A parallel multi-unit resource deadlock detection algorithm with O ( log 2 ( min ( m , n ) ) ) overall run-time complexity

Current MPSoCs typically consist of less than a dozen processing units. Future MPSoCs are likely to integrate many more. With this trend, dozens of applications can be running on an MPSoC concurrently and application deadlock on MPSoCs will become a severe problem. To address the application deadloc...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Journal of parallel and distributed computing Ročník 71; číslo 7; s. 938 - 954
Hlavní autoři: Xiao, Xiang, Lee, Jaehwan John
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier Inc 01.07.2011
Témata:
Algorithms
Chip multiprocessor
Classification
Complexity
Computational efficiency
Deadlock detection
Deadlock detection in hardware
Digital logic design
Graph traversing
Graphs
Hardware
Multi-unit resource systems
Parallel algorithm
Reachability computation
Real-time embedded systems
RTOS
Run time (computers)
Running
RTOS
Multi-unit resource systems
Digital logic design
Parallel algorithm
Chip multiprocessor
Reachability computation
Deadlock detection in hardware
Deadlock detection
Real-time embedded systems
Graph traversing
ISSN:0743-7315, 1096-0848
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Current MPSoCs typically consist of less than a dozen processing units. Future MPSoCs are likely to integrate many more. With this trend, dozens of applications can be running on an MPSoC concurrently and application deadlock on MPSoCs will become a severe problem. To address the application deadlock problem in current and future MPSoCs, this article proposes a parallel multi-unit resource deadlock detection algorithm, incorporating four contributions: (1) a classification of resource events that enables each category of events to be handled efficiently, (2) a parallel node hopping mechanism that explores the entire graph exponentially in parallel to obtain information about reachable processes of every resource, (3) an innovative hardware implementation of the node hopping mechanism using bit-wise computations, and (4) proofs of correctness and run-time complexity of the proposed algorithm. Based on information about reachable processes as well as sink nodes in the graph, the proposed algorithm detects deadlock in O(1) run-time. Compared with the worst case run-time of any previous algorithm, which employs a single scheme to handle all resource events, ours is considerably reduced to O ( log 2 ( min ( m , n ) ) ) when implemented in hardware, where m and n are the number of processes and resources, respectively. ► Parallel multi-unit resource deadlock detection algorithm with O ( log 2 ( min ( m , n ) ) ) overall run-time complexity, where m and n are the number of processes and resources, respectively. ► Utilizing a classification of resource events that enables each category of events to be handled efficiently in parallel. ► Parallel node hopping mechanism that explores the entire multi-unit resource allocation graph (RAG) exponentially in parallel to obtain information about reachable processes of every resource. ► Innovative parallel hardware implementation of the node hopping mechanism using bit-wise computations.► Proofs of the correctness and run-time complexity of the proposed algorithm.
Bibliografie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0743-7315
1096-0848
DOI:10.1016/j.jpdc.2011.01.001