Filters








3,773 Hits in 8.8 sec

"Hot spot" contention and combining in multistage interconnection networks

Gregory F. Pfister, V. Alan Norton
1985 IEEE transactions on computers  
The combining of messages within a multistage switching network has been proposed [1], [11], [14] to reduce memory contention in highly parallel shared-memory multi processors, especially for shared lock  ...  The potential degradation due to even moderate hot spot traffic was found to be very signifi cant, severely degrading all memory access, not just access to shared lock locations, due to an effect we call  ...  Wong, who performed numerous simulation experiments on the effects of hot spots and combining with a variety of switch designs.  ... 
doi:10.1109/tc.1985.6312198 fatcat:zhjdkwutvbfbhmziyv5by4b2di

Topology agnostic hot-spot avoidance with InfiniBand

Abhinav Vishnu, Matthew Koop, Adam Moody, Amith Mamidala, Sundeep Narravula, Dhabaleswar K. Panda
2009 Concurrency and Computation  
To make the matters worse, the deterministic routing nature of InfiniBand limits the application from an effective use of multiple paths transparently to avoid the hot-spot(s) in the network.  ...  Leveraging the LMC mechanism to avoid the hot-spot(s) in the network is the focus of this paper.  ...  The upper-level application may include programming models (MPI [3, 4] , distributed shared memory, global arrays and other PGAS languages).  ... 
doi:10.1002/cpe.1359 fatcat:hak5tghzcvg6rpb6cw4x4v5joa

Comparative performance evaluation of hot spot contention between MIN-based and ring-based shared-memory architectures

Xiaodong Zhang, Yong Yan, R. Castaneda
1995 IEEE Transactions on Parallel and Distributed Systems  
the presence of memory hot spots.  ...  a hot spot, Ti is an average delay of unsuccessful memory access in the presence of a hot spot, and &is an average delay time spent in state b in Fig. 1 .  ...  Since 1987, he has published extensively in the areas of parallel and distributed computing, performance evaluation, operating systems and algorithm analysis.  ... 
doi:10.1109/71.406963 fatcat:rfvjjxgr5zeh3evjf35gfdrnzy

Hot-Spot Avoidance With Multi-Pathing Over InfiniBand: An MPI Perspective

A. Vishnu, M. Koop, A. Moody, A. R. Mamidala, S. Narravula, D. K. Panda
2007 Seventh IEEE International Symposium on Cluster Computing and the Grid (CCGrid '07)  
To make matters worse, the deterministic routing nature of In-finiBand limits the application from effective use of multiple paths transparently and avoid the hot-spots in the network.  ...  In this paper, we design a new MPI functionality, which provides hot-spot avoidance for different communications, without apriori knowledge of the pattern.  ...  Hence, this policy benefits in the presence of hot-spots, however does not lead to performance degradation in their absence.  ... 
doi:10.1109/ccgrid.2007.60 dblp:conf/ccgrid/VishnuKMMNP07 fatcat:jlbhdiajkja7fmzpjm253nlqga

Using a multipath network for reducing the effects of hot spots

Mu-Cheng Wang, H.J. Siegel, M.A. Nichols, S. Abraham
1995 IEEE Transactions on Parallel and Distributed Systems  
Zndex Terms-Extra stage cube; hot spot; interconnection network; multistage interconnection network; parallel processing. Dr. Nichols is a member of the IEEE Computer Society and the ACM.  ...  This research investigates how the Extra Stage Cube multipath MIN can reduce the detrimental effects of tree saturation caused by hot spots.  ...  We thank the referees for their helpful comments.  ... 
doi:10.1109/71.372775 fatcat:xqyn7pumzbckrm2mydpyoe3axe

