Two Levels Hashing Function in Consistent Hashing Algorithm

Size: px
Start display at page:

Download "Two Levels Hashing Function in Consistent Hashing Algorithm"

Transcription

1 1 Yihua Lan, 2Xiaopu Ma, 3Yong Zhang, 4Haozheng Ren, 5Chao Yin, 6Huaifei Hu School of Computer Engineering, Huaihai Institute of Technology, Lianyungang, China, 2 School of Computer and Information Technology, Nanyang Normal University, Nanyang, China, 3 School of Computer Engineering, Huaihai Institute of Technology, Lianyungang, China, 4 School of Computer Engineering, Huaihai Institute of Technology, Lianyungang, China, 5 School of Computer Science, Huazhong University of Science and Technology, Wuhan, P.R.China, *6,Corresponding Author School of Biological Science and Medical Engineering, South-entral University For Nationalities, Wuhan, P.R.China, 1 Abstract This article focused on consisting hashing algorithm. Traditional hashing function in consistent hashing is linear hashing, and linear hashing needs more virtual nodes to guarantee even data distribution. We designed a new hashing method two levels hashing. This new hashing method uses two hashing functions to eliminate the count of virtual nodes. First level hashing function transfer raw data to more even distributed data, and the second level hashing uses linear hashing method. After two levels hashing, data can distribute more plain than only linear hashing method used, so less virtual nodes can used. Our experimental results show that two levels hashing function can do better than current linear hashing function. Keywords: Hash Buckets, Consistent Hash, Data Distribution 1. Introduction In the circumstance of distributed systems, especially in cloud computing and cloud storage, data distribution is always in a dilemma: efficiency and balanced distribution [1-2]. Considering the efficiency, data need to be stored in the storage devices which as near as better, so less network band rate usage and less latency will occur. But for the device living time, all storages should be balanced access, and the data also should be distributed balanced. Xianhui Li et al. [3] proposed a new data distribution algorithm named consistent hashing to solve the problem of data balanced distribution. Consistent hashing can distribute data in a real number circle range, and for N storage devices, this range divided into N sub-ranges averagely. Every storage device was looked as a hash bucket, and each hash bucket occupies a sub-range. Consistent hashing algorithm has many virtues such as: distribute data evenly [4], flexible in adding more storage devices [5]. But for a linear hashing function, all the data cannot distribute evenly in a real number circle range. Normally, some sub-ranges have intensive data and some sub-ranges have little data. To solve this problem, the concept of virtual node was introduced [6]. Each hashing bucket is divided into M virtual nodes, and totally, N storage devices has M N virtual nodes. Experientially, M N must larger than 200. Too many virtual nodes affect the efficiency of the storage system. In this paper, we proposed a new hashing algorithm to decrease the number of virtual nodes. Our hashing algorithm uses two levels hashing function to map data into a real number circle range, and this method can map data more evenly than one level linear hashing function, so less virtual nodes are introduced. 2. Consistent hashing algorithm and linear hashing function Consistent hashing algorithm is widely used in distributed systems, such as Dynamo [7], Cassandra [8] and DHTbd [9]. International Journal of Digital Content Technology and its Applications(JDCTA) Volume6,Number20,November 2012 doi: /jdcta.vol6.issue

2 With a glance of consistent hashing algorithm, consistent hashing uses circle hash space to replace linear hash space in common hash algorithm. Suppose the algorithm uses 64 bits to form a circle hash space, then the circle space from 0 through 264-1, each object i is hashed as a number in the space, and each storage node X is hashed another number in the space, as Figure 1. Consistent hashing algorithm shows. Objects and nodes use the same hashing function, like equation (1) and equation (2) shows. Hash(object i ) key i (1) Hash(node X ) key X (2) Figure 1. 0~264-1 form a circle hash space and data objects and nodes are hashed into the same space Each node forms a bucket, and each data object stores in the node which is the first node when data object rotate clockwise, as figure 2. Data store in node shows. Figure 2. Data objects store in the node which is the first node in clockwise direction This simple algorithm has the fatal problem: data objects perhaps not distribute evenly in all the arcs which two nodes adjacent. This will cause different node stores different number of data objects. To solve this problem, virtual node is introduced. A real node is split into M virtual nodes, and all the M virtual nodes spread on the circle evenly. In Figure 3. A real node is split into 2 virtual nodes, two objects are stored in two virtual nodes: v2 and v4, these two virtual nodes point to the real node n1 and n2. 466

3 Figure 3. A real node is split into 2 virtual nodes, and data objects store in virtual nodes, mapping into real nodes. Virtual nodes can solve the problem of uneven distribution, as Figure 4. Data objects store in virtual nodes illustrates. In Figure 4(a). without virtual nodes, data object 1 is stored in node A, while data object 2, data object 3 and data object 4 are stored in node C, two nodes stores no equal number of data objects. In Figure4 (b). with virtual nodes, data object 1 is stored in virtual node A2, data object 2 is stored in virtual node A1, data object 3 is stored in virtual node C1 and data object 4 is stored in virtual node C2. Each virtual node stores 1 data object, and two virtual nodes mapped in a real node. So each real node stores 2 data objects. Data objects are distributed evenly in all real nodes. (a) Without virtual nodes (b) With virtual nodes Figure 4. Data objects store in virtual nodes, with virtual nodes, data distribute evenly in two real nodes Generally, for the purpose of even distribution, the circle should have 200~300 virtual nodes. And the hashing function is linear hashing function. Linear hashing functions are widely used in different hashing algorithm. For a key(k) of a data, common linear hashing function is use mod operate about K, that is: V K mod N (3) In equation (3), N buckets in the system and V is the bucket number which data K stored. In real storage system, consistent hashing algorithm has two versions: FNV-1 and FNV-1a, Sheepdog uses FNV-1a. The algorithm of FNV-1a is described as Algorithm 1. FNV-1a algorithm. 467

