Filters








107 Hits in 3.0 sec

Energy-Efficient Deflection-based On-chip Networks: Topology, Routing, Flow Control [article]

Rachata Ausavarungnirun, Onur Mutlu
2021 arXiv   pre-print
high performance and energy efficiency under both low and high network loads, this chapter discusses critical drawbacks of traditional bufferless designs and describes recent research works focusing on two  ...  These two figures provide quantitative evidence that the network is deadlock-free and livelock-free. Several observations are in order: 1.  ...  We introduce two mechanisms to provide a deterministic guarantee of livelock-free operation in [23, 24] .  ... 
arXiv:2112.02516v2 fatcat:vhrsn7lknncgnaxf7ycw6ylzam

An approach to verification and validation of a reliable multicasting protocol

John R. Callahan, Todd L. Montgomery
1996 Proceedings of the 1996 international symposium on Software testing and analysis - ISSTA '96  
As a result, the ring is wedged in a livelock state because sites cannot fill some empty slots in their Ordering Queues.  ...  The original algorithm, which we call the Token Ring Protocol (TRP), allows sites in a packet-switching network to establish a token ring for distributing responsibility for acknowledgments.  ... 
doi:10.1145/229000.226316 dblp:conf/issta/CallahanM96 fatcat:xr4i7jkmsndqdk7ninlzsfcc5q

An approach to verification and validation of a reliable multicasting protocol

John R. Callahan, Todd L. Montgomery
1996 Software engineering notes  
As a result, the ring is wedged in a livelock state because sites cannot fill some empty slots in their Ordering Queues.  ...  The original algorithm, which we call the Token Ring Protocol (TRP), allows sites in a packet-switching network to establish a token ring for distributing responsibility for acknowledgments.  ... 
doi:10.1145/226295.226316 fatcat:ggqxhvrbzrenreeri4iq5bz5yy

CHIPPER: A low-complexity bufferless deflection router

Chris Fallin, Chris Craik, Onur Mutlu
2011 2011 IEEE 17th International Symposium on High Performance Computer Architecture  
Explicit mechanisms are required for livelock freedom due to the non-minimal nature of deflection.  ...  We introduce three key insights: first, that deflection routing port allocation maps naturally to a permutation network within the router; second, that livelock freedom requires only an implicit token-passing  ...  Acknowledgments We thank the anonymous reviewers for their feedback. We gratefully acknowledge members of the SAFARI research  ... 
doi:10.1109/hpca.2011.5749724 dblp:conf/hpca/FallinCM11 fatcat:4birjeoienflhn7n4lz7bkqqqi

Error- and flow-control protocols for terabit optical networks

Ted H. Szymanski, Michael R. Feldman, Richard L. Li, W. Brian Matkin, Suning Tang
2000 Optoelectronic Interconnects VII; Photonics Packaging and Integration II  
The design of an optical image guide network for distributed multiprocessing is described. The network supports multiple high bandwidth rings between workstations over distances of 10s of meters.  ...  The designs of efficient error and flow control protocols for such networks are proposed and analysed.  ...  The channels supporting long packets should use a "token ring" media access (MAC) protocol, rather then a slotted ring protocol, to minimize latency.  ... 
doi:10.1117/12.384387 fatcat:wftluuuuzff5hewalqjf24of54

Design and Evaluation of Hierarchical Rings with Deflection Routing

Rachata Ausavarungnirun, Chris Fallin, Xiangyao Yu, Kevin Kai-Wei Chang, Greg Nazario, Reetuparna Das, Gabriel H. Loh, Onur Mutlu
2014 2014 IEEE 26th International Symposium on Computer Architecture and High Performance Computing  
When inter-ring buffers are full, our design simply deflects flits so that they circle the ring and try again, which eliminates the need for in-ring buffering.  ...  Second, we introduce two simple mechanisms that together provide an end-to-end delivery guarantee within the entire network (despite any deflections that occur) without impacting the critical path or latency  ...  ACKNOWLEDGMENTS We thank the reviewers and SAFARI members for their feedback. We acknowledge the support of AMD, IBM, Intel, and Qualcomm.  ... 
doi:10.1109/sbac-pad.2014.31 dblp:conf/sbac-pad/AusavarungnirunFYCNDLM14 fatcat:oniwh2qpi5bg7owgscmw4jivu4

NOC based Router Architecture to Reduce Power using Round Robin Algorithm

Nanammal V
2020 International Journal for Research in Applied Science and Engineering Technology  
To schedule the processes, a round-robin scheduler generally allocate time for sharing, giving each job based on time slot (its allowance of CPU time) and interrupts the duty if it's not completed by them  ...  Deadlock and Livelock freedom in Routing is one of current issue in NoC routing. The Deadlock and livelock freedom prohibiting certain routing turns.  ...  Router architecture is designed in such a way that routers are forced to perform two major functions: Process routable protocol and routing protocol, to determine their best path.  ... 
doi:10.22214/ijraset.2020.5129 fatcat:uva2os4oyjehxm4j252uxg4vhm

Achieving both High Energy Efficiency and High Performance in On-Chip Communication using Hierarchical Rings with Deflection Routing [article]

Rachata Ausavarungnirun, Chris Fallin, Xiangyao Yu, Kevin Kai-Wei Chang, Greg Nazario, Reetuparna Das, Gabriel H. Loh, Onur Mutlu
2016 arXiv   pre-print
When inter-ring buffers are full, our design simply deflects flits so that they circle the ring and try again, which eliminates the need for in-ring buffering.  ...  Second, we introduce two simple mechanisms that provides an end-to-end delivery guarantee within the entire network without impacting the critical path or latency of the vast majority of network traffic  ...  Acknowledgments We thank the reviewers and SAFARI members for their feedback. We acknowledge the support of AMD, IBM, Intel, and Qualcomm.  ... 
arXiv:1602.06005v1 fatcat:xnudls56ozdhbfeo6cpgecqulq

