The Lightweight Distributed Metric Service (LDMS) provides an unprecedented ability to collect system data at resolutions necessary for detecting features and events of interest and to respond on meaningful timescales.
Key features include:
LDMS is available open source at https://github.com/ovis-hpc/ovis
More information can be found at the github wiki, including:
Damian Dechev - University of Central Florida
Tom Tucker - Open Grid Computing
Jim Brandt - Sandia National Laboratories
Eric Roman - Lawrence Berkeley National Laboratory
Ann Gentile - Sandia National Laboratories
ldmscon2019 at easychair dot org