AD Tool: TAMC
Introduction
Applications
Tools
Research Groups
Workshops
Publications
My Account
About

TAMC


Summary:
TAMC is a source-to-source AD-tool for FORTRAN-77 programs. The generated code propagates derivatives in forward (tangent linear) or reverse (adjoint) mode. TAMC is very flexible thanks to many options and user directives.

URL: http://www.autodiff.com/tamc

Developers:
Mode: Forward
 
Method: Source transformation
 
supported Language: Fortran77

Reference:
Ralf Giering
Tangent Linear and Adjoint Model Compiler, Users Manual
Center for Global Change Sciences, Department of Earth, Atmospheric, and Planetary Science, MIT, 1997

Ralf Giering, Thomas Kaminski
Recipes for Adjoint Code Construction
Article in ACM Transactions on Mathematical Software, 1998



Features:

Fortran 77 Support:
TAMC supports almost the full FORTRAN-77 standard and also some common extenstions.
Fortran 90 Support:
Some Fortran-90 extensions to FORTRAN-77 are supported.
Array assignments and WHERE statements are supported.
Most intrinsic functions are handled.
Analysises:
TAMC normalizes the code and applies a control flow analysis.
TAMC applies an intraprocedural data dependence
and an interprocedural data flow analysis.
Given the independent and dependent variables
of the specified top-level routine,
TAMC determines all active routines and variables
and produces derivative code only for those.
Directives
TAMC accepts several directives.
Using the reverse mode storing of variables instead of recomputation
is controlled by directives.

Multi level checkpointing can be implemented by splitting
a loop and inserting directives.

Black box (library) routines are handled using flow information
given by directives.


AwardsHeinz-Billing-Award 1995 for the advancement of scientific computing

Supported Platforms:
  • Application Server


Licensing: free with restrictions

References on TAMC in our publication database:  131

10+
#Pubs
0
5
7
7
12
14
14
30
19
11
10
1
1
'96 '97 '98 '99 '00 '01 '02 '03 '04 '05 '06 '08
Year

Related Research Groups:

  

LinkedIn:    Contact:
autodiff.org
Username:
Password:
(lost password)