An Evaluation of Network Stack Parallelization Strategies in Modern Operating Systems

As technology trends push future microprocessors toward chip multiprocessor designs, operating system network stacks must be parallelized in order to keep pace with improvements in network bandwidth. There are two competing strategies for stack parallelization. Message-parallel network stacks use concurrent threads to carry out network operations on independent messages (usually packets), whereas connection-parallel stacks map operations to groups of connections and permit concurrent processing on independent connection groups. Connection-parallel stacks can use either locks or threads to serialize access to connection groups. This paper evaluates these parallel stack organizations using a modern operating system and chip multiprocessor hardware.

Compared to uniprocessor kernels, all parallel stack organizations incur additional locking overhead, cache inefficiencies, and scheduling overhead. However, the organizations balance these limitations differently, leading to variations in peak performance and connection scalability. Lock-serialized connection-parallel organizations reduce the locking overhead of message-parallel organizations by using many connection groups and eliminate the expensive thread handoff mechanism of thread-serialized connection-parallel organizations. The resultant organization outperforms the others, delivering 5.4 Gb/s of TCP throughput for most connection loads and providing a 126% throughput improvement versus a uniprocessor for the heaviest connection loads.

OpenVZ

OpenVZ es una tecnología de virtualización en el nivel de sistema operativo para Linux. OpenVZ permite que un servidor físico ejecute múltiples instancias de sistemas operativos aislados, conocidos como Servidores Privados Virtuales (SPV o VPS en inglés) o Entornos Virtuales (EV).

Si se lo compara a máquinas virtuales tales como VMware, VirtualBox y las tecnologías de virtualización tales como Xen, OpenVZ ofrece menor flexibilidad en la elección del sistema operativo: tanto los huéspedes como los anfitriones deben ser Linux (aunque las distribuciones de GNU/Linux pueden ser diferentes en diferentes EVs). Sin embargo, la virtualización en el nivel de sistema operativo de OpenVZ proporciona mejor rendimiento, escalabilidad, densidad, administración de recursos dinámicos, y facilidad de administración que las alternativas.

OpenVZ es una base de Virtuozzo que es un software comercial desarrollado por SWsoft, Inc., OpenVZ es un producto de software libre y licenciado bajo los términos de la licencia GNU GPL versión 2.

OpenVZ consiste del núcleo y de herramientas en el nivel de usuario.

Against Anarchist Apartheid

questingenlightenment:

Group 1

• Pierre-Joseph Proudhon
• Josiah Warren

• Stephen Pearl Andrews
• Ezra Heywood
• Anselme Bellegarrigue
• Lysander Spooner
• Benjamin Tucker
• Francis D. Tandy
• John Henry Mackay
• Voltairine de Cleyre (early)
• Franz Oppenheimer

Group 2 

• Gustave de Molinari
• Herbert Spencer (early)
• Auberon Herbert
• Wordsworth Donisthorpe
• Rose Wilder Lane
• Robert LeFevre
• Murray Rothbard
• David Friedman
• Randy Barnett
• Samuel E. Konkin 3.0
• Hans-Hermann Hoppe

It’s obvious what the two lists have in common: all the names on both lists belong to thinkers who have favoured radically free markets and the abolition of the state – hence, one might infer, market anarchists.

But it’s quite common in left-anarchist circles to insist that while the Group 1 thinkers are genuine anarchists, those in Group 2 are not true anarchists at all – on the grounds that true anarchists must oppose not only the state but also capitalism. Group 1, we’re told, is commendably anti-capitalist and so authentically anarchist; but the members of Group 2 exclude themselves from the anarchist ranks by their advocacy of capitalism. (I’m not sure into which group geolibs like Albert J. Nock and Frank Chodorov, or migrating thinkers like Karl Hess, are supposed to fall, so I left their names off.)

Read More

(Fuente: aaeblog.com, vía foucaultofyou)

proclus:

Iran Claims New Cyber Attack On Its Nuclear Plants, Blames US and Allies - Slashdot http://ping.fm/i9qBk

Tahoe: A Secure Distributed Filesystem

Abstract

The “Tahoe” project is a distributed filesystem, which safely stores files on multiple machines to protect against hardware failures. Cryptographic tools are used to ensure integrity and confidentiality, and a decentralized architecture minimizes single points of failure. Files can be accessed through a web interface or native system calls (via FUSE). Fine-grained sharing allows individual files or directories to be delegated by passing short URI-like strings through email. Tahoe grids are easy to set up, and can be used by a handful of friends or by a large company for thousands of customers.