Spelling suggestions: "subject:"aperating systems"" "subject:"boperating systems""
211 |
Architectures for Service Differentiation in Overloaded Internet ServersVoigt, Thiemo January 2002 (has links)
Web servers become overloaded when one or several server resources such as network interface, CPU and disk become overutilized. Server overload leads to low server throughput and long response times experienced by the clients. Traditional server design includes only marginal or no support for overload protection. This thesis presents the design, implementation and evaluation of architectures that provide overload protection and service differentiation in web servers. During server overload not all requests can be processed in a timely manner. Therefore, it is desirable to perform service differentiation, i.e., to service requests that are regarded as more important than others. Since requests that are eventually discarded also consume resources, admission control should be performed as early as possible in the lifetime of a web transaction. Depending on the workload, some server resources can be overutilized while the demand on other resources is low because certain types of requests utilize one resource more than others. The implementation of admission control in the kernel of the operating system shows that this approach is more efficient and scalable than implementing the same scheme in user space. We also present an admission control architecture that performs admission control based on the current server resource utilization combined with knowledge about resource consumption of requests. Experiments demonstrate more than 40% higher throughput during overload compared to a standard server and several magnitudes lower response times. This thesis also presents novel architectures and implementations of operating system support for predictable service guarantees. The Nemesis operating system provides applications with a guaranteed communication service using the developed TCP/IP implementation and the scheduling of server resources. SILK (Scout in the Linux kernel) is a new networking stack for the Linux operating system that is based on the Scout operating system. Experiments show that SILK enables prioritizing and other forms of service differentiation between network connections while running unmodified Linux applications.
|
212 |
KernTune: self-tuning Linux kernel performance using support vector machines.Yi, Long. January 2006 (has links)
<p>Self-tuning has been an elusive goal for operating systems and is becoming a pressing issue for modern operating systems. Well-trained system administrators are able to tune an operating system to achieve better system performance for a specific system class. Unfortunately, the system class can change when the running applications change. The model for self-tuning operating system is based on a monitor-classify-adjust loop. The idea of this loop is to continuously monitor certain performance metrics, and whenever these change, the system determines the new system class and dynamically adjusts tuning parameters for this new class. This thesis described KernTune, a prototype tool that identifies the system class and improves system performance automatically. A key aspect of KernTune is the notion of Artificial Intelligence oriented performance tuning. Its uses a support vector machine to identify the system class, and tunes the operating system for that specific system class. This thesis presented design and implementation details for KernTune. It showed how KernTune identifies a system class and tunes the operating system for improved performance.</p>
|
213 |
Equité d'accès aux ressources dans les systèmes partagés best-effortGoichon, François 16 December 2013 (has links) (PDF)
Au cours de la dernière décennie, l'industrie du service informatique s'est métamorphosée afin de répondre à des besoins client croissants en termes de disponibilité, de performance ou de capacité de stockage des systèmes informatisés. Afin de faire face à ces demandes, les hébergeurs d'infrastructures ont naturellement adopté le partage de systèmes où les charges de travail de différents clients sont exécutées simultanément. Cette technique, mutualisant les ressources à disposition d'un système entre les utilisateurs, permet aux hébergeurs de réduire le coût de maintenance de leurs infrastructures, mais pose des problèmes d'interférence de performance et d'équité d'accès aux ressources. Nous désignons par le terme systèmes partagés best-effort les systèmes dont la gestion de ressources est centrée autour d'une maximisation de l'usage des ressources à disposition, tout en garantissant une répartition équitable entre les différents utilisateurs. Dans ce travail, nous soulignons la possibilité pour un utilisateur abusif d'attaquer les ressources d'une plateforme partagée afin de réduire de manière significative la qualité de service fournie aux autres utilisateurs concurrents. Le manque de métriques génériques aux différentes ressources, ainsi que le compromis naturel entre équité et optimisation des performances forment les causes principales des problèmes rencontrés dans ces systèmes. Nous introduisons le temps d'utilisation comme métrique générique de consommation des ressources, métrique s'adaptant aux différentes ressources gérées par les systèmes partagés best-effort. Ceci nous amène à la spécification de couches de contrôles génériques, transparentes et automatisées d'application de politiques d'équité garantissant une utilisation maximisée des ressources régulées. Notre prototype, implémenté au sein du noyau Linux, nous permet d'évaluer l'apport de notre approche pour la régulation des surcharges d'utilisation mémoire. Nous observons une amélioration significative de la performance d'applications typiques des systèmes partagés best-effort en temps de contention mémoire. De plus, notre technique borne l'impact d'applications abusives sur d'autres applications légitimes concurrentes, puisque l'incertitude sur les durées d'exécution est naturellement amoindrie.
|
214 |
Automatically increasing fault tolerance in distributed systemsBazzi, Rida Adnan January 1994 (has links)
No description available.
|
215 |
A spectral method for mapping dataflow graphsElling, Volker Wilhelm January 1998 (has links)
No description available.
|
216 |
On concurrency control in logbased databasesGong, Guohui January 1999 (has links)
No description available.
|
217 |
Shared state management for time-sensitive distributed applicationsKrishnaswamy, Vijaykumar January 2001 (has links)
No description available.
|
218 |
A framework for the development of wide area distributed applicationsTopol, Brad Byer January 1998 (has links)
No description available.
|
219 |
Support for fault-tolerant computations in distributed object systemsChelliah, Muthusamy January 1996 (has links)
No description available.
|
220 |
Implementing and programming weakly consistent memoriesJohn, Ranjit 12 1900 (has links)
No description available.
|
Page generated in 0.1167 seconds