Description
Alexis Bandet, INRIA
This is the report for the project 'Optimization of Fault-Tolerance Strategies for Workflow Applications'
Checkpoint operations are periodic and high-volume I/O operations and, as such, are particularly sensitive to interferences. Indeed, HPC applications execute on dedicated nodes but share the I/O system. As a consequence, interferences surge when several applications perform I/O...
Open experimental platforms for Computer Science systems research, like the Chameleon and Grid’5000/FIT testbeds, are a critical tool not only for the support of computer science experimentation but also a key enabler of reproducibility. One of the perennial challenges that scientific instruments of this type grapple with are how they should evolve to support the emergent needs of research....
Scientists have lots of data that they need to store, transport, and use. Lossy compression could be the solution, but there are 32+ compressors, each with its own interface and the interfaces of the most recent compressors often evolve. Moreover, compressors are missing key features: provenance and configuration parameter optimization. LibPressio addresses all these issues by providing a...
This project-talk shall give an update on the current status of the CI-HPC project within JLESC. In the last JLESC-meeting some issues and aspects of CI-HPC were raised, that have been taken care of. Two shall be presented here.
First, an approach to combine best of both worlds from GitHub and GitLab: The large community and visibility of GitHub with the rich feature set that is available...