Filters








2,838 Hits in 3.3 sec

Brushwood: Distributed Trees in Peer-to-Peer Systems [chapter]

Chi Zhang, Arvind Krishnamurthy, Randolph Y. Wang
2005 Lecture Notes in Computer Science  
We show how to implement several p2p applications using distributed tree structures.  ...  At each lookup step, a search request is routed to a processor managing a search region close to the query point. Such regions are searched in the order of expanding dis- tances from the query point.  ...  Insertion and lookup of a point is straight-forward.  ... 
doi:10.1007/11558989_5 fatcat:zpgt3nlc6rbc5h7det2kx5nza4

F-Chord: Improved uniform routing on Chord

G. Cordasco, L. Gargano, A. Negro, V. Scarano, M. Hammar
2008 Networks  
The schemes, based on the Fibonacci number system, allow to improve on the maximum/average number of hops for lookups and the routing table size per node.  ...  The search schemes used in F-Chord is close to optimal but not for all values of N . Fig. 1 . 1 The three cases in the proof of Theorem 1.  ...  Our starting point is Fibonacci search [2] . Our results: Let F ib(i) be the i-th Fibonacci number.  ... 
doi:10.1002/net.20253 fatcat:dwz7vjiegrho3ipvkqpv5xmj5i

F-Chord: Improved Uniform Routing on Chord [chapter]

Gennaro Cordasco, Luisa Gargano, Mikael Hammar, Alberto Negro, Vittorio Scarano
2004 Lecture Notes in Computer Science  
The schemes, based on the Fibonacci number system, allow to improve on the maximum/average number of hops for lookups and the routing table size per node.  ...  The search schemes used in F-Chord is close to optimal but not for all values of N . Fig. 1 . 1 The three cases in the proof of Theorem 1.  ...  Our starting point is Fibonacci search [2] . Our results: Let F ib(i) be the i-th Fibonacci number.  ... 
doi:10.1007/978-3-540-27796-5_9 fatcat:6hpwbbv4n5a5bnzawr757qrtg4

Dynamic pipelining

Jahangir Hasan, T. N. Vijaykumar
2005 Computer communication review  
A truly scalable IP-lookup scheme must address five challenges of scalability, namely: routing-table size, lookup throughput, implementation cost, power dissipation, and routing-table update cost.  ...  We not only present a theoretical analysis but also evaluate SDP and a number of previous schemes using detailed hardware simulation.  ...  However, using the same stride is neither necessary nor optimal in terms of storage space or lookup delay.  ... 
doi:10.1145/1090191.1080116 fatcat:25posayqbbgmlgzavsnorxz2fu

Fast Longest Prefix Matching by Exploiting SIMD Instructions

Yukito Ueno, Ryo Nakamura, Yohei Kuga, Hiroshi Esaki
2020 IEEE Access  
Longest prefix matching (LPM) is a fundamental process in IP routing used not only in traditional hardware routers but also in software middleboxes.  ...  Moreover, Spider achieves 5,074 million lookups per second with 16 CPU cores, which is equivalent to the processing capacity of 3.4 Tbps in short packets; the performance opens up the possibility of packet  ...  In addition, we minimize the amount of searching on the routing table by employing an optimization called direct pointing [10] .  ... 
doi:10.1109/access.2020.3023156 fatcat:w4edreiwjve57mncd7msqm37f4

Dynamic pipelining

Jahangir Hasan, T. N. Vijaykumar
2005 Proceedings of the 2005 conference on Applications, technologies, architectures, and protocols for computer communications - SIGCOMM '05  
A truly scalable IP-lookup scheme must address five challenges of scalability, namely: routing-table size, lookup throughput, implementation cost, power dissipation, and routing-table update cost.  ...  We not only present a theoretical analysis but also evaluate SDP and a number of previous schemes using detailed hardware simulation.  ...  However, using the same stride is neither necessary nor optimal in terms of storage space or lookup delay.  ... 
doi:10.1145/1080091.1080116 dblp:conf/sigcomm/HasanV05 fatcat:4rzckocdfjfl7pwcupvpsivk2u

Approximate Geospatial Joins with Precision Guarantees

Andreas Kipf, Harald Lang, Varun Pandey, Raul Alexandru Persa, Peter Boncz, Thomas Neumann, Alfons Kemper
2018 2018 IEEE 34th International Conference on Data Engineering (ICDE)  
Our technique uses a quadtree-based hierarchical grid to approximate polygons and stores these approximations in a specialized radix tree.  ...  An especially challenging problem are joins between streaming points and static polygons. Since points are not known beforehand, they cannot be indexed.  ...  Likewise, lookups are in O(k) with k being the key length as opposed to the O(log n) runtime complexity of binary search that could be used on a sorted vector.  ... 
doi:10.1109/icde.2018.00150 dblp:conf/icde/KipfLPPB0K18 fatcat:6n6hzspda5eslgpuoq22nibahe

Cimple: Instruction and Memory Level Parallelism [article]

Vladimir Kiriansky, Haoran Xu, Martin Rinard, Saman Amarasinghe
2018 arXiv   pre-print
., by using wide superscalar pipelines and vector execution units, as well as deep buffers for in-flight memory requests.  ...  We demonstrate state-of-the-art performance on core algorithms used in in-memory databases that operate on arrays, hash tables, trees, and skip lists.  ...  Short-lived coroutines suffer from branch mispredictions on stage selection, and using a switch statement today leaves to compiler optimizations, preferably profile guided, to decide between using a jump  ... 
arXiv:1807.01624v1 fatcat:p57lcohyh5e2rdebufx5v2lxta

