Title: Towards the Incorporation of Dynamic Adaptation into Operating Systems
Speaker: Patricia Teller, University of Texas at El Paso
Date/Time: Monday, August 28, 2006, 1:00 – 2:00 pm
Location: CSRI Building, Room 90 (Sandia NM)
Brief Abstract: In the context of the DAiSES (Dynamic Adaptability in Support of Extreme Scale) research project, which is funded by the Department of Energy Office of Science, we are investigating ways to incorporate adaptation into operating systems, either by varying parameter values or policies at runtime. As compared to conventional operating systems with statically defined parameters and policies, such operating systems offer the potential for improved performance. In response to changing workload
characteristics or requirements, operating system (OS) adaptation is meant to dynamically ?customize? the OS in an attempt to provide ?best? service, based on predefined criteria, for the active workload. This phase of our research focuses on conventional operating systems, but hopefully this experience will lay the foundation for addressing adaptation in operating systems for extreme scale systems.
Current DAiSES research activities focus on three adaptation targets: disk scheduling, virtual memory management, and file I/O. These initial targets were chosen based on evidence of the potential performance gains that could be achieved by varying associated parameters and policies. To improve performance in any one of these ways, we study the correlation among workload characteristics/requirements, the relevant OS parameter
values and/or policies, and achievable performance gains as measured by predefined performance metrics. These relationships, which are challenging to establish, in conjunction with the system state, measured at runtime, dictate the adaptation process.
Thus far, disk scheduling has received most of our attention, and will be a major focus of this talk. It is a target for OS adaptation via policy change, while, for example, virtual memory management is a target for adaptation of OS parameter values. In terms of disk scheduling, we have designed a new disk scheduling strategy that leverages a fair queuing discipline and implements a fair scheduling algorithm that can satisfy different performance requirements for different concurrently executing applications. Because this strategy is fair in terms of allocated disk time, it provides performance isolation among applications and, therefore, predictable disk performance for each application, which facilitates providing quality of service guarantees. Currently, this strategy is being used to support adaptation of policies.
CSRI POC: Rolf Riesen, (505) 845-7363 |