Fortran compiler from
the Free Software Foundation, and the FLANG compiler from the LLVM
project.
<!--l. 87--><p class="indent" > Previous approaches have been based on mixing Fortran&#x00A0;95, with its support for
href="userhtmlli3.html#XRouXiaXu:11">19</a>]</span>; sufficient support for Fortran&#x00A0;2008 is now
available from many compilers, including recent versions of the GNU Fortran
compiler from the Free Software Foundation, the FLANG compiler from the
LLVM project, and the Intel OneAPI compiler. The README file contains
a list of compilers against which we have successfully tested the current
release.
<!--l. 91--><p class="indent" > Previous approaches have been based on mixing Fortran&#x00A0;95, with its support for
object-based design, with other languages; these have been advocated by a number
of authors, e.g.&#x00A0;<span class="cite">[<a
href="userhtmlli3.html#Xmachiels">16</a>]</span>. Moreover, the Fortran&#x00A0;95 facilities for dynamic memory
management and interface overloading greatly enhance the usability of the PSBLAS
subroutines. In this way, the library can take care of runtime memory requirements
that are quite difficult or even impossible to predict at implementation or
compilation time.
<!--l. 97--><p class="indent" > The presentation of the PSBLAS library follows the general structure of the
href="userhtmlli3.html#Xmachiels">16</a>]</span>. The Fortran&#x00A0;95 facilities for dynamic memory management and
interface overloading ensure that the library can take care of runtime memory
requirements that are quite difficult or even impossible to predict at implementation
or compilation time.
<!--l. 99--><p class="indent" > The presentation of the PSBLAS library follows the general structure of the
proposal for serial Sparse BLAS&#x00A0;<span class="cite">[<a
href="userhtmlli3.html#Xsblas97">8</a>,&#x00A0;<a
href="userhtmlli3.html#Xsblas02">9</a>]</span>, which in its turn is based on the proposal for
@@ -58,16 +59,15 @@ BLAS on dense matrices&#x00A0;<span class="cite">[<a
href="userhtmlli3.html#XBLAS1">15</a>,&#x00A0;<a
href="userhtmlli3.html#XBLAS2">5</a>,&#x00A0;<a
href="userhtmlli3.html#XBLAS3">6</a>]</span>.
<!--l. 102--><p class="indent" > The applicability of sparse iterative solvers to many different areas causes
some terminology problems because the same concept may be denoted
through different names depending on the application area. The PSBLAS
features presented in this document will be discussed referring to a finite
difference discretization of a Partial Differential Equation (PDE). However,
the scope of the library is wider than that: for example, it can be applied
to finite element discretizations of PDEs, and even to different classes of
problems such as nonlinear optimization, for example in optimal control
problems.
<!--l. 112--><p class="indent" > The design of a solver for sparse linear systems is driven by many conflicting
<!--l. 104--><p class="indent" > The applicability of sparse iterative solvers to many different areas causes some
terminology problems because the same concept may be denoted by different names
depending on the application area. The PSBLAS features presented in this document
will be discussed taking as a reference a finite difference discretization of a Partial
Differential Equation (PDE). However, the scope of the library is wider than that: it
can be applied to finite element and other discretizations of PDEs, and even to
different classes of problems such as nonlinear optimization, for example in optimal
control problems.
<!--l. 114--><p class="indent" > The design of a solver for sparse linear systems is driven by many conflicting
objectives, such as limiting occupation of storage resources, exploiting regularities in
the input data, exploiting hardware characteristics of the parallel platform. To
@@ -88,12 +88,12 @@ applications.
<!--l. 129--><div class="crosslinks"><p class="noindent">[<a
<!--l. 131--><div class="crosslinks"><p class="noindent">[<a
href="userhtmlse2.html" >next</a>] [<a
href="userhtmlli2.html" >prev</a>] [<a
href="userhtmlli2.html#tailuserhtmlli2.html" >prev-tail</a>] [<a
href="userhtmlse1.html" >front</a>] [<a
href="userhtml.html#userhtmlse1.html" >up</a>] </p></div>
<!--l. 129--><p class="indent" > <a
<!--l. 131--><p class="indent" > <a
id="tailuserhtmlse1.html"></a>
</body></html>