4 Algorithm 1. FNV-1a algorithm Require: offset_basis: a constant base number FNV_prime: a large prime Output: Hash: hashing value of a key(data object or node) 1. Hash = offset_basis; 2. for each key to be hashed 3. { 4. Hash = Hash xor key; 5. Hash = Hash * FNV_prime; 6. } 7. return Hash; Different real system can chose different offset_basis and FNV_prime, for Sheepdog, the two parameter is offset_basis=0xcbf29ce ull and FNV_prime= xb3. So the expression of FNV-1a in Sheepdog is linear, that means the data distribution on the circle will be the same with data trace. If we want to eliminate the number of virtual nodes, we should smooth the data distribution and let it be more even. 3. Two levels hashing function Two levels hashing function can smooth the data trace to be more even. C.Raczy et al. [10] proved all the data distribution can be transferred into Gaussian Distribution. For Gaussian distribution, center data is density and other zone is sparse. We can use first level hashing function to separate density data, and the second level hashing function uses linear hashing function to map data area into circle area First level hashing function of data objects The most intuitionistic method of transfer Gaussian Distributed data into even data is use reverse function of Gaussian Distribution. But the reverse function is not easy to be calculated, so the alternative is use another function. Tangent function has the similar distribution of Gaussian Distribution, though we can use the reverse function of tangent function to transfer data into even distribution. There are some problems to use reverse function of tangent. The first one is which is the center of data? There are two solutions: the first one is adjusting the data center according to the history, and another is set data center manual. The first solution requires 1 day data history at least, and marks these data with a flag. These data use only linear hashing function. The other data can use two levels hashing function to map. When read or write data, if I/O found the flag is true, linear hashing function is used, while the flag is false, two levels hashing function are used. The other solution requires administrator set the data center, and the administrator should be sophisticated. Right data center will achieve the purpose and wrong data center can only cause data distribution more badly. The second problem the reverse function of tangent s value area is π/2 through π/2, and linear hashing function uses these values as input parameters, so well defined linear hashing function should be chosen. The third problem is deal with fraction. Reverse function of tangent will cause fraction, but in the circle of consistent hashing algorithm, integers are required. The strategy of transfer fractions to integers is cut the number less than certain number. It has no effect when data set is large enough. We use data s address A as the key, and A is from 0 through the whole capacity of storage, suppose it is T. We choose C as the data center, and use a zoom factor εto adjust address range, so the key is: key A C (4) The first level hashing function can calculate the result Hash1: 468

5 Hash1 arctan(key ) (5) 3.2. Second level hashing function of data objects The second level hashing function is linear hashing function. It clips a parameter from π/2 through π/2 to a field [0, 2R-1]. As section 3.1illustrated, the parameters should cut the number less than a certain number S, for more accuracy, the certain number should more less. For example, the certain number S is , and if the parameter is π/4, it will be For easy calculation, S should be 10-I, and I is a integer larger than 0. The easiest linear hashing function is as equation (3) shows, so we can use this equation as our second level hashing function. Hash (Hash1 / 2) 10I mod 2R (6) 3.3. Virtual nodes hashing Two levels hashing function also need virtual nodes, but less than linear hashing. Virtual nodes distribute evenly on the circle area. We can use another simple way to distribute them. Suppose there are N real nodes, and every real nodes has M virtual nodes, so every virtual node has an extent of 2R/(M N). For real node i, its virtual node j, the hash value is: HashNode j N i 2R M N (7) For R = 160, N = 4, M = 8, the circle will be as Figure 5. Virtual nodes distribution on the ring shows. Figure 5. Virtual nodes distribution on the ring 4. Experiments and results We choose different data trace to show that our method will be less virtual nodes than traditional hashing function in consistent hashing algorithm Traces used in the article All the trace used in the article is block level trace: TPC-C is a database trace, which collected by the Performance Evaluation laboratory at Brigham Young University [11]. MSR-Cambridge is a serial 469

6 of traces provided by Microsoft researchers [12]. MSR-Cambridge has many volume-based sub-traces, we choose Hm, which hardware monitoring server, volume 0; MDS, which is Media server, volume 0 and prxy, which is Firewall/web proxy server, volume 0. Table 1. Traces summarize shows the characters of different traces. Trace name Lowest data Address TPC-C HM Prxy MDS Table 1. Traces summarize Highest data Address Total request count Following 4 figures illustrate requested data address distributions of four traces. Four numbers in the legend is the address interval of sample. For example, TPC-C has 200 requests have the same data address between 1200 and 1300, then the Request Count is 200 when Data Address is 1200, and this counterpart legend is 100. From the figures, we can see all data are distributed with Gaussian Distribution, and we can use two levels hashing function to relocate all the data. First thing to do is confirm the data center. We can manual set data center as Table 2. Data center of traces. Here all the data centers are proper data centers. Trace name TPC-C HM Prxy MDS Table 2. Data center of Traces Data center address Figure 6. Data distribution of TPC-C Figure 7. Data distribution of HM 470

7 Figure 8. Data distribution of Prxy Figure 9. Data distribution of MDS 4.2. Compare of two levels hashing function and linear hashing function We use our two levels hashing function compared with Sheepdog s FNV-1a hashing function. The first level hashing function used here according to equation (4) and (5), the second level hashing function we used here accord equation (6). Virtual nodes hashing we used here all according equation (7). Although Sheepdog use different virtual nodes hashing function with equation (7), but for comparative, we also use equation (7) here to guarantee all the virtual nodes are distributed evenly on the circle. For testing the virtual node number, we choose 4 real nodes, that is node0, node1, node2 and node3. Each real node has the same count of virtual nodes. In the experiments, we use VN = 8, VN = 20 and VN = 50 to compare different virtual nodes count. 471

8 Figure 10. Data distribution in real nodes with the trace of TPC-C Figure 11. Data distribution in real nodes with the trace of HM Figure 12. Data distribution in real nodes with the trace of Prxy 472