Rescore in a Flash: Compact, Cache Efficient Hashing Data Structures for n-Gram Language Models

Grant P. Strimel, Ariya Rastrow, Gautam Tiwari, Adrien Piérard, Jon Webb
2020 Interspeech 2020  
DashHashLM introduces several optimizations to language model compression which are designed to minimize expected memory accesses.  ...  We introduce DashHashLM, an efficient data structure that stores an n-gram language model compactly while making minimal trade-offs on runtime lookup latency.  ...  Use a minimal perfect hash function h * to lookup w's state id h * (w). In an offset index array I, lookup I[h * (w)] and I[h * (w) + 1] , the subarray boundaries for outgoing arcs from w. 4.  ... 
doi:10.21437/interspeech.2020-1939 dblp:conf/interspeech/StrimelRTPW20 fatcat:djxokflyo5dtdb2v6m7xloqgr4

Cimple: instruction and memory level parallelism

Vladimir Kiriansky, Haoran Xu, Martin Rinard, Saman Amarasinghe
2018 Proceedings of the 27th International Conference on Parallel Architectures and Compilation Techniques - PACT '18  
., by using wide superscalar pipelines and vector execution units, as well as deep buffers for inflight memory requests.  ...  We demonstrate state-of-the-art performance on core algorithms used in in-memory databases that operate on arrays, hash tables, trees, and skip lists.  ...  Short-lived coroutines suffer from branch mispredictions on stage selection, and using a switch statement today leaves to compiler optimizations, preferably profile guided, to decide between using a jump  ... 
doi:10.1145/3243176.3243185 dblp:conf/IEEEpact/KirianskyXRA18 fatcat:m7vzcvzjufcbbfu3it2vgsoxk4

Prefix Cuttings for Packet Classification with Fast Updates

2014 KSII Transactions on Internet and Information Systems  
Previous packet classification algorithms have mainly focused on search speed and memory usage, while overlooking update performance.  ...  Finally, we use the heuristic of information entropy-based bit partition to choose some specific bits of IP prefix to split the ruleset into subsets.  ...  Previous packet classification research centers on how to use heuristics to find the optimal way to locate the point [2] .  ... 
doi:10.3837/tiis.2014.04.016 fatcat:zqussnciwbcr5mw7onqeje55li

Benchmarking Learned Indexes [article]

Ryan Marcus, Andreas Kipf, Alexander van Renen, Mihail Stoian, Sanchit Misra, Alfons Kemper, Thomas Neumann, Tim Kraska
2020 arXiv   pre-print
Using four real-world datasets, we demonstrate that learned index structures can indeed outperform non-learned indexes in read-only in-memory workloads over a dense array.  ...  Hashing provides O(1) time point lookups.  ...  Next, RS performs a binary search between the two offsets on the sorted array of spline points to locate the two spline points that encompass the lookup key.  ... 
arXiv:2006.12804v2 fatcat:76kciq3s3vh4zbbsofgokusoqq

Silhouette lookup for monocular 3D pose tracking

Nicholas R. Howe
2007 Image and Vision Computing  
For silhouette retrieval, the experiments below use the chamfer distance between the boundary points of the observed silhouette and the boundary of the library silhouettes.  ...  Indeed, the metrics used can be computed rapidly enough that linear search is used for all of the experiments presented below.  ... 
doi:10.1016/j.imavis.2005.10.006 fatcat:ywmxcd6es5cbxjurztkk4pni5i

Concurrent Range Reporting in Two-Dimensional Space [chapter]

Peyman Afshani, Cheng Sheng, Yufei Tao, Bryan T. Wilkinson
2013 Proceedings of the Twenty-Fifth Annual ACM-SIAM Symposium on Discrete Algorithms  
., S L of points in R d with a total of N points.  ...  Finally, we give an optimal data structure for three-sided ranges for the case L = O(log N ).  ...  These boundary keys can be found efficiently using the down pointers. First, the (colorc) boundary group of root(T ) is retrieved by searching B root(T ) in O(log L) time.  ... 
doi:10.1137/1.9781611973402.73 dblp:conf/soda/AfshaniSTW14 fatcat:uoeftq3yrfdrpe753la4uoxfxy

BPF+

Andrew Begel, Steven McCanne, Susan L. Graham
1999 Proceedings of the conference on Applications, technologies, architectures, and protocols for computer communication - SIGCOMM '99  
The optimization phase of the compiler uses a flowgraph set relation called edge dominators and the novel application of an optimization technique that we call "redundant predicate elimination," in which  ...  code to migrate across protection boundaries, and a just-in-time assembler to convert byte codes to efficient native code.  ...  The original BPF architecture and optimization framework benefited from many fruitful design discussions with Van Jacobson, Vern Paxson, and Craig Leres.  ... 
doi:10.1145/316188.316214 dblp:conf/sigcomm/BegelMG99 fatcat:t6vagsxdv5efhi6nyp5k7w5lui
« Previous Showing results 1 — 15 out of 2,838 results