High Volume Services
This project, in its early stages has to do with all things surrounding high availability services.
This includes the management of services, the economics and the performance tuning of services.
Service provision is a delicate interplay between supply and demand.
The ability to deliver reliable service levels depends on both the
resources that are available in the data centre and the pattern of
demand driven by the users. We try to understand how
the incoming demand for a service affects one's ability to formulate
rational Service Level Objectives and hence deliver the appropriate
supply response to the customers with appropriate Service Level
Agreements (SLA).
It is common to have several, sometimes many, servers in parallel,
sharing the load of incoming traffic. Indeed modern operating systems
and CPU architectures are designed with multi-threading in mind, so
the simple view taken by basic queueing models is far from being an
accurate reflection of task scheduling. |
