Filters








861 Hits in 3.7 sec

Brief announcement

Joseph Izraelevitz, Michael L. Scott
2014 Proceedings of the 26th ACM symposium on Parallelism in algorithms and architectures - SPAA '14  
In this paper, we introduce two new FIFO dual queues.  ...  Like all dual queues, they arrange for dequeue operations to block when the queue is empty, and to complete in the original order when data becomes available.  ...  SINGLE POLARITY DUAL RING QUEUE In our single polarity dual ring queue (SPDQ), each ring in the list has a single polarity-it can hold only data or only antidata.  ... 
doi:10.1145/2612669.2612711 dblp:conf/spaa/IzraelevitzS14 fatcat:wjxxk2geajay3jz75mb5weomhe

Generality and Speed in Nonblocking Dual Containers

Joseph Izraelevitz, Michael L. Scott
2017 ACM Transactions on Parallel Computing  
Section 4 (also a previous brief announcement [Izraelevitz and Scott 2014b] and a technical report [Izraelevitz and Scott 2014d]) presents FIFO dual queues that significantly improve on the performance  ...  Section 3 (earlier versions of which appeared as a brief announcement [Izraelevitz and Scott 2014a] and a technical report [Izraelevitz and Scott 2014c]) introduces a generic construction for building  ...  FC Dual Queue: A best-effort implementation of a flat-combining dual queue, using the methodology of Hendler et al. [2010b] . LCRQ: The fast total ring queue of Morrison and Afek [2013] .  ... 
doi:10.1145/3040220 fatcat:mlgs6rvma5e6rfx43ghntdrk5u

PROTECTORATION: a fast and efficient multiple-failure recovery technique for resilient packet ring using dark fiber

M. Maier, M. Herzog, M. Scheutzow, M. Reisslein
2005 Journal of Lightwave Technology  
The two protection methods wrapping and steering used in IEEE 802.17 resilient packet ring (RPR) provide fast but very inefficient and limited network failure recovery.  ...  The proposed protectoration technique 1) combines the fast recovery time of protection and the bandwidth efficiency of restoration, 2) provides full recovery from multiple link and node failures, 3) builds  ...  Similar to RPR, ring homed nodes operate in the single-queue mode or dual-queue mode.  ... 
doi:10.1109/jlt.2005.856165 fatcat:itlehtbbvbagtgd6woeqjbqnu4

RINGOSTAR: an evolutionary performance-enhancing WDM upgrade of IEEE 802.17 resilient packet ring

M. Herzog, M. Maier
2006 IEEE Communications Magazine  
Overview of RPR -RPR is an optical dual-fiber bidirectional ring network where each fiber ring carries a single wavelength channel.  ...  Specifically, for each ring a node has one or two transit queues for in-transit traffic, one transmission queue for locally generated data packets, one reception queue for packets destined for the local  ... 
doi:10.1109/mcom.2006.1593544 fatcat:2bbbxcufvrczbnl2fdulgqwbbm

Horizon

Božidar Radunović, Christos Gkantsidis, Dinan Gunawardena, Peter Key
2008 Proceedings of the 14th ACM international conference on Mobile computing and networking - MobiCom '08  
Performance Metrics Our test-bed has a form of a ring, and offers two disjoint paths between any node-pairs.  ...  In practice, we use a similar dual optimization problem as in Section 2 to design the algorithms for flow and next hop selection, but we optimize the price calculation to avoid excessive queueing; Section  ... 
doi:10.1145/1409944.1409973 dblp:conf/mobicom/RadunovicGGK08 fatcat:tu4smz2ohne5rp4jpddp5uobeq

A lossless switch for data acquisition networks

Grzegorz Jereczek, Giovanna Lehmann Miotto, David Malone, Miroslaw Walukiewicz
2015 2015 IEEE 40th Conference on Local Computer Networks (LCN)  
Output queueing (OQ) using hardware queues of a NIC in (a), virtual output queueing (VOQ) using hardware queues of a NIC in (b), and DAQ-dedicated queueing using software queues (DPDK rings -daqrings)  ...  Output queueing (OQ) using hardware queues of a NIC in (a), virtual output queueing (VOQ) using hardware queues of a NIC in (b), and DAQ-dedicated queueing using software queues (DPDK rings -daqrings)  ... 
doi:10.1109/lcn.2015.7366370 dblp:conf/lcn/JereczekMMW15 fatcat:ckogbydvvveoleqdyqlhtqv3wi

Mathematical decomposition techniques for distributed cross-layer optimization of data networks

B. Johansson, P. Soldati, M. Johansson
2006 IEEE Journal on Selected Areas in Communications  
In this paper, we review and explain three distinct techniques that can be used to engineer utility-maximizing protocols: primal, dual, and cross decomposition.  ...  The MVC dual subproblem (10) is identified with the S-TDMA dual subproblem (27) . Hence, the S-TDMA algorithm is an MVC algorithm and convergence follows from [20] .  ...  Similarly, dual decomposition can be seen as letting the and updates run on a fast time-scale, while the variables are updated slowly. V.  ... 
doi:10.1109/jsac.2006.879364 fatcat:53onmspkgnbylmi5eeg7xhqtrq

ABSTRACTING LOW-LEVEL APIS FOR CROSS-PLATFORM GRAPHICS APPLICATION DEVELOPMENT
ABSTRAGIROVANIE NIZKOUROVNEVYKh API PRI SOZDANII KROSSPLATFORMENNYKh GRAFIChESKIKh PRILOZhENII