9 Figure 13. Data distribution in real nodes with the trace of MDS The upper 4 figures show that at the same virtual nodes count, two levels hashing can achieve more even data distribution than linear hashing. In another words, with the same distribution, two level hashing need less virtual nodes than linear hashing Parameters effects on experimental results Parameters can affect experimental results. First parameter affect two hashing methods is circle range, here is identified by R. If R is large enough, all the data addresses will less than 2R, and mod operation will not affect the result, so more virtual node can achieve even distribution. In equation (4), we use a zoom factorεto adjust address range, this is very important. If key larger than 100, all the value of arctan(key) will close to π/2, so the data will fall in a single virtual node, that is to say, most data will be stored in a real node, this is not the purpose of our design. And badly, almost all the data address is larger than C+100. We should use this zoom factor to adjust the data address. To choose this zoom factor, we should know the data address edge of a trace. For example, in Figure 6. Data distribution of TPC-C, the address can be seen as the edge of intensive requests and sparse requests, and the data center of TPC-C is , the difference of edge and center is , so the εcan be set as 1/ The third parameter can be seen in equation (6). I also can be seen as a zoom factor to enlarge data address. For the value of Hash1 is from -π/2 through π/2, so concentrated data address will cause the problem that all the data fall in the same virtual node. Proper selected I can enlarge data address to proper range. Generally speaking, I cannot be too small, but larger I affect little on the results. We choose I=5 or 6 in experiments. 5. Conclusions From the experimental results, we can see two levels hashing can obviously decrease the count of virtual nodes in consistent hashing. Less virtual nodes bring the benefit of efficiency, but two levels hashing should be more calculation. Function arctan(x) is a time consuming operation. Time consuming is for CPU, but for storage devices, more even cause more living time. 6. References [1] Anmin Ding, "On the MGIS-Oriented Design Web Service", JDCTA: International Journal of Digital Content Technology and its Applications, Vol. 5, No. 6, pp. 8 ~ 15, 2011 [2] Xiujuan LI Tao CAI Shiguang JU, "The Hash Postfix Table Based Metadata Management Algorithm for Mass Storage System", JDCTA: International Journal of Digital Content Technology and its Applications, Vol. 4, No. 9, pp. 97 ~ 105,

10 [3] Xianhui Li, Cuihua Ren, Menglong Yue, A Distributed Real-time Database Index Algorithm Based on B+ Tree and Consistent Hashing, In International Conference on Advances in Engineering 2011, Volume 24, pp , 2011 [4] Zhenbin Yan, Wenzheng Li, Research of a scheduling and load balancing scheme based on large-scale distributed systems, in Software Engineering and Service Science (ICSESS), 2012 IEEE 3rd International Conference, pp 22-24, June 2012 [5] Bong Jun Ko, Vasileios Pappas, Ramya Raghavendra, Yang Song, Raheleh B. Dilmaghani, Kangwon Lee, Dinesh Verma, An information-centric architecture for data center networks, in ICN '12 Proceedings of the second edition of the ICN workshop on Information-centric networking, pp 79-84, 2012 [6] Thorsten Schutt, Florian Schintke, Alexander Reinefeld, Structured Overlay without Consistent Hashing: Empirical Results, in the Sixth IEEE International Symposium of Cluster Computing and the Grid, CCGRID 06, 2006 [7] Giuseppe DeCandia, Deniz Hastorun, Madan Jampani, Gunavardhan Kakulapati, Avinash Lakshman, Alex Pilchin, Swaminathan Sivasubramanian, Peter Vosshall and Werner Vogels, Dynamo: Amazon s Highly Available Key-value Store, In Proceedings of twenty-first ACM SIGOPS symposium on Operating systems principles, pp , 2007 [8] Avinash Lakshman, Prashant Malik, Cassandra-a decentralized structured storage system, in ACM SIGOPS Operating systems review, Volume 44, issue 2, pp , 2010 [9] George Parisis, George Xylomenos and Theodore Apostolopoulos, DHTbd: A Reliable Blockbased Storage System for High Performance Clusters, In Proceedings of IEEE/ACM CCGRID,2011 [10] C. Raczy, G.Tan and J. Yu, A Sort-Based DDM Matching Algorithm for HLA, Journal of ACM Transactions on Modeling and Computer Simulation(TOMACS) Volume 15, Issue 1, pp , January 2005 [11] TPC-C block i/o trace. Performance Evaluation Laboratory, Brigham Young University, [12] D. Narayanan, A. Donnelly, and A. Rowstron, Write off-loading: Practical Power Management for Enterprise Storage, ACM Transactions on Storage, vol. 4, pp. 1 23, Nov

Dynamo: Amazon s Highly Available Key-value Store

Dynamo: Amazon s Highly Available Key-value Store Dynamo: Amazon s Highly Available Key-value Store Giuseppe DeCandia, Deniz Hastorun, Madan Jampani, Gunavardhan Kakulapati, Avinash Lakshman, Alex Pilchin, Swaminathan Sivasubramanian, Peter Vosshall and

More information

Dynamo: Amazon s Highly Available Key-value Store

Dynamo: Amazon s Highly Available Key-value Store Dynamo: Amazon s Highly Available Key-value Store Giuseppe DeCandia, Deniz Hastorun, Madan Jampani, Gunavardhan Kakulapati, Avinash Lakshman, Alex Pilchin, Swaminathan Sivasubramanian, Peter Vosshall and

More information

The material in this lecture is taken from Dynamo: Amazon s Highly Available Key-value Store, by G. DeCandia, D. Hastorun, M. Jampani, G.

