New version of BFQ, benchmark suite and experimental results
ثبت نشده
چکیده
BFQ (Budget Fair Queueing) is a production-quality, proportional-share disk scheduler with a relatively large user base. Part of its success is due to a set of simple heuristics that we added to the original algorithm about one year ago. These heuristics are the main focus of this document. The first heuristic enriches BFQ with one of the most desirable properties for a desktop or handheld system: responsiveness. The remaining heuristics improve the robustness of BFQ across heterogeneous devices, and help BFQ to preserve a high throughput under demanding workloads. To measure the performance of these heuristics we have implemented a suite of micro and macro benchmarks mimicking several real-world tasks, and have run it on three different systems with a single rotational disk. We have also compared our results against Completely Fair Queueing (CFQ), the default Linux disk scheduler. As a result of our heuristics: 1) whatever the disk load is, interactive applications are virtually as responsive as if the disk was idle; 2) latencies comparable to CFQ are still guaranteed to time-sensitive, non-interactive applications, as, e.g., audio and video players; 3) a high throughput is achieved also in the presence of many concurrent requests and sudden increases of the workload. textbf
منابع مشابه
PARSEC 2.0: A New Benchmark Suite for Chip-Multiprocessors
The second version of the Princeton Application Repository for Shared-Memory Computers (PARSEC) has been released. PARSEC is a benchmark suite for Chip-Multiprocessors (CMPs) that focuses on emerging applications. It includes a diverse set of workloads from different domains such as interactive animation or systems applications that mimic large-scale commercial workloads. The next version of PA...
متن کاملTest Selection on Extended Finite State Machines with Provable Guarantees
Building high confidence regression test suites to validate new system versions is a challenging problem. A modelbased approach to build a regression test suite from a given test suite is described. The generated test suite includes every test that will traverse a change performed to produce the new version, and consists of only such tests to reduce the testing costs. Finite state machines exte...
متن کاملThe GemsTD benchmark suite
We introduce version 1.0 of the GemsTD benchmark suite and report the performance on several different architectures for the cases in this suite. TRITA-PDC-2003:2 • ISSN 1401-2731 • ISRN KTH/PDC/R--03/2--SE
متن کاملChaotic Genetic Algorithm based on Explicit Memory with a new Strategy for Updating and Retrieval of Memory in Dynamic Environments
Many of the problems considered in optimization and learning assume that solutions exist in a dynamic. Hence, algorithms are required that dynamically adapt with the problem’s conditions and search new conditions. Mostly, utilization of information from the past allows to quickly adapting changes after. This is the idea underlining the use of memory in this field, what involves key design issue...
متن کاملThe Genesis distributed memory benchmarks
The genesis distributed-memory benchmarks represent a signiicant step forward in the evaluation of distributed-memory MIMD systems. The initial version of this benchmark suite was augmented and modiied in order to evaluate the suprenum computer. This augmented suite has now been further modiied to allow the benchmarks to be run on several important distributed systems without further code modii...
متن کامل