Cache-Sensitive Pre-runtime Scheduling
نویسندگان
چکیده
Daniel K astner Stephan Thesing Universitat des Saarlandes / Fachbereich Informatik Postfach 15 11 50 / D-66041 Saarbr ucken / Germany Phone: +49 681 302 5589 Fax: +49 681 302 3065 fkaestner,[email protected] http://www.cs.uni-sb.de/~fkaestner,thesingg Abstract We present a novel pre-runtime scheduling method for uniprocessors which precisely incorporates the e ects of task switching on the processor cache into its decisions. Tasks are modelled as a sequence of non preemtable segments with precedence constraints. The cache behavior of each task segment is statically determined by abstract interpretation. For the sake of e ciency, the scheduling algorithm uses a heuristically guided search strategy.
منابع مشابه
Combining Locality Analysis with Online Proactive Job Co-scheduling in Chip Multiprocessors
The shared-cache contention on Chip Multiprocessors causes performance degradation to applications and hurts system fairness. Many previously proposed solutions schedule programs according to runtime sampled cache performance to reduce cache contention. The strong dependence on runtime sampling inherently limits the scalability and effectiveness of those techniques. This work explores the combi...
متن کاملCAPS: Contention-Aware Proactive Scheduling for CMPs
Many Chip Multiprocessors (CMPs) rely on shared caches to hide the latency of inter-thread communications as well as to improve effective memory bandwidth. Yet along comes cache contention, which often results in cache thrashing and severe performance degradation. Because of the variety of programs, a suitable schedule can often alleviate the issues significantly. However, it remains an open qu...
متن کاملCache and Pipeline Sensitive Fixed Priority Scheduling for Preemptive Real-Time Systems
Current schedulability analyses for preemptive systems consider cache behaviour by adding preemption caused cache reload costs. Thereby, they ignore the fact that delays due to cache misses often have a reduced impact because of pipeline effects. In this paper, these methods are called isolated. Pipeline-related preemption costs are not considered at all in current schedulability analyses. This...
متن کاملMemory Conscious Scheduling and Processor Allocation on NUMA Architrchitectures
Operating system abstractions do not always meet the needs of a language or applications designer. A lack of efficiency and functionality in scheduling mechanisms can be filled by an application-specific runtime environment providing mechanisms for dynamic processor allocation and memory conscious scheduling. We believe that a synergistic approach that involves three components, the operating s...
متن کاملRuntime Data Flow Graph Scheduling of Matrix Computations with Multiple Hardware Accelerators
Abstract In our previous work, we have presented a systematic methodology for parallelizing dense matrix computations using a separation of concerns between the code that implements a linear algebra algorithm and a runtime system that exploits parallelism for which only relatively simple scheduling algorithms were used to parallelize a wide range of dense matrix computations. We have extended t...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 1998