The material in this lecture is taken from Dynamo: Amazon s Highly Available Key-value Store, by G. DeCandia, D. Hastorun, M. Jampani, G. The material in this lecture is taken from Dynamo: Amazon s Highly Available Key-value Store, by G. DeCandia, D. Hastorun, M. Jampani, G. Kakulapati, A. Lakshman, A. Pilchin, S. Sivasubramanian, P. Vosshall,

More information

Load Balancing Technology Based On Consistent Hashing For Database Cluster Systems

Load Balancing Technology Based On Consistent Hashing For Database Cluster Systems National Conference on Information Technology and Computer Science (CITCS 2012) Load Balancing Technology Based On Consistent Hashing For Database Cluster Systems Zhenguo Xuan Dept. of Computer Science,

More information

A Cloud Storage Adaptable to Read-Intensive and Write-Intensive Workload

A Cloud Storage Adaptable to Read-Intensive and Write-Intensive Workload DEIM Forum 2011 C3-3 152-8552 2-12-1 E-mail: {nakamur6,shudo}@is.titech.ac.jp.,., MyCassandra, Cassandra MySQL, 41.4%, 49.4%.,, Abstract A Cloud Storage Adaptable to Read-Intensive and Write-Intensive

More information

CS555: Distributed Systems [Fall 2017] Dept. Of Computer Science, Colorado State University

CS555: Distributed Systems [Fall 2017] Dept. Of Computer Science, Colorado State University CS 555: DISTRIBUTED SYSTEMS [REPLICATION & CONSISTENCY] Frequently asked questions from the previous class survey Shrideep Pallickara Computer Science Colorado State University L25.1 L25.2 Topics covered

More information

GreenCHT: A Power-Proportional Replication Scheme for Consistent Hashing based Key Value Storage Systems

GreenCHT: A Power-Proportional Replication Scheme for Consistent Hashing based Key Value Storage Systems GreenCHT: A Power-Proportional Replication Scheme for Consistent Hashing based Key Value Storage Systems Nannan Zhao 12, Jiguang Wan 12, Jun Wang 3, and Changsheng Xie 12 1 Wuhan National Laboratory for

More information

CS 223 Final Project CuckooRings: A Data Structure for Reducing Maximum Load in Consistent Hashing

CS 223 Final Project CuckooRings: A Data Structure for Reducing Maximum Load in Consistent Hashing CS 223 Final Project CuckooRings: A Data Structure for Reducing Maximum Load in Consistent Hashing Jonah Kallenbach and Ankit Gupta May 2015 1 Introduction Cuckoo hashing and consistent hashing are different

More information

Cassandra- A Distributed Database

Cassandra- A Distributed Database Cassandra- A Distributed Database Tulika Gupta Department of Information Technology Poornima Institute of Engineering and Technology Jaipur, Rajasthan, India Abstract- A relational database is a traditional

More information

CS555: Distributed Systems [Fall 2017] Dept. Of Computer Science, Colorado State University

CS555: Distributed Systems [Fall 2017] Dept. Of Computer Science, Colorado State University CS 555: DISTRIBUTED SYSTEMS [DYNAMO & GOOGLE FILE SYSTEM] Frequently asked questions from the previous class survey What s the typical size of an inconsistency window in most production settings? Dynamo?

More information

CSE-E5430 Scalable Cloud Computing Lecture 10

CSE-E5430 Scalable Cloud Computing Lecture 10 CSE-E5430 Scalable Cloud Computing Lecture 10 Keijo Heljanko Department of Computer Science School of Science Aalto University keijo.heljanko@aalto.fi 23.11-2015 1/29 Exam Registering for the exam is obligatory,

More information

CS655: Advanced Topics in Distributed Systems [Fall 2013] Dept. Of Computer Science, Colorado State University

CS655: Advanced Topics in Distributed Systems [Fall 2013] Dept. Of Computer Science, Colorado State University CS 655: ADVANCED TOPICS IN DISTRIBUTED SYSTEMS Shrideep Pallickara Computer Science Colorado State University PROFILING HARD DISKS L4.1 L4.2 Characteristics of peripheral devices & their speed relative

More information

Reprise: Stability under churn (Tapestry) A Simple lookup Test. Churn (Optional Bamboo paper last time)

Reprise: Stability under churn (Tapestry) A Simple lookup Test. Churn (Optional Bamboo paper last time) EECS 262a Advanced Topics in Computer Systems Lecture 22 Reprise: Stability under churn (Tapestry) P2P Storage: Dynamo November 20 th, 2013 John Kubiatowicz and Anthony D. Joseph Electrical Engineering

More information

Distributed Hash Tables Chord and Dynamo

Distributed Hash Tables Chord and Dynamo Distributed Hash Tables Chord and Dynamo (Lecture 19, cs262a) Ion Stoica, UC Berkeley October 31, 2016 Today s Papers Chord: A Scalable Peer-to-peer Lookup Service for Internet Applications, Ion Stoica,

More information

Evaluation of the Huawei UDS cloud storage system for CERN specific data

Evaluation of the Huawei UDS cloud storage system for CERN specific data th International Conference on Computing in High Energy and Nuclear Physics (CHEP3) IOP Publishing Journal of Physics: Conference Series 53 (4) 44 doi:.88/74-6596/53/4/44 Evaluation of the Huawei UDS cloud

More information

DISTRIBUTED FILE SYSTEMS CARSTEN WEINHOLD

DISTRIBUTED FILE SYSTEMS CARSTEN WEINHOLD Department of Computer Science Institute of System Architecture, Operating Systems Group DISTRIBUTED FILE SYSTEMS CARSTEN WEINHOLD OUTLINE Classical distributed file systems NFS: Sun Network File System

More information

A New Architecture of HSS Based on Cloud Computing

A New Architecture of HSS Based on Cloud Computing A New Architecture of HSS Basedon Cloud Computing Tianpu Yang, Xiangming Wen, Yong Sun, Zhenmin Zhao, Yuedui Wang School of Information and Communication Engineering Beijing University of Posts and Telecommunications

More information

DISTRIBUTED FILE SYSTEMS CARSTEN WEINHOLD

DISTRIBUTED FILE SYSTEMS CARSTEN WEINHOLD Department of Computer Science Institute of System Architecture, Operating Systems Group DISTRIBUTED FILE SYSTEMS CARSTEN WEINHOLD OUTLINE Classical distributed file systems NFS: Sun Network File System

More information

An Efficient Distributed B-tree Index Method in Cloud Computing

An Efficient Distributed B-tree Index Method in Cloud Computing Send Orders for Reprints to reprints@benthamscience.ae The Open Cybernetics & Systemics Journal, 214, 8, 32-38 32 Open Access An Efficient Distributed B-tree Index Method in Cloud Computing Huang Bin 1,*

