Using Coarrays to Parallelize Legacy Fortran Applications: Strategy and Case Study

Table 1

Modernization steps: horizontal lines indicate partial ordering.

Step	Details
1	Set up automated builds via CMake^1 and version control via Git^2.
2	Convert fixed- to free-source format via “convert.f90” by Metcalf^3.
3	Replace goto with do while for main loop termination.
4	Enforce type/kind/rank consistency of arguments and return values by wrapping all procedures in a module.
5	Eliminate implicit typing.
6	Replace data statements with parameter statements.
7	Replace write-access to common blocks with module variables.
8	Replace keyboard input with default initializations.
9	Set up automated, extensible tests for accuracy and performance via OOP and CTest^1.
10	Make all procedures outside of the main program pure.
11	Eliminate actual/dummy array shape inconsistencies by passing array subsections to assumed-shape arrays.
12	Replace static memory allocation with dynamic allocation.
13	Replace loops with array assignments.
14	Expose greater parallelism by unrolling the nested loops in the particle set-up.
15	Balance the work distribution by spreading particles across images during set-up.
16	Exploit a Fortran 2015 collective procedure to gather statistics.
17	Study and tune performance with TAU^4.

http://www.cmake.org/.
²http://git-scm.com/.
ftp://ftp.numerical.rl.ac.uk/pub/MandR/convert.f90.
⁴http://tau.uoregon.edu/.