On Bufferless Routing of Variable Length Messages in Leveled Networks

@article{Bhatt1996OnBR,
  title={On Bufferless Routing of Variable Length Messages in Leveled Networks},
  author={Sandeep N. Bhatt and Gianfranco Bilardi and Geppino Pucci and Abhiram G. Ranade and Arnold L. Rosenberg and Eric J. Schwabe},
  journal={IEEE Trans. Computers},
  year={1996},
  volume={45},
  pages={714-729}
}
We study the most general communication paradigm on a multiprocessor, wherein each processor has a distinct message (of possibly distinct lengths) for each other processor. We study this paradigm, which we call chatting, on multiprocessors that do not allow messages once dispatched ever to be delayed on their routes. By insisting on oblivious routes for messages, we convert the communication problem to a pure scheduling problem. We introduce the notion of a virtual chatting schedule, and we… 
View via Publisher
doi.ieeecomputersociety.org
19 Citations
Highly Influential Citations
1
Background Citations
8
Methods Citations
1

Figures from this paper

19 Citations

Citation Type

Citation Type

Scheduling Time-Constrained Communication in Linear Networks
TLDR
It is shown thatbuffering can improve throughput in general by a logarithmic factor (but no more), but that in several significant special cases, such as when all messages can be released immediately, buffering can help by only a small constant factor.
Scheduling Time-Constrained Communication in Linear Networks
TLDR
The problem of centrally scheduling multiple messages in a linear network, when each message has both a release time and a deadline is studied, and it is shown that buffering can improve throughput in general by a logarithmic factor, but that in several significant special cases, Buffering can help by only a small constant factor.
Message scheduling on trees under a generalized line-communication model
  • D. Barth, M. Valencia-Pabon
  • Computer Science
    Proceedings Fourth International Symposium on Parallel Architectures, Algorithms, and Networks (I-SPAN'99)
  • 1999
TLDR
It is shown that the problem of finding optimal schedules to route any set of messages under both the generalized line-communication model and the bufferless routing model is NP-hard even if restricted to binary trees.
Methods and problems of communication in usual networks
Time-constrained scheduling of weighted packets on trees and meshes
TLDR
This paper provides constant factor approximation algorithms for the time-constrained scheduling problem with weighted packets on trees and meshes and provides logarithmic approximations for the same problems in the buffered case.
Efficient bufferless packet switching on trees and leveled networks
Universal Bufierless Routing
TLDR
A new deterministic technique that constructs a universal bufierless algorithm by emulating a universalbufiered algorithm, which is to replace packet bufiering with packet circulation on regions of the network.
Precedence-Constrained Task Allocation onto Point-to-Point Networks for Pipelined Execution
TLDR
A heuristic procedure based on preclustering, recursive mincut bipartitioning, and iterative improvement is proposed to reduce the maximum contention due to communication in the network, increasing the likelihood that messages meet their deadlines.
Universal Bufferless Routing
TLDR
A new deterministic technique is constructed that constructs a universal bufferless algorithm by emulating a universal buffered algorithm to replace packet buffering with packet circulation on regions of the network.
Modeling parallel bandwidth: local vs. global restrictions
TLDR
This paper focuses on models of parallel computation that account for bandwidth limitations as an aggregate parameter m < p, such that the p processors can send at most m messages in total at each step.
...
...

References

SHOWING 1-10 OF 24 REFERENCES
Randomized Routing and Sorting on Fixed-Connection Networks
TLDR
A randomized on-line algorithm for scheduling any set of N packets whose paths have congestion c on any bounded-degree leveled network with depth L in O(c + L + log N) steps, using constant-size queues.
A theory of wormhole routing in parallel computers
TLDR
Simulation results suggest that the idea of random initial delays is not only useful for theoretical analysis but may actually improve the performance of wormhole routing algorithms.
Routing, merging and sorting on parallel models of computation
TLDR
It is shown that log log n - log log r is asymptotically optimal for rn processors to merge two sorted lists of n elements and is able to achieve such an efficient sort via Valiant's parallel merging algorithm.
A Scheme for Fast Parallel Communication
TLDR
There is a distributed randomized algorithm that can route every packet to its destination without two packets passing down the same wire at any one time, and finishes within time $O(\log N)$ with overwhelming probability for all such routing requests.
Packet Routing in Networks with Long Wires
TLDR
These results improve upon previous routing results which assume that unit time suffices to traverse a wire of any length and yield improved results for job-shop scheduling as long as a technical restriction on the job- shop problem is incorporated.
Deadlock-Free Message Routing in Multiprocessor Interconnection Networks
TLDR
A deadlock-free routing algorithm can be generated for arbitrary interconnection networks using the concept of virtual channels, which is used to develop deadlocked routing algorithms for k-ary n-cubes, for cube-connected cycles, and for shuffle-exchange networks.
Exact analysis of hot-potato routing
  • U. Feige, P. Raghavan
  • Computer Science
    Proceedings., 33rd Annual Symposium on Foundations of Computer Science
  • 1992
TLDR
The authors consider a form of packet routing known as hot potato routing or deflection routing, whose striking feature is that there are no buffers at intermediate nodes, and give a simple deterministic algorithm that on a n*n torus will route a random instance in 2n+O(log n) steps with high probability.
Fast algorithms for bit-serial routing on a hypercube
TLDR
The algorithm is adaptive and it is shown that this is necessary to achieve the logarithmic speedup, and generalize the Borodin-Hopcroft lower bound on oblivious routing by proving that any randomized oblivious algorithm on a polylogarithic degree network requires at least Ω(log2N/log logN) bit steps with high probability for almost all permutations.
An Adaptive and Fault Tolerant Wormhole Routing Strategy for k-Ary n-Cubes
TLDR
The concept of virtual channels is extended to multiple virtual communication systems that provide adaptability and fault tolerance in addition to being deadlock-free, and virtual interconnection networks allowing adaptive, deadlocks-free routing are examined.
An optimal synchronizer for the hypercube
TLDR
A novel technique for constructing network synchronizers is presented, developed from some basic relationships between synchronizers and the structure of a t-spanning subgraph over the network, and a synchronizer for the hypercube with optimal time and communication complexities is obtained.
...
...