More information

Decentralized Approach for Balancing Load in Dynamic Cloud Environment

Decentralized Approach for Balancing Load in Dynamic Cloud Environment Decentralized Approach for Balancing Load in Dynamic Cloud Environment Karthick Smiline Britto.J 1, PrittoPaul.P 2 ME, Department of CSE, Velammal Engineering College, Anna University, Chennai, India 1

More information

A Review to the Approach for Transformation of Data from MySQL to NoSQL

A Review to the Approach for Transformation of Data from MySQL to NoSQL A Review to the Approach for Transformation of Data from MySQL to NoSQL Monika 1 and Ashok 2 1 M. Tech. Scholar, Department of Computer Science and Engineering, BITS College of Engineering, Bhiwani, Haryana

More information

Improving Logical Clocks in Riak with Dotted Version Vectors: A Case Study

Improving Logical Clocks in Riak with Dotted Version Vectors: A Case Study Improving Logical Clocks in Riak with Dotted Version Vectors: A Case Study Ricardo Gonçalves Universidade do Minho, Braga, Portugal, tome@di.uminho.pt Abstract. Major web applications need the partition-tolerance

More information

NOSQL DATABASE PERFORMANCE BENCHMARKING - A CASE STUDY

NOSQL DATABASE PERFORMANCE BENCHMARKING - A CASE STUDY STUDIA UNIV. BABEŞ BOLYAI, INFORMATICA, Volume LXIII, Number 1, 2018 DOI: 10.24193/subbi.2018.1.06 NOSQL DATABASE PERFORMANCE BENCHMARKING - A CASE STUDY CAMELIA-FLORINA ANDOR AND BAZIL PÂRV Abstract.

More information

Evaluating Auto Scalable Application on Cloud

Evaluating Auto Scalable Application on Cloud Evaluating Auto Scalable Application on Cloud Takashi Okamoto Abstract Cloud computing enables dynamic scaling out of system resources, depending on workloads and data volume. In addition to the conventional

More information

CAP Theorem, BASE & DynamoDB

CAP Theorem, BASE & DynamoDB Indian Institute of Science Bangalore, India भ रत य व ज ञ न स स थ न ब गल र, भ रत DS256:Jan18 (3:1) Department of Computational and Data Sciences CAP Theorem, BASE & DynamoDB Yogesh Simmhan Yogesh Simmhan

More information

L22: NoSQL. CS3200 Database design (sp18 s2) 4/5/2018 Several slides courtesy of Benny Kimelfeld

L22: NoSQL. CS3200 Database design (sp18 s2)   4/5/2018 Several slides courtesy of Benny Kimelfeld L22: NoSQL CS3200 Database design (sp18 s2) https://course.ccs.neu.edu/cs3200sp18s2/ 4/5/2018 Several slides courtesy of Benny Kimelfeld 2 Outline 3 Introduction Transaction Consistency 4 main data models

More information

A Multipath AODV Reliable Data Transmission Routing Algorithm Based on LQI

A Multipath AODV Reliable Data Transmission Routing Algorithm Based on LQI Sensors & Transducers 2014 by IFSA Publishing, S. L. http://www.sensorsportal.com A Multipath AODV Reliable Data Transmission Routing Algorithm Based on LQI 1 Yongxian SONG, 2 Rongbiao ZHANG and Fuhuan

More information

Introduction to store data in Redis, a persistent and fast key-value database

Introduction to store data in Redis, a persistent and fast key-value database AMICT 2010-2011. pp. 39 49 39 Introduction to store data in Redis, a persistent and fast key-value database Matti Paksula Department of Computer Science, University of Helsinki P.O.Box 68, FI-00014 University

More information

Research on Heterogeneous Communication Network for Power Distribution Automation

Research on Heterogeneous Communication Network for Power Distribution Automation 3rd International Conference on Material, Mechanical and Manufacturing Engineering (IC3ME 2015) Research on Heterogeneous Communication Network for Power Distribution Automation Qiang YU 1,a*, Hui HUANG

More information

International Journal of Advanced Networking & Applications (IJANA) ISSN:

International Journal of Advanced Networking & Applications (IJANA) ISSN: Secured Load Re-Deployment of File Chunks in Distributed File System Roopadevi D S, Mrs. Nandini G B.E,M.Tech(Ph.D) PG Student, Assistant Professor,Department of Computer Science, RajaRajeswari College

More information

Lecture 18: Peer-to-Peer applications

Lecture 18: Peer-to-Peer applications Lecture 18: Peer-to-Peer applications Anirudh Sivaraman 2017/12/10 In this lecture, we ll talk about peer-to-peer (P2P) applications. These are applications in which all end hosts are treated equally:

More information

Hashing Techniques. Material based on slides by George Bebis

Hashing Techniques. Material based on slides by George Bebis Hashing Techniques Material based on slides by George Bebis https://www.cse.unr.edu/~bebis/cs477/lect/hashing.ppt The Search Problem Find items with keys matching a given search key Given an array A, containing

More information

Fast K-nearest neighbors searching algorithms for point clouds data of 3D scanning system 1

Fast K-nearest neighbors searching algorithms for point clouds data of 3D scanning system 1 Acta Technica 62 No. 3B/2017, 141 148 c 2017 Institute of Thermomechanics CAS, v.v.i. Fast K-nearest neighbors searching algorithms for point clouds data of 3D scanning system 1 Zhang Fan 2, 3, Tan Yuegang

More information

ProRenaTa: Proactive and Reactive Tuning to Scale a Distributed Storage System

ProRenaTa: Proactive and Reactive Tuning to Scale a Distributed Storage System ProRenaTa: Proactive and Reactive Tuning to Scale a Distributed Storage System Ying Liu, Navaneeth Rameshan, Enric Monte, Vladimir Vlassov, and Leandro Navarro Ying Liu; Rameshan, N.; Monte, E.; Vlassov,

More information

Video annotation based on adaptive annular spatial partition scheme

Video annotation based on adaptive annular spatial partition scheme Video annotation based on adaptive annular spatial partition scheme Guiguang Ding a), Lu Zhang, and Xiaoxu Li Key Laboratory for Information System Security, Ministry of Education, Tsinghua National Laboratory

