Results 1 - 10 of 76

Dryad: Distributed Data-Parallel Programs from Sequential Building Blocks

by Michael Isard, Mihai Budiu, Yuan Yu, Andrew Birrell, Dennis Fetterly - In EuroSys , 2007
"... Dryad is a general-purpose distributed execution engine for coarse-grain data-parallel applications. A Dryad applica-tion combines computational “vertices ” with communica-tion “channels ” to form a dataflow graph. Dryad runs the application by executing the vertices of this graph on a set of availa ..."
Abstract - Cited by 762 (27 self) - Add to MetaCart
of available computers, communicating as appropriate through files, TCP pipes, and shared-memory FIFOs. The vertices provided by the application developer are quite simple and are usually written as sequential programs with no thread creation or locking. Concurrency arises from Dryad scheduling vertices to run

Dynamics of Random Early Detection

by Dong Lin, Robert Morris - In Proceedings of ACM SIGCOMM , 1997
"... In this paper we evaluate the effectiveness of Random Early Detection (RED) over traffic types categorized as nonadaptive, fragile and robust, according to their responses to congestion. We point out that RED allows unfair bandwidth sharing when a mixture of the three traffic types shares a link Thi ..."
Abstract - Cited by 465 (1 self) - Add to MetaCart
additional queueing delays inside the network These improvements are demonstrated by simulations of TCP and UDP traffic. FRED does not make any assumptions about queueing architecture: it will work with a FIFO gateway. FRED’s peractive-jlow accounting uses memory in proportion to the total number of b

Buffer Management for Shared-Memory ATM Switches

by Mutlu Arpaci , John A. Copeland - IEEE COMMUNICATIONS SURVEYS & TUTORIALS , 2000
"... In the shared-memory switch architecture, output links share a single large memory, in which logical FIFO queues are assigned to each link. Although memory sharing can provide a better queuing performance than physically separated buffers, it requires carefully designed buffer management schemes ..."
Abstract - Cited by 15 (1 self) - Add to MetaCart
In the shared-memory switch architecture, output links share a single large memory, in which logical FIFO queues are assigned to each link. Although memory sharing can provide a better queuing performance than physically separated buffers, it requires carefully designed buffer management schemes

Message-Passing Concurrency on Shared-Memory Multiprocessors

by Charles Eric Laforest , 2007
"... This paper presents an exploration of message-passing concurrency on a commodity Linux shared-memory multiprocessor. While message-passing has more overhead than shared-memory multithreading, it exhibits a greater scalability. Much of this desirable property is derived from the generality and ease o ..."
Abstract - Add to MetaCart
This paper presents an exploration of message-passing concurrency on a commodity Linux shared-memory multiprocessor. While message-passing has more overhead than shared-memory multithreading, it exhibits a greater scalability. Much of this desirable property is derived from the generality and ease

Correct and Efficient Bounded FIFO Queues

by Nhat Minh Lê, Adrien Guatto, Albert Cohen, Antoniu Pop , 2013
"... Bounded single-producer single-consumer FIFO queues are one of the simplest concurrent data-structure, and they do not require more than sequential consistency for correct operation. Still, sequential consistency is an unrealistic hypothesis on shared-memory multiprocessors, and enforcing it throug ..."
Abstract - Cited by 1 (0 self) - Add to MetaCart
Bounded single-producer single-consumer FIFO queues are one of the simplest concurrent data-structure, and they do not require more than sequential consistency for correct operation. Still, sequential consistency is an unrealistic hypothesis on shared-memory multiprocessors, and enforcing

RTLinux POSIX API for IO on Real-time FIFOs and Shared Memory

by Cort Dougan, Matt Sherer , 2003
"... The primary means of communication between # threads and Linux processes are RTLinux FIFOs. These have been limited by the fact that there was a static number of them and their names were restricted to /dev/rtf0 through /dev/rtf.... Starting with RTLinux/Pro # 1.2 and soon Open RTLinux we have ..."
Abstract - Cited by 1 (0 self) - Add to MetaCart
added a means for creating RTLinux FIFOs and shared memory with arbitrary names and locations that can be operated on with ordinary UNIX semantics.

A Simple, Fast and Scalable Non-Blocking Concurrent FIFO Queue for Shared Memory Multiprocessor Systems

by Philippas Tsigas, Yi Zhang - in Proceedings of the 13th ACM Symposium on Parallel Algorithms and Architectures , 2001
"... A non-blocking FIFO queue algorithm for multiprocessor shared memory systems is presented in this paper. The algorithm is very simple, fast and scales very well in both symmetric and non-symmetric multiprocessor shared memory systems. Experiments on a 64-node SUN Enterprise 10000 -- a symmetric mult ..."
Abstract - Cited by 42 (7 self) - Add to MetaCart
A non-blocking FIFO queue algorithm for multiprocessor shared memory systems is presented in this paper. The algorithm is very simple, fast and scales very well in both symmetric and non-symmetric multiprocessor shared memory systems. Experiments on a 64-node SUN Enterprise 10000 -- a symmetric

ABSTRACT

by Michael Isard, Andrew Birrell, Mihai Budiu, Dennis Fetterly, Yuan Yu
"... Dryad is a general-purpose distributed execution engine for coarse-grain data-parallel applications. A Dryad application combines computational “vertices ” with communication “channels ” to form a dataflow graph. Dryad runs the application by executing the vertices of this graph on a set of availabl ..."
Abstract - Add to MetaCart
of available computers, communicating as appropriate through files, TCP pipes, and shared-memory FIFOs. The vertices provided by the application developer are quite simple and are usually written as sequential programs with no thread creation or locking. Concurrency arises from Dryad scheduling vertices to run

Building FIFO and Priority-Queuing Spin Locks from Atomic Swap

by TravisS. Craig , 1993
"... We present practical new algorithms for FIFO or priority-ordered spin locks on sharedmemory multiprocessors with an atomic swap instruction. Different versions of these queuing spin locks are designed for machines with coherent-cache and NUMA memory models. We include extensions to provide nested ..."
Abstract - Cited by 46 (0 self) - Add to MetaCart
We present practical new algorithms for FIFO or priority-ordered spin locks on sharedmemory multiprocessors with an atomic swap instruction. Different versions of these queuing spin locks are designed for machines with coherent-cache and NUMA memory models. We include extensions to provide

A Simple, Fast and Scalable Non-Blocking Concurrent FIFO Queue for Shared Memory Multiprocessor Systems

by unknown authors
"... A non-blocking FIFO queue algorithm for multiprocessor shared memory systems is presented in this paper. The algorithm is very simple, fast and scales very well in both symmetric and non-symmetric multiprocessor shared memory systems. Experiments on a 64-node SUN Enterprise 10000; a symmetric multip ..."
Abstract - Add to MetaCart
A non-blocking FIFO queue algorithm for multiprocessor shared memory systems is presented in this paper. The algorithm is very simple, fast and scales very well in both symmetric and non-symmetric multiprocessor shared memory systems. Experiments on a 64-node SUN Enterprise 10000; a symmetric
Results 1 - 10 of 76