Avoiding hot-spots on two-level direct networks

Abhinav Bhatele, Nikhil Jain, William D. Gropp, Laxmikant V. Kale
2011 Proceedings of 2011 International Conference for High Performance Computing, Networking, Storage and Analysis on - SC '11  
The presence of multiple levels in this design leads to hot-spots on a few links when processes are grouped together at the lowest level to minimize total communication volume.  ...  IBM's PERCS topology and the dragonfly network discussed in the DARPA exascale hardware study are examples of this design.  ...  Acknowledgments This research was supported in part by the Blue Waters sustainedpetascale computing project (which is supported by the NSF grant OCI 07-25070 and the state of Illinois) and by a DOE Grant  ... 
doi:10.1145/2063384.2063486 dblp:conf/sc/BhateleJGK11 fatcat:m5fwcxh5hbe2jirouvmkwwj6zy

Decomposing Data-Centric Storage Query Hot-Spots in Sensor Networks

Mohamed Aly, Panos K. Chrysanthis, Kirk Pruhs
2006 2006 Third Annual International Conference on Mobile and Ubiquitous Systems: Networking & Services  
Arising when a large percentage of queries is accessing data stored in few sensor nodes, query hot-spots reduce the Quality of Data (QoD) and the lifetime of the sensor network.  ...  Experimental evaluation illustrates the efficiency of ZP/ZPR in decomposing query hot-spots while increasing QoD as well as energy savings by balancing energy consumption among sensor nodes.  ...  ZPR is meant to work in parallel with ZP such that the combination of ZP/ZPR can efficiently handle the decomposition task of query hot-spots of different sizes.  ... 
doi:10.1109/mobiq.2006.340396 dblp:conf/mobiquitous/AlyCP06 fatcat:olragv33vzf6tlu4qwiacnhtee

Decomposing Data-Centric Storage Query Hot-Spots in Sensor Networks

Mohamed Aly, Panos K. Chrysanthis, Kirk Pruhs
2006 2006 3rd Annual International Conference on Mobile and Ubiquitous Systems - Workshops  
Arising when a large percentage of queries is accessing data stored in few sensor nodes, query hot-spots reduce the Quality of Data (QoD) and the lifetime of the sensor network.  ...  Experimental evaluation illustrates the efficiency of ZP/ZPR in decomposing query hot-spots while increasing QoD as well as energy savings by balancing energy consumption among sensor nodes.  ...  ZPR is meant to work in parallel with ZP such that the combination of ZP/ZPR can efficiently handle the decomposition task of query hot-spots of different sizes.  ... 
doi:10.1109/mobiqw.2006.361728 fatcat:ydbhgg7oebdmnilcjgnnbjuqja

Per-station throughput fairness in a WLAN hot-spot with TCP traffic

A. Detti, N. Blefari-Melazzi, I. Habib, A. Ordine
2011 Computer Networks  
Our model accounts for the effect of several parameters and scenarios that no other model captures in a single analytic framework.  ...  First, we introduce two findings: i) changing the physical transmission rate of the stations does not impact the per-station fairness; ii) packet queuing occurs in the uplink buffer of a station and has  ...  Review of models on TCP fairness in 802.11 Hot-Spots The fairness performance can be evaluated as a ratio between goodputs of individual connections.  ... 
doi:10.1016/j.comnet.2011.01.004 fatcat:k5lkewma5bh25ji6exgfnnzsv4

Beyond Hot Spots: Biases in Antibody Somatic Hypermutation and Implications for Vaccine Design

Chaim A. Schramm, Daniel C. Douek
2018 Frontiers in Immunology  
Newer models that estimate the propensity for mutation for every possible 5-or 7-nucleotide motif have emphasized the complexity of SHM targeting and suggested possible new hot spot motifs.  ...  However, this comes at the cost of merging the effects of two distinct biological processes, the generation of mutations, and the selection acting on those mutations.  ...  The presence or absence of a microsequence hot spot also clearly impacts the observed differential substitution rates at some positions, such as S29 in VH5-51, which forms an (44) , but only about 50%  ... 
doi:10.3389/fimmu.2018.01876 pmid:30154794 fatcat:d7p5xabojnckhkms3rsan2retu