V. D. Shakaev, О. А. Shabalina
2014 Vestnik komp iuternykh i informatsionnykh tekhnologii  
Fixed-size queue The fixed-size queue is internally implemented as a ring-buffer of elements, using separate indices for read and write operations.  ...  (Illustration after [18]) Figure 3 . 3 1 depicts a simplified design for a dual-core PC. Fig. 3. 1 : 1 Simplified overview of the PC architecture for a dual-core PC.  ... 
doi:10.14489/vkit.2014.08.pp.011-016 fatcat:haexayusvnfe3pf7dmo6oldija

Simulation of networks of spiking neurons: A review of tools and strategies [article]

R. Brette, M. Rudolph, T. Carnevale, M. Hines, D. Beeman, J. M. Bower, M. Diesmann, A. Morrison, P. H. Goodman, F. C. Harris Jr., M. Zirpe, T. Natschlager, D. Pecevski (+8 others)
2007 arXiv   pre-print
In the simple case of identical transmission delays, the data structure for the queue can be just a FIFO queue (first in, first out), which has fast implementations (Cormen et al, 2001) .  ...  Some connectivity patterns (e.g. all-to-all, ring, etc) are also included.  ... 
arXiv:q-bio/0611089v2 fatcat:wc22vb75zrb6xj66ngffn6gjuy

Access Networks [chapter]

2005 Connection-Oriented Networks  
Step decreases are announced through a management cell. A step decrease is always accepted.  ...  These congestion control schemes are known as fast bandwidth allocation schemes.  ... 
doi:10.1002/0470016361.ch11 fatcat:kah6edcdxjgbrpc4ihj7ccvrku

Simulation of networks of spiking neurons: A review of tools and strategies

Romain Brette, Michelle Rudolph, Ted Carnevale, Michael Hines, David Beeman, James M. Bower, Markus Diesmann, Abigail Morrison, Philip H. Goodman, Frederick C. Harris, Milind Zirpe, Thomas Natschläger (+10 others)
2007 Journal of Computational Neuroscience  
for the queue can be just a FIFO queue (First In, First Out) , which has fast implementations (Cormen et al. 2001) .  ...  Some connectivity patterns (e.g. all-to-all, ring, etc.) are also included.  ... 
doi:10.1007/s10827-007-0038-6 pmid:17629781 pmcid:PMC2638500 fatcat:ddtyniiynvcf3co43v5euiacty

Time-Predictable Computer Architecture

Martin Schoeberl
2009 EURASIP Journal on Embedded Systems  
To model this queue for a WCET tool, we need to cut the maximum history and assume an empty queue at such a cut point. In [33] , the authors model the 4-byte-long prefetch queue of an Intel 80188.  ...  Intel recently announced a CMP system named Larrabee [46] . Larrabee is intended as a replacement for graphic processing units from other vendors.  ... 
doi:10.1155/2009/758480 fatcat:4mxu4useyngezo5busxvnx3ize

Structural Overview of ISP Networks [chapter]

Robert D. Doverspike, K. K. Ramakrishnan, Chris Chase
2010 Computer Communications and Networks  
If an attribute of the route changes, the originating router announces it again, replacing the previous announcement.  ...  If we reexamine the above TDM metro access descriptions, we find that there are many restoration mechanisms, such as dual homing of the ARs to the BRs and SONET rings in the metro network.  ... 
doi:10.1007/978-1-84882-828-5_2 fatcat:s7uk64c2xbgpxj2k2vqmg2h3ae

Mastering Complex Control in MOBA Games with Deep Reinforcement Learning

Deheng Ye, Zhao Liu, Mingfei Sun, Bei Shi, Peilin Zhao, Hao Wu, Hongsheng Yu, Shaojie Yang, Xipeng Wu, Qingwei Guo, Qiaobo Chen, Yinyuting Yin (+6 others)
2020 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
Our algorithm includes several novel strategies, including control dependency decoupling, action mask, target attention, and dual-clip PPO, with which our proposed actor-critic network can be effectively  ...  The gradients in the RL learners are averaged through the ring allreduce algorithm (Sergeev and Balso 2018).  ...  In order to generate episodes efficiently, we build a CPU version of the fast inference library FeatherCNN 1 .  ... 
doi:10.1609/aaai.v34i04.6144 fatcat:cwwwohycizhybpflig24r2kev4

Mastering Complex Control in MOBA Games with Deep Reinforcement Learning [article]

Deheng Ye, Zhao Liu, Mingfei Sun, Bei Shi, Peilin Zhao, Hao Wu, Hongsheng Yu, Shaojie Yang, Xipeng Wu, Qingwei Guo, Qiaobo Chen, Yinyuting Yin (+6 others)
2020 arXiv   pre-print
Our algorithm includes several novel strategies, including control dependency decoupling, action mask, target attention, and dual-clip PPO, with which our proposed actor-critic network can be effectively  ...  In order to generate episodes efficiently, we build a CPU version of the fast inference library FeatherCNN 2 .  ...  The gradients in the RL learners are averaged through the ring allreduce algorithm (Sergeev and Balso 2018) .  ... 
arXiv:1912.09729v3 fatcat:wwra4rqnajer3cip3lkr4zgd64
« Previous Showing results 1 — 15 out of 861 results