Mining Software Usage with the Automatic Library Tracking Database (ALTD)

Conference · Tue Jan 01 04:00:00 EST 2013

OSTI ID:1091676

Hadri, Bilel ^[1]; Fahey, Mark R ^[1]

---

1. ORNL

Tracking software usage is important for HPC centers, computer vendors, code developers and funding agencies to provide more efficient and targeted software support, and to forecast needs and guide HPC software effort towards the Exascale era. However, accurately tracking software usage on HPC systems has been a challenging task. In this paper, we present a tool called Automatic Library Tracking Database (ALTD) that has been developed and put in production on several Cray systems. The ALTD infrastructure prototype automatically and transparently stores information about libraries linked into an application at compilation time and also the executables launched in a batch job. We will illustrate the usage of libraries, compilers and third party software applications on a system managed by the National Institute for Computational Sciences.

• Research Organization: Oak Ridge National Laboratory (ORNL)
• Sponsoring Organization: ME USDOE - Office of Management, Budget, and Evaluation; ORNL work for others
• DOE Contract Number: AC05-00OR22725
• OSTI ID: 1091676
• Country of Publication: United States
• Language: English