More information

SOLVING LOAD REBALANCING FOR DISTRIBUTED FILE SYSTEM IN CLOUD

SOLVING LOAD REBALANCING FOR DISTRIBUTED FILE SYSTEM IN CLOUD 1 SHAIK SHAHEENA, 2 SD. AFZAL AHMAD, 3 DR.PRAVEEN SHAM 1 PG SCHOLAR,CSE(CN), QUBA ENGINEERING COLLEGE & TECHNOLOGY, NELLORE 2 ASSOCIATE PROFESSOR, CSE, QUBA ENGINEERING COLLEGE & TECHNOLOGY, NELLORE 3

More information

The Improvement and Implementation of the High Concurrency Web Server Based on Nginx Baiqi Wang1, a, Jiayue Liu2,b and Zhiyi Fang 3,*

The Improvement and Implementation of the High Concurrency Web Server Based on Nginx Baiqi Wang1, a, Jiayue Liu2,b and Zhiyi Fang 3,* Computing, Performance and Communication systems (2016) 1: 1-7 Clausius Scientific Press, Canada The Improvement and Implementation of the High Concurrency Web Server Based on Nginx Baiqi Wang1, a, Jiayue

More information

Enhancing the Query Performance of NoSQL Datastores using Caching Framework

Enhancing the Query Performance of NoSQL Datastores using Caching Framework Enhancing the Query Performance of NoSQL Datastores using Caching Framework Ruchi Nanda #1, Swati V. Chande *2, K.S. Sharma #3 #1,# 3 Department of CS & IT, The IIS University, Jaipur, India *2 Department

More information

A Security Audit Module for HBase

A Security Audit Module for HBase 2016 Joint International Conference on Artificial Intelligence and Computer Engineering (AICE 2016) and International Conference on Network and Communication Security (NCS 2016) ISBN: 978-1-60595-362-5

More information

Robust Frequency-tuned Salient Region Detection

Robust Frequency-tuned Salient Region Detection Robust Frequency-tuned Salient Region Detection 1 Yong Zhang, 2 Yihua Lan, 3 Haozheng Ren, 4 Ming Li 1 School of Computer Engineering, Huaihai Institute of Technology, Lianyungang, China, zhyhglyg@126.com

More information

Managing Object Versioning in Geo-Distributed Object Storage Systems

Managing Object Versioning in Geo-Distributed Object Storage Systems Managing Object Versioning in Geo-Distributed Object Storage Systems João Neto KTH joaon@kth.se Vianney Rancurel Vinh Tao Scality Scality and UPMC-LIP6 vianney.rancurel@scality.com vinh.tao@lip6.fr ABSTRACT

More information

Making Sense of NoSQL Dan McCreary, Kelly-McCreary & Associates. Minnesota Web Design Community Meetup Monday, February 3rd, :00pm to 8:30pm

Making Sense of NoSQL Dan McCreary, Kelly-McCreary & Associates. Minnesota Web Design Community Meetup Monday, February 3rd, :00pm to 8:30pm Making Sense of NoSQL Dan McCreary, Kelly-McCreary & Associates Minnesota Web Design Community Meetup Monday, February 3rd, 2014 6:00pm to 8:30pm Food for thought What percentage of database transactions

More information

SQL Query Optimization on Cross Nodes for Distributed System

SQL Query Optimization on Cross Nodes for Distributed System 2016 International Conference on Power, Energy Engineering and Management (PEEM 2016) ISBN: 978-1-60595-324-3 SQL Query Optimization on Cross Nodes for Distributed System Feng ZHAO 1, Qiao SUN 1, Yan-bin

More information

A Data Classification Algorithm of Internet of Things Based on Neural Network

A Data Classification Algorithm of Internet of Things Based on Neural Network A Data Classification Algorithm of Internet of Things Based on Neural Network https://doi.org/10.3991/ijoe.v13i09.7587 Zhenjun Li Hunan Radio and TV University, Hunan, China 278060389@qq.com Abstract To

More information

Analyzing and Improving Load Balancing Algorithm of MooseFS

Analyzing and Improving Load Balancing Algorithm of MooseFS , pp. 169-176 http://dx.doi.org/10.14257/ijgdc.2014.7.4.16 Analyzing and Improving Load Balancing Algorithm of MooseFS Zhang Baojun 1, Pan Ruifang 1 and Ye Fujun 2 1. New Media Institute, Zhejiang University

More information

AN EFFICIENT DEADLOCK DETECTION AND RESOLUTION ALGORITHM FOR GENERALIZED DEADLOCKS. Wei Lu, Chengkai Yu, Weiwei Xing, Xiaoping Che and Yong Yang

AN EFFICIENT DEADLOCK DETECTION AND RESOLUTION ALGORITHM FOR GENERALIZED DEADLOCKS. Wei Lu, Chengkai Yu, Weiwei Xing, Xiaoping Che and Yong Yang International Journal of Innovative Computing, Information and Control ICIC International c 2017 ISSN 1349-4198 Volume 13, Number 2, April 2017 pp. 703 710 AN EFFICIENT DEADLOCK DETECTION AND RESOLUTION

More information

A Static-Dynamic Conjunct Windows Process Integrity Detection Model

A Static-Dynamic Conjunct Windows Process Integrity Detection Model A Static-Dynamic Conjunct Windows Process Integrity Detection Model Fei Chen 1, Yi Li 1, Tong Zhang 1, Kehe Wu 1, 1 North China Electric Power University, Department of Control and Computer Engineering,

More information

A Proxy-based Query Aggregation Method for Distributed Key-Value Stores

A Proxy-based Query Aggregation Method for Distributed Key-Value Stores A Proxy-based Query Aggregation Method for Distributed Key-Value Stores Daichi Kawanami, Masanari Kamoshita, Ryota Kawashima and Hiroshi Matsuo Nagoya Institute of Technology, in Nagoya, Aichi, 466-8555,

