Paper Search Console

Home Search Page About Contact

Journal Title

Title of Journal: J Supercomput

Search In Journal Title:

Abbravation: The Journal of Supercomputing

Search In Journal Abbravation:

Publisher

Springer US

Search In Publisher:

DOI

10.1002/chin.198943187

Search In DOI:

ISSN

1573-0484

Search In ISSN:
Search In Title Of Papers:

Conditional forwarding simple flow control to inc

Authors: Zhigang Yu Xinyu Wang Kele Shen
Publish Date: 2016/01/04
Volume: 72, Issue: 2, Pages: 639-653
PDF Link

Abstract

Fully adaptive routing algorithm has been widely used by modern commercial supercomputers due to its high path diversity However fully adaptive routing algorithm is prone to deadlock especially when wormhole switching with nonatomic virtual channel VC allocation is employed Nonatomic VC allocation means that a VC can be immediately allocated once the tail flit of last packet arrives Duato’s theory gives a general methodology for deadlockfree fully adaptive routing design by dividing VCs into escape and adaptive ones and prohibiting packets from using adaptive VCs after using escape VCs However this prohibition usually induces adaptivity loss and performance degradation To address these issues we extend Duato’s theory and propose conditional forwarding flow control CFFC for packet residing in escape VC and requesting adaptive VC it can be forwarded if the requested adaptive VC has enough free buffers to hold the whole packet By allowing packets to regain adaptivity CFFC enables the design of a fully adaptive routing algorithm with high routing adaptivity By supporting nonatomic VC allocation CFFC maintains efficient VC utilization We prove that CFFC will not introduce deadlock if the routing algorithm is deadlockfree using nonatomic VC allocation ie the routing subfunction applied in escape VCs is connected and deadlockfree Simulation results show that our proposed method exhibits higher VC utilization and performs averagely 148  better than existing fully adaptive routing algorithmsWe sincerely thank the anonymous reviewers for their helpful comments and suggestions This work is supported in part by the National Science Foundation of China under grants 61171121 61402086 61572279 and Scientific Research Foundation of Liaoning Provincial Education Department No L2015165 and DUFE Excellent Talents Project No DUFE2015R06

Keywords:

References

.
Search In Abstract Of Papers:
Other Papers In This Journal:
  1. Fault-tolerant broadcast in anonymous systems
  2. A hyper-heuristic approach for resource provisioning-based scheduling in grid environment
  3. Constructing data supply chain based on layered PROV
  4. A localization algorithm for large scale mobile wireless sensor networks: a learning approach
  5. Comment on “A general model for long-tailed network traffic approximation”
  6. Direct private query in location-based services with GPU run time analysis
  7. A randomized clustering of anonymous wireless ad hoc networks with an application to the initialization problem
  8. GPU-based parallel genetic approach to large-scale travelling salesman problem
  9. SENFIS: a Sensor Node File System for increasing the scalability and reliability of Wireless Sensor Networks applications
  10. The panconnectivity and the pancycle-connectivity of the generalized base- b hypercube
  11. 3-Hierarchical resource management model on web grid service architecture
  12. Research challenges and perspectives on Wisdom Web of Things (W2T)
  13. A novel framework for social web forums’ thread ranking based on semantics and post quality features
  14. A proposal on location data supplementing information transfer method over MANET
  15. A framework for SaaS software packages evaluation and selection with virtual team and BOCR of analytic network process
  16. Real-time blind audio source separation: performance assessment on an advanced digital signal processor
  17. High-efficient video compression for social multimedia distribution
  18. RCB: leakage-resilient authenticated encryption via re-keying
  19. A framework for efficient performance prediction of distributed applications in heterogeneous systems
  20. Reducing the search space by closure and simplification paradigms
  21. P2P resource searching with Cloning Random Walker assisted by Weakly Connected Dominating Set
  22. Cost and energy aware service provisioning for mobile client in cloud computing environment
  23. PerContRep: a practical reputation system for pervasive content services
  24. Task scheduling for heterogeneous computing systems
  25. Integrating 3D image descriptors of margin sharpness and texture on a GPU-optimized similar pulmonary nodule retrieval engine
  26. Exploring the speedups of embedded microprocessor systems utilizing a high-performance coprocessor data-path
  27. Tuning remote GPU virtualization for InfiniBand networks
  28. Energy efficient utilization of resources in cloud computing systems
  29. The election problem in asynchronous distributed systems with bounded faulty processes
  30. Dynamic-width reconfigurable parallel prefix circuits
  31. Editorial: special issue on parallel nature-inspired optimization
  32. Combating I-O bottleneck using prefetching: model, algorithms, and ramifications
  33. Editorial of special section on advanced in high performance, algorithm, and framework for future computing
  34. Hierarchical multichannel-based integrated smart metering infrastructure
  35. A learning automata-based heuristic algorithm for solving the minimum spanning tree problem in stochastic graphs
  36. Distributed identification of the lineality space of a cone
  37. On studying business models in mobile social networks based on two-sided market (TSM)
  38. Reconfiguration process and routing management for service platforms
  39. Scheduling array redistribution with virtual channel support
  40. A case study of Grid Computing and computer algebra: parallel Gröbner Bases and Characteristic Sets
  41. A rhythm-based access control system for secure wellness services
  42. A novel Adaptive Cluster Transformation (ACT)-based intrusion tolerant architecture for hybrid information technology
  43. Panconnectivity of Cartesian product graphs
  44. Efficient search algorithms for RNAi target detection
Search Result: