Techniques for Scaling and Tuning Concurrent Systems
Event details
| Date | 08.06.2016 |
| Hour | 10:00 › 12:00 |
| Speaker | Oana Balmau |
| Location | |
| Category | Conferences - Seminars |
EDIC Candidacy Exam
Exam President: Prof. Christos Kozyrakis
Thesis Director: Prof. Willy Zwaenepoel
Thesis co-director: Prof. Rachid Guerraoui
Co-examiner: Prof. Katerina Argyraki
Background papers:
GroupLens: applying collaborative filtering to Usenet news.
ProteusTM: Abstraction Meets Performance in Transactional Memory.
Scaling concurrent log-structured data stores
Abstract
To process and benefit from the very large amount of data being generated, it is appealing to leverage the prevalence of commodity multicore machines. Tuning and scaling complex concurrent algorithms and systems are challenging tasks, which significantly hamper their wide adoption. In this proposal, we discuss techniques for improving the scalability of concurrent systems and automatically identifying good system configurations for given workloads.
Exam President: Prof. Christos Kozyrakis
Thesis Director: Prof. Willy Zwaenepoel
Thesis co-director: Prof. Rachid Guerraoui
Co-examiner: Prof. Katerina Argyraki
Background papers:
GroupLens: applying collaborative filtering to Usenet news.
ProteusTM: Abstraction Meets Performance in Transactional Memory.
Scaling concurrent log-structured data stores
Abstract
To process and benefit from the very large amount of data being generated, it is appealing to leverage the prevalence of commodity multicore machines. Tuning and scaling complex concurrent algorithms and systems are challenging tasks, which significantly hamper their wide adoption. In this proposal, we discuss techniques for improving the scalability of concurrent systems and automatically identifying good system configurations for given workloads.
Practical information
- General public
- Free
Contact
- Cecilia Chapuis EDIC