More information

A Hybrid Scheme for Object Allocation in a Distributed Object-Storage System

A Hybrid Scheme for Object Allocation in a Distributed Object-Storage System A Hybrid Scheme for Object Allocation in a Distributed Object-Storage System Fang Wang **, Shunda Zhang, Dan Feng, Hong Jiang, Lingfang Zeng, and Song Lv Key Laboratory of Data Storage System, Ministry

More information

Load Balancing For Cloud-Based Dynamic Data Processing

Load Balancing For Cloud-Based Dynamic Data Processing Journal of Computer Sciences Investigations Load Balancing For Cloud-Based Dynamic Data Processing Talal Talib Jameel Department of Dentistry, Al Yarmouk University College, Baghdad, Iraq Article history

More information

Dynamo: Amazon s Highly- Available Key- Value Store

Dynamo: Amazon s Highly- Available Key- Value Store Dynamo: Amazon s Highly- Available Key- Value Store Giuseppe DeCandia, Deniz Hastorun, Madan Jampani, Gunavardhan KakulapaD, Avinash Lakshman, Alex Pilchin, Swaminathan Sivasubramanian, Peter Vosshall

More information

Scalability In Peer-to-Peer Systems. Presented by Stavros Nikolaou

Scalability In Peer-to-Peer Systems. Presented by Stavros Nikolaou Scalability In Peer-to-Peer Systems Presented by Stavros Nikolaou Background on Peer-to-Peer Systems Definition: Distributed systems/applications featuring: No centralized control, no hierarchical organization

More information

Video Inter-frame Forgery Identification Based on Optical Flow Consistency

Video Inter-frame Forgery Identification Based on Optical Flow Consistency Sensors & Transducers 24 by IFSA Publishing, S. L. http://www.sensorsportal.com Video Inter-frame Forgery Identification Based on Optical Flow Consistency Qi Wang, Zhaohong Li, Zhenzhen Zhang, Qinglong

More information

From Paxos to CORFU: A Flash-Speed Shared Log

From Paxos to CORFU: A Flash-Speed Shared Log From Paxos to CORFU: A Flash-Speed Shared Log Dahlia Malkhi dalia@microsoft.com Vijayan Prabhakaran vijayanp@microsoft.com Mahesh Balakrishnan maheshba@microsoft.com Ted Wobber wobber@microsoft.com John

More information

A method of three-dimensional subdivision of arbitrary polyhedron by. using pyramids

A method of three-dimensional subdivision of arbitrary polyhedron by. using pyramids 5th International Conference on Measurement, Instrumentation and Automation (ICMIA 2016) A method of three-dimensional subdivision of arbitrary polyhedron by using pyramids LIU Ji-bo1,a*, Wang Zhi-hong1,b,

More information

Distributed Systems. 17. Distributed Lookup. Paul Krzyzanowski. Rutgers University. Fall 2016

Distributed Systems. 17. Distributed Lookup. Paul Krzyzanowski. Rutgers University. Fall 2016 Distributed Systems 17. Distributed Lookup Paul Krzyzanowski Rutgers University Fall 2016 1 Distributed Lookup Look up (key, value) Cooperating set of nodes Ideally: No central coordinator Some nodes can

More information

On the Expansion of Access Bandwidth of Manufacturing Cloud Core Network

On the Expansion of Access Bandwidth of Manufacturing Cloud Core Network 1288 JOURNAL OF SOFTWARE, VOL. 9, NO. 5, MAY 2014 On the Expansion of Access Bandwidth of Manufacturing Cloud Core Network Hongyao Ju Zhejiang Textile & Fashion College, NingBo 315211, P.R.China Email:

More information

A Structured Overlay for Non-uniform Node Identifier Distribution Based on Flexible Routing Tables

A Structured Overlay for Non-uniform Node Identifier Distribution Based on Flexible Routing Tables A Structured Overlay for Non-uniform Node Identifier Distribution Based on Flexible Routing Tables Takehiro Miyao, Hiroya Nagao, Kazuyuki Shudo Tokyo Institute of Technology 2-12-1 Ookayama, Meguro-ku,

More information

Hash Tables Outline. Definition Hash functions Open hashing Closed hashing. Efficiency. collision resolution techniques. EECS 268 Programming II 1

Hash Tables Outline. Definition Hash functions Open hashing Closed hashing. Efficiency. collision resolution techniques. EECS 268 Programming II 1 Hash Tables Outline Definition Hash functions Open hashing Closed hashing collision resolution techniques Efficiency EECS 268 Programming II 1 Overview Implementation style for the Table ADT that is good

More information

A Novel Extreme Point Selection Algorithm in SIFT

A Novel Extreme Point Selection Algorithm in SIFT A Novel Extreme Point Selection Algorithm in SIFT Ding Zuchun School of Electronic and Communication, South China University of Technolog Guangzhou, China zucding@gmail.com Abstract. This paper proposes

More information

An Adaptive Threshold LBP Algorithm for Face Recognition

An Adaptive Threshold LBP Algorithm for Face Recognition An Adaptive Threshold LBP Algorithm for Face Recognition Xiaoping Jiang 1, Chuyu Guo 1,*, Hua Zhang 1, and Chenghua Li 1 1 College of Electronics and Information Engineering, Hubei Key Laboratory of Intelligent

More information

Two-Mode Data Distribution Scheme for Heterogeneous Storage in Data Centers

Two-Mode Data Distribution Scheme for Heterogeneous Storage in Data Centers Two-Mode Data Distribution Scheme for Heterogeneous Storage in Data Centers Wei Xie, Jiang Zhou, Mark Reyes, Jason Noble and Yong Chen Department of Computer Science, Texas Tech University, Lubbock, TX

More information

EFFICIENT ROUTING OF LOAD BALANCING IN GRID COMPUTING

EFFICIENT ROUTING OF LOAD BALANCING IN GRID COMPUTING EFFICIENT ROUTING OF LOAD BALANCING IN GRID COMPUTING MOHAMMAD H. NADIMI-SHAHRAKI*, FARAMARZ SAFI, ELNAZ SHAFIGH FARD Department of Computer Engineering, Najafabad branch, Islamic Azad University, Najafabad,

