Filters








7,025 Hits in 3.6 sec

More Analysis of Double Hashing for Balanced Allocations [article]

Michael Mitzenmacher
2015 arXiv   pre-print
For balanced allocations, keys are hashed into a hash table where each bucket can hold multiple keys, and each key is placed in the least loaded of d choices.  ...  It has been shown previously that asymptotically the performance of double hashing and fully random hashing is the same in the balanced allocation paradigm using fluid limit methods.  ...  We attempt to make an analogous argument in the double hashing setting for balanced allocations.  ... 
arXiv:1503.00658v1 fatcat:mkolh6dtvzbs3hdiakyrn4rstm

More Analysis of Double Hashing for Balanced Allocations

Michael Mitzenmacher
2015 2016 Proceedings of the Thirteenth Workshop on Analytic Algorithmics and Combinatorics (ANALCO)  
For balanced allocations, keys are hashed into a hash table where each bucket can hold multiple keys, and each key is placed in the least loaded of d choices.  ...  It has been shown previously that asymptotically the performance of double hashing and fully random hashing is the same in the balanced allocation paradigm using fluid limit methods.  ...  We attempt to make an analogous argument in the double hashing setting for balanced allocations.  ... 
doi:10.1137/1.9781611974324.1 dblp:conf/analco/Mitzenmacher16 fatcat:e3cru7b3nzbdjhufehr2tbjihm

Balanced Allocations and Double Hashing [article]

Michael Mitzenmacher
2014 arXiv   pre-print
Double hashing has recently found more common usage in schemes that use multiple hash functions.  ...  We first perform an empirical study showing that, surprisingly, the performance difference between double hashing and fully random hashing appears negligible in the standard balanced allocation paradigm  ...  Conclusion We have first demonstrated empirically that using double hashing with balanced allocation processes (e.g., the power of (more than) two choices), surprisingly, does not noticeably change performance  ... 
arXiv:1209.5360v4 fatcat:5mkninmy3nh35bwnilm4g3ipeu

Performance Analysis of a Load Balancing Hash-Join Algorithm for a Shared Memory Multiprocessor

Edward Omiecinski
1991 Very Large Data Bases Conference  
In this paper, we propose a new version of the hashbased join algorithm that balances the load between the processors, for any given bucket, in a shared everything environment.  ...  We develop an analytical model of the cost of the algorithm and implement the algorithm on a shared memory multiprocessor machine.  ...  For the double skew case, the decision should be made based on the combined sizes, Hence, more than 6 processors should be allocated to the skewed bucket.  ... 
dblp:conf/vldb/Omiecinski91 fatcat:a5i4gh6ybvhnxcn7aknsm5ewau

Data Structure: Theoretical Approach

Durgesh Raghuvanshi
2018 International Journal of Trend in Scientific Research and Development  
For example, relational databases commonly use B-tree indexes for data retrieval, while compiler implementations usually use hash tables to look up identifiers.  ...  Data structures serve as the basis for abstract data types (ADT). "The ADT defines the logical form of the data type. The data structure implements the physical form of the data type."  ...  AVL tree permits difference (balance factor) to be only 1.If the difference in the height of left and right sub-trees is more than 1, the tree is balanced using some rotation techniques.  ... 
doi:10.31142/ijtsrd18977 fatcat:zg6qkvkf2zdztj5fac4eds5wda

OpSparse: a Highly Optimized Framework for Sparse General Matrix Multiplication on GPUs [article]

Zhaoyang Du, Yijin Guan, Tianchan Guan, Dimin Niu, Linyong Huang, Hongzhong Zheng, Yuan Xie
2022 arXiv   pre-print
The state-of-the-art SpGEMM libraries (i.e., nsparse and spECK) adopt several algorithms to tackle the challenges of global load balance, local load balance, and allocation of the result matrix.  ...  global memory usage of the metadata, and 5) improving the execution parallelism by overlapping global memory allocation with kernel execution.  ...  These experiments numerically show the performance improvements of our proposed optimizations as compared to the inefficient implementations in the state-of-the-art SpGEMM libraries.  ... 
arXiv:2206.07244v1 fatcat:5afqbc36ibbrzgoerussz2trnq

Persistent Memory Hash Indexes: An Experimental Evaluation

Daokun Hu, Zhiwen Chen, Jianbing Wu, Jianhua Sun, Hao Chen
2021 Proceedings of the VLDB Endowment  
With our in-depth analysis, we identify design trade-offs and good paradigms in prior arts, and suggest desirable optimizations and directions for the future development of PM-based hash tables.  ...  In particular, we focus on the evaluation of six state-of-the-art hash tables including Level hashing, CCEH, Dash, PCLHT, Clevel, and SOFT, with real PM hardware.  ...  We thank the anonymous reviewers for their valuable suggestions in improving this paper.  ... 
dblp:journals/pvldb/HuCW0021 fatcat:xifixwwcirc6vpnpr2aak4od2a

Trustworthy computing for secure smart cities

Wathiq Mansoor, Vijayakumar Varadarajan
2021 Journal of Ambient Intelligence and Smart Environments  
The paper "M2FBalancer: A mist-assisted Fog computing-based load balancing strategy for smart cities" by Tripathy et al. implements an optimization strategy applying a dynamic resource allocation method  ...  We hope that researchers will benefit from the papers in this issue and find more motivation to pay attention to this important need.  ...  Acknowledgements Wathiq Mansoor wishes to express his gratitude to the University of Dubai for supporting faculty to contribute to international research collaborations.  ... 
doi:10.3233/ais-210597 fatcat:m2fpg4rkarewjcld22rcvkpywu