Comparative modeling and evaluation of CC-NUMA and COMA on hierarchical ring architectures

Xiaodong Zhang, Yong Yan
1995 IEEE Transactions on Parallel and Distributed Systems  
Parallel computing performance on scalable share& memory architectures is affected by the structure of the interconnection networks linking processors to memory modules and on the efficiency of the memory  ...  Cache Coherence Nonuniform Memory Access (CC-NUMA) and Cache Only Memory Access (COMA) are two effective memory systems, and the hierarchical ring structure is an efficient interco~ection network in hardware  ...  Effects on normal parallel computations in cool nodes In this experiment we measured the effects that a hot spot may have on a matrix multiplication application.  ... 
doi:10.1109/71.476171 fatcat:rgobzhwvnfcqjbd7mrumanmooe

Performance evaluation of I/O traffic and placement of I/O nodes on a high performance network

S. Coll, F. Petrini, E. Frachtenberg, A. Hoisie
2002 Proceedings 16th International Parallel and Distributed Processing Symposium  
A common trend in the design of large-scale clusters is to use a highperformance data network to integrate the processing nodes in a single parallel computer.  ...  The effects of using dedicated I/O nodes or overlapping I/O and computation on the I/O nodes are also analyzed.  ...  Acknowledgements The authors would like to thank the Quadrics team, David Addison, Jon Beecroft, Robin Crook, Moray McLaren, David Hewson, Duncan Roweth and John Taylor, for their invaluable support.  ... 
doi:10.1109/ipdps.2002.1016565 dblp:conf/ipps/CollPFH02 fatcat:kju6fjeq6zdqpo2c7r2jg4uhee

Human serine racemase structure/activity relationship studies provide mechanistic insight and point to position 84 as a hot spot for β-elimination function

David L. Nelson, Greg A. Applegate, Matthew L. Beio, Danielle L. Graham, David B. Berkowitz
2017 Journal of Biological Chemistry  
In the most extreme case of functional reprogramming, the S84D mutant displays a dramatic reversal of ␤-elimination substrate specificity in favor of L-serine over the normally preferred Lserine-O-sulfate  ...  Computational modeling pointed to a key role for residue Arg-135 in binding and properly positioning the L-THA and L-serine-O-sulfate substrates and the L-erythro-␤-hydroxyaspartate inhibitor.  ...  Acknowledgment-Facilities in this work were supported by a Grantin-Aid from the National Institutes of Health Grant RR016544.  ... 
doi:10.1074/jbc.m117.777904 pmid:28696262 pmcid:PMC5572919 fatcat:bgvs6lyoenafxdonadcmphf2zi

Dynamic Resource Allocation for Overload Avoidance and Green Cloud Computing

Saima Israil, Rajeev Pandey, Uday Chourasia
2015 International Journal of Computer Applications  
At present, load balancing is one of the challenging issues in cloud computing environment.  ...  Resource multiplexing is done through the virtualization technology in cloud computing. Virtualization technology acts as a backbone for provisioning requirements of the cloud based solutions.  ...  Hot spot Solver: The hot spot solver in VM Scheduler detects if the resource utilization of any PM is above the hot threshold (i.e., a hot spot).  ... 
doi:10.5120/ijca2015906234 fatcat:2zi3cyfycrfbhgaw43qrxlam2y

Evaluating and designing software mutual exclusion algorithms on shared-memory multiprocessors

X. Zhang, Y. Yan, R. Castaneda
1996 IEEE Parallel & Distributed Technology Systems & Applications  
A mutual exclusion operation generates a hot spot in a system. Hot-spot effects might degrade overall network traffic, not just the traffic to shared variables.  ...  Our previous research indicates that ring-network transactions in the KSR-1 decrease no more than SO% in the presence of memory hot spots.1o This compares with a 300% latency change in the TC2OOCi.  ... 
doi:10.1109/88.481663 fatcat:rdpg2oxmofas5cdeu5btuzhox4
« Previous Showing results 1 — 15 out of 3,773 results