This paper proposes new protocols for the interconnection of FIFO- and causal-ordered broadcast systems, thus increasing their scalability. They use several interconnection links between systems, which avoids bottleneck problems due to the network traffic, since messages are not forced to go throughout a single link but instead through the several links we establish. General architectures to interconnect FIFO- and causal-ordered systems are proposed. Failure management is also discussed and a performance analysis is given, detailing the benefits introduced by these interconnection approaches that are able to easily increase the resulting interconnection bandwidth
This thesis describes a family of VLSI chips designed to link a number of processors on a one-to-one...
Interconnection networks have a key role in distributed processing systems of today but to follow th...
When any protocol is implemented FIFO design is the mandatory module, which provides delay compensat...
One of the most important abstractions for designing distributed programs is the broadcast facility....
Communication protocols designed for database applications are not necessarily suitable for other ap...
A large amount of work has been invested in devising al-gorithms to implement distributed shared mem...
It is desirable to be able to combine ease of implemen-tation of a network with the ability to analy...
We present an implementation of a multicast network of processors. The processors are connected in a...
It is desirable to be able to combine ease of implementation of a network with the ability to analyz...
Distributed programming has to face problems due to asynchronism of underlying communication network...
Several protocols have been proposed to enable efficient multi-point communication (multicast or br...
International audienceFIFO broadcast provides application ordering semantics of messages broadcast b...
This paper presents the slotted-FIFO communication mode that supports communication primitives for t...
Link striping algorithms are often used to overcome transmission bottlenecks in computer networks. H...
A number of broadcast protocols that are reliable subject to a variety of ordering and delivery guar...
This thesis describes a family of VLSI chips designed to link a number of processors on a one-to-one...
Interconnection networks have a key role in distributed processing systems of today but to follow th...
When any protocol is implemented FIFO design is the mandatory module, which provides delay compensat...
One of the most important abstractions for designing distributed programs is the broadcast facility....
Communication protocols designed for database applications are not necessarily suitable for other ap...
A large amount of work has been invested in devising al-gorithms to implement distributed shared mem...
It is desirable to be able to combine ease of implemen-tation of a network with the ability to analy...
We present an implementation of a multicast network of processors. The processors are connected in a...
It is desirable to be able to combine ease of implementation of a network with the ability to analyz...
Distributed programming has to face problems due to asynchronism of underlying communication network...
Several protocols have been proposed to enable efficient multi-point communication (multicast or br...
International audienceFIFO broadcast provides application ordering semantics of messages broadcast b...
This paper presents the slotted-FIFO communication mode that supports communication primitives for t...
Link striping algorithms are often used to overcome transmission bottlenecks in computer networks. H...
A number of broadcast protocols that are reliable subject to a variety of ordering and delivery guar...
This thesis describes a family of VLSI chips designed to link a number of processors on a one-to-one...
Interconnection networks have a key role in distributed processing systems of today but to follow th...
When any protocol is implemented FIFO design is the mandatory module, which provides delay compensat...