Tree-Chain: A Fast Lightweight Consensus Algorithm for IoT Applications [article]

Ali Dorri, Raja Jurdak
2020 arXiv   pre-print
code based on the hash of their public key.  ...  is selected randomly based on the most significant characters of the hash function output (known as consensus code), and ii) blockchain level where validator is randomly allocated to a particular consensus  ...  • Load balancing: This metric evaluates the impact of the load balancing algorithm by studying the processing overhead on val j before and after running the load balancing algorithm.• Double spending  ... 
arXiv:2005.09443v1 fatcat:2gd7k6cugff4xpkqbnihxd2evy

A Distributed Market Framework for Large-Scale Resource Sharing [chapter]

Marian Mihailescu, Yong Meng Teo
2010 Lecture Notes in Computer Science  
We use mechanism design to create a pricing scheme that allocates a request for multiple resource types, by trading economic efficiency for computational efficiency, strategy-proof and budget-balance.  ...  In this paper, we propose a scalable distributed market framework for the allocation of shared resources in large distributed systems.  ...  Acknowledgments This work is supported by the Singapore Ministry of Education under AcRF grant number R-252-000-339-112.  ... 
doi:10.1007/978-3-642-15277-1_40 fatcat:4fczx5xrzvg5ldbvfrg4ocuks4

Flow Management at Multi-Gbps: Tradeoffs and Lessons Learned [chapter]

Georges Nassopulos, Dario Rossi, Francesco Gringoli, Lorenzo Nava, Maurizio Dusi, Pedro Maria Santiago del Rio
2014 Lecture Notes in Computer Science  
hashing functions for keeping a table of flows and updating it by adding, searching and removing flows.  ...  [31] , to improve performance of generic traffic monitoring and analysis tools.  ...  Interestingly, the performance loss is more contained in case of balanced trees, which are inherently more robust to misconfigurations.  ... 
doi:10.1007/978-3-642-54999-1_1 fatcat:tgceyx56t5hafhcvdbujlxzecq

Extendible hashing---a fast access method for dynamic files

Ronald Fagin, Jurg Nievergelt, Nicholas Pippenger, H. Raymond Strong
1979 ACM Transactions on Database Systems  
We study, by analysis and simulation, the performance of extendible hashing.  ...  This approach simultaneously solves the problem of making hash tables that are extendible and of making radix search trees that are balanced.  ...  changing the O(1) access time characteristic of hashing towards the O(n) time characteristic of sequential allocation.  ... 
doi:10.1145/320083.320092 fatcat:5mf5c7vmcrbotezzqb6nkltqgu

High-performance sparse matrix-matrix products on Intel KNL and multicore architectures [article]

Yusuke Nagasaka, Satoshi Matsuoka, Ariful Azad, Aydın Buluç
2018 arXiv   pre-print
Although many SpGEMM algorithms have been proposed, hardware specific optimizations for multi- and many-core processors are lacking and a detailed analysis of their performance under various use cases  ...  A critical finding is that hash-table-based SpGEMM gets a significant performance boost if the nonzeros are not required to be sorted within each row of the output matrix.  ...  The other is that we allocate large enough memory space for output matrix and compute. The former requires more computation cost, and the latter uses much more memory space.  ... 
arXiv:1804.01698v2 fatcat:pqsq7dkfardhtkff5r7ic4fsue

High-Performance Sparse Matrix-Matrix Products on Intel KNL and Multicore Architectures

Yusuke Nagasaka, Satoshi Matsuoka, Ariful Azad, Aydın Buluç
2018 Proceedings of the 47th International Conference on Parallel Processing Companion - ICPP '18  
Although many SpGEMM algorithms have been proposed, hardware specific optimizations for multi-and many-core processors are lacking and a detailed analysis of their performance under various use cases and  ...  A critical finding is that hash-table-based SpGEMM gets a significant performance boost if the nonzeros are not required to be sorted within each row of the output matrix.  ...  The other is that we allocate large enough memory space for output matrix and compute. The former requires more computation cost, and the latter uses much more memory space.  ... 
doi:10.1145/3229710.3229720 dblp:conf/icppw/NagasakaMAB18 fatcat:5gjsb4a7bvdsdflnvp52adek4e

Partitioning Trillion-Edge Graphs in Minutes

George M. Slota, Sivasankaran Rajamanickam, Karen Devine, Kamesh Madduri
2017 2017 IEEE International Parallel and Distributed Processing Symposium (IPDPS)  
The empirical evaluation with various graphs shows that the proposed method produces higherquality partitions than the state-of-the-art distributed graph partitioning algorithms.  ...  The performance evaluation shows that the space efficiency of the proposed method is an orderof-magnitude better than the existing algorithms, keeping its time efficiency comparable.  ...  Each line type (solid, dotted, double) represents the initial partition for allocation process. Each color (blue, red, green) is the final partitioning result.  ... 
doi:10.1109/ipdps.2017.95 dblp:conf/ipps/SlotaRDM17 fatcat:osvwtvqz35dkdh6dhvhcemih4i
« Previous Showing results 1 — 15 out of 7,025 results