A case for hierarchical rings with deflection routing: An energy-efficient on-chip communication substrate

Rachata Ausavarungnirun, Chris Fallin, Xiangyao Yu, Kevin Kai-Wei Chang, Greg Nazario, Reetuparna Das, Gabriel H. Loh, Onur Mutlu
2016 Parallel Computing  
When inter-ring buffers are full, our design simply deflects flits so that they circle the ring and try again, which eliminates the need for in-ring buffering.  ...  Second, we introduce two simple mechanisms that together provide an end-to-end delivery guarantee within the entire network (despite any deflections that occur) without impacting the critical path or latency  ...  Acknowledgments We thank the reviewers and SAFARI members for their feedback. We acknowledge the support of AMD, IBM, Intel, and Qualcomm.  ... 
doi:10.1016/j.parco.2016.01.009 fatcat:x5ia7fpsl5auhjs6h24qxbzg4y

Bufferless and Minimally-Buffered Deflection Routing [chapter]

Chris Fallin, Greg Nazario, Xiangyao Yu, Kevin Chang, Rachata Ausavarungnirun, Onur Mutlu
2013 Routing Algorithms in Networks-on-Chip  
Acknowledgments We thank the anonymous reviewers of our conference papers CHIPPER [17] and MinBD [20] for their feedback.  ...  We gratefully acknowledge members of the SAFARI group and Michael Papamichael at CMU for feedback. Chris Fallin is currently supported by an NSF Graduate Research Fellowship (Grant No. 0946825).  ...  Livelock and Deadlock-free Operation MinBD provides livelock-free delivery of flits using Golden Packet and buffer redirection.  ... 
doi:10.1007/978-1-4614-8274-1_10 fatcat:edzwfbsgo5fi7idw45nnswthya

Net Reductions for LTL Model-Checking [chapter]

Javier Esparza, Claus Schröter
2001 Lecture Notes in Computer Science  
We present a set of reduction rules for LTL model-checking of 1-safe Petri nets.  ...  Our reduction techniques are of two kinds: (1) Linear programming techniques which are based on well-known Petri net techniques like invariants and implicit places, and (2) local net reductions.  ...  We would like to thank Keijo Heljanko for valuable comments and sending us an implementation of the LTL model-checker.  ... 
doi:10.1007/3-540-44798-9_25 fatcat:jw73b2pqqjcahkzr5z5obkn47m

Shared-memory mutual exclusion: major research trends since 1986

James H. Anderson, Yong-Jik Kim, Ted Herman
2003 Distributed computing  
Dijkstra's algorithm, which is based on an earlier two-process algorithm by Dekker, is livelockfree but not starvation-free.  ...  Some authors use the term "deadlock-freedom" to denote the progress property termed livelock-freedom here.  ...  Acknowledgement: We are grateful to Faith Fich for her detailed comments on several earlier drafts of this paper.  ... 
doi:10.1007/s00446-003-0088-6 fatcat:g57iqdjp5neu7eu6sm5hpamohi

Atomic Coherence: Leveraging nanophotonics to build race-free cache coherence protocols

Dana Vantrease, Mikko H. Lipasti, Nathan Binkert
2011 2011 IEEE 17th International Symposium on High Performance Computer Architecture  
Once issued, requests follow a predictable race-free path. Because requests are guaranteed not to race, coherence protocols are simpler and protocol extensions are straightforward.  ...  We then show how, in the absence of races, it is easy to add support for speculative coherence and improve performance by up to 70%.  ...  Fig. 1 : 1 L2 Transitions for Two Protocols.  ... 
doi:10.1109/hpca.2011.5749723 dblp:conf/hpca/VantreaseLB11 fatcat:qjeosry5gnf2jog7gu7mmske6e

MinBD: Minimally-Buffered Deflection Routing for Energy-Efficient Interconnect

Chris Fallin, Greg Nazario, Xiangyao Yu, Kevin Chang, Rachata Ausavarungnirun, Onur Mutlu
2012 2012 IEEE/ACM Sixth International Symposium on Networks-on-Chip  
ACKNOWLEDGMENTS We thank the anonymous reviewers for their feedback. We gratefully acknowledge members of the SAFARI group and Michael Papamichael at CMU for feedback.  ...  Livelock and Deadlock-free Operation MinBD provides livelock-free delivery of flits using Golden Packet and buffer redirection.  ...  The (4, 1) point represents the minimum buffer size for deadlock-free operation (two message classes [16] , times two to avoid routing deadlock [20] ).  ... 
doi:10.1109/nocs.2012.8 dblp:conf/nocs/FallinNYCAM12 fatcat:lj2x4dei7rah7kiieguohuwece

Coherence Ordering for Ring-based Chip Multiprocessors

Michael R. Marty, Mark D. Hill
2006 2006 39th Annual IEEE/ACM International Symposium on Microarchitecture (MICRO'06)  
In this work, we propose a new class of ring protocols, RING-ORDER, in which requests complete in ring position order to achieve two benefits.  ...  Existing cache coherence protocols for rings either establish a (total) ordering point (ORDERING-POINT) or use a greedy order (GREEDY-ORDER) with unbounded retries.  ...  Finally we thank Virtutech, the Wisconsin Condor group, and the Wisconsin Computer Systems Lab for their support.  ... 
doi:10.1109/micro.2006.14 dblp:conf/micro/MartyH06 fatcat:hqykjtz7ozhxjdyfnht5rm4jxu
« Previous Showing results 1 — 15 out of 107 results