AbstractThis paper studies the implicit costs of synchronization and the possible gains arising from avoiding synchronization in asynchronous environments. An asynchronous generalization of the PRAM model called the APRAM model is used and appropriate complexity measures are defined. The advantage that asynchrony provides is illustrated by analyzing two algorithms: a parallel summation algorithm which proceeds along an implicit complete binary tree and a recursive doubling algorithm which proceeds along a linked list
In this paper, we present a cost model taking into account properties of present time machines. This...
The directed acyclic graph (DAG) associated with a parallel al-gorithm captures the order in which s...
Composite synchronization isa new algorithm that com-bines localized asynchronous coordination, with...
AbstractThis paper studies the implicit costs of synchronization and the possible gains arising from...
Many constant time algorithms for various problems have been developed for the reconfigurable mesh (...
AbstractSeveral recent papers have introduced asynchronous shared memory parallel models in an attem...
We introduce a model of parallel computation that retains the ideal properties of the PRAM by using ...
AbstractWe consider the problem of simulating synchronous computations on asynchronous shared memory...
It is well known that synchronization and communication delays are the major sources of performance ...
So far, we have mainly studied synchronous algorithms. Generally, asynchro-nous algorithms are more ...
We introduce a model of parallel computation that retains the ideal properties of the PRAM by using ...
This extended abstract introduces and evaluates a new model of parallel computation, called the Para...
In this thesis, we present a body of work on the performance and convergence properties of asynchron...
Asynchronous iterations arise naturally on parallel computers if one wants to minimize idle times. T...
textabstractThe parallel execution of branch and bound algorithms can result in seemingly unreasonab...
In this paper, we present a cost model taking into account properties of present time machines. This...
The directed acyclic graph (DAG) associated with a parallel al-gorithm captures the order in which s...
Composite synchronization isa new algorithm that com-bines localized asynchronous coordination, with...
AbstractThis paper studies the implicit costs of synchronization and the possible gains arising from...
Many constant time algorithms for various problems have been developed for the reconfigurable mesh (...
AbstractSeveral recent papers have introduced asynchronous shared memory parallel models in an attem...
We introduce a model of parallel computation that retains the ideal properties of the PRAM by using ...
AbstractWe consider the problem of simulating synchronous computations on asynchronous shared memory...
It is well known that synchronization and communication delays are the major sources of performance ...
So far, we have mainly studied synchronous algorithms. Generally, asynchro-nous algorithms are more ...
We introduce a model of parallel computation that retains the ideal properties of the PRAM by using ...
This extended abstract introduces and evaluates a new model of parallel computation, called the Para...
In this thesis, we present a body of work on the performance and convergence properties of asynchron...
Asynchronous iterations arise naturally on parallel computers if one wants to minimize idle times. T...
textabstractThe parallel execution of branch and bound algorithms can result in seemingly unreasonab...
In this paper, we present a cost model taking into account properties of present time machines. This...
The directed acyclic graph (DAG) associated with a parallel al-gorithm captures the order in which s...
Composite synchronization isa new algorithm that com-bines localized asynchronous coordination, with...