System Support for Handling Transiency in Data Centers and Cloud Systems
Event details
| Date | 10.06.2014 |
| Hour | 10:00 |
| Speaker | Prashant SHENOY, University of Massachusetts |
| Location | |
| Category | Conferences - Seminars |
Modern distributed applications are built using the implicit assumption that the underlying data center servers will be stable and normally available, barring for occasional faults. In many emerging scenarios, however, data centers and clouds only provide transient, rather than continuous, availability of their servers. Transiency in modern distributed systems arises in many contexts, such as green data centers powered using renewable intermittent sources cloud platforms that provide lower-cost spot server instances which can be revoked from their users, or smart data centers that voluntarily curtail their energy usage when signaled by the smart electric grid.
In this talk, I will argue for treating transiency as a first-class design concern when building modern distributed systems and applications. I will present bounded virtual machine migration as a general mechanism for handling transiency in a broad class of systems and discuss how such a mechanism can be employed to design higher-level techniques to handle transiency in data centers and cloud systems. I will end with some preliminary results that demonstrate the benefits and feasibility of our approach.
In this talk, I will argue for treating transiency as a first-class design concern when building modern distributed systems and applications. I will present bounded virtual machine migration as a general mechanism for handling transiency in a broad class of systems and discuss how such a mechanism can be employed to design higher-level techniques to handle transiency in data centers and cloud systems. I will end with some preliminary results that demonstrate the benefits and feasibility of our approach.
Links
Practical information
- General public
- Free
Organizer
- Babak Falsafi
Contact
- Sylvie Thomet