More information

Dynamo: Amazon s Highly Available Key-value Store

Dynamo: Amazon s Highly Available Key-value Store Dynamo: Amazon s Highly Available Key-value Store Giuseppe DeCandia, Deniz Hastorun, Madan Jampani, Gunavardhan Kakulapati, Avinash Lakshman, Alex Pilchin, Swaminathan Sivasubramanian, Peter Vosshall and

More information

Towards Optimal Data Replication Across Data Centers

Towards Optimal Data Replication Across Data Centers Towards Optimal Data Replication Across Data Centers Fan Ping, Xiaohu Li, Christopher McConnell, Rohini Vabbalareddy, Jeong-Hyon Hwang Department of Computer Science University at Albany - SUNY {apping,

More information

Research on Load Balancing in Task Allocation Process in Heterogeneous Hadoop Cluster

Research on Load Balancing in Task Allocation Process in Heterogeneous Hadoop Cluster 2017 2 nd International Conference on Artificial Intelligence and Engineering Applications (AIEA 2017) ISBN: 978-1-60595-485-1 Research on Load Balancing in Task Allocation Process in Heterogeneous Hadoop

More information

Experimental Study on Neighbor Selection Policy for Phoenix Network Coordinate System

Experimental Study on Neighbor Selection Policy for Phoenix Network Coordinate System Experimental Study on Neighbor Selection Policy for Phoenix Network Coordinate System Gang Wang, Shining Wu, Guodong Wang, Beixing Deng, Xing Li Tsinghua National Laboratory for Information Science and

More information

WITH the rapidly increasing usage of internet based services,

WITH the rapidly increasing usage of internet based services, PHD QUALIFYING EXAM, 2014; SCHOOL OF INFORMATICS AND COMPUTING, INDIANA UNIVERSITY 1 A Survey on SEDA and Using SEDA to Optimize HDFS Read Operation Isuru Suriarachchi, School of Informatics and Computing,

More information

A New Method Of VPN Based On LSP Technology

A New Method Of VPN Based On LSP Technology 2nd Joint International Information Technology, Mechanical and Electronic Engineering Conference (JIMEC 2017) A New Method Of VPN Based On LSP Technology HaiJun Qing 1, 2 1, 2, ChaoXiang Liang, LiPing

More information

A Test Sequence Generation Method Based on Dependencies and Slices Jin-peng MO *, Jun-yi LI and Jian-wen HUANG

A Test Sequence Generation Method Based on Dependencies and Slices Jin-peng MO *, Jun-yi LI and Jian-wen HUANG 2017 2nd International Conference on Advances in Management Engineering and Information Technology (AMEIT 2017) ISBN: 978-1-60595-457-8 A Test Sequence Generation Method Based on Dependencies and Slices

More information

A Two-layer Geo-cloud based Dynamic Replica Creation Strategy

A Two-layer Geo-cloud based Dynamic Replica Creation Strategy Appl. Math. Inf. Sci. 8, No. 1, 431-4 (2014) 431 Applied Mathematics & Information Sciences An International Journal http://dx.doi.org/10.12785/amis/080154 A Two-layer Geo-cloud based Dynamic Replica Creation

More information

Distributed Systems. 16. Distributed Lookup. Paul Krzyzanowski. Rutgers University. Fall 2017

Distributed Systems. 16. Distributed Lookup. Paul Krzyzanowski. Rutgers University. Fall 2017 Distributed Systems 16. Distributed Lookup Paul Krzyzanowski Rutgers University Fall 2017 1 Distributed Lookup Look up (key, value) Cooperating set of nodes Ideally: No central coordinator Some nodes can

More information

Study on Image Position Algorithm of the PCB Detection

Study on Image Position Algorithm of the PCB Detection odern Applied cience; Vol. 6, No. 8; 01 IN 1913-1844 E-IN 1913-185 Published by Canadian Center of cience and Education tudy on Image Position Algorithm of the PCB Detection Zhou Lv 1, Deng heng 1, Yan

More information

Enhanced Paxos Commit for Transactions on DHTs

Enhanced Paxos Commit for Transactions on DHTs Konrad-Zuse-Zentrum für Informationstechnik Berlin Takustraße 7 D-14195 Berlin-Dahlem Germany FLORIAN SCHINTKE, ALEXANDER REINEFELD, SEIF HARIDI, THORSTEN SCHÜTT Enhanced Paxos Commit for Transactions

More information

Lecture 16. Reading: Weiss Ch. 5 CSE 100, UCSD: LEC 16. Page 1 of 40

Lecture 16. Reading: Weiss Ch. 5 CSE 100, UCSD: LEC 16. Page 1 of 40 Lecture 16 Hashing Hash table and hash function design Hash functions for integers and strings Collision resolution strategies: linear probing, double hashing, random hashing, separate chaining Hash table

More information

The 3D Terrain Interactive Technique Based on Gesture Recognition Yanyan Li1, a, Xiaomeng Xu2, b, Jiayu Sun3, c, Haimeng Zhao4, d*

The 3D Terrain Interactive Technique Based on Gesture Recognition Yanyan Li1, a, Xiaomeng Xu2, b, Jiayu Sun3, c, Haimeng Zhao4, d* Joint International Information Technology, Mechanical and Electronic Engineering Conference (JIMEC 2016) The 3D Terrain Interactive Technique Based on Gesture Recognition Yanyan Li1, a, Xiaomeng Xu2,

More information

Jun Li, Ph.D. School of Computing and Information Sciences Phone:

Jun Li, Ph.D. School of Computing and Information Sciences Phone: Jun Li, Ph.D. School of Computing and Information Sciences Phone: + 1-305-348-4964 Florida International University Email: junli @ cs. fiu. edu 11200 SW 8th St, ECS 380, Miami, FL 33199 Web: http://users.cs.fiu.edu/

More information

TSAR : A Two Tier Sensor Storage Architecture using Interval Skip Graphs

TSAR : A Two Tier Sensor Storage Architecture using Interval Skip Graphs TSAR : A Two Tier Sensor Storage Architecture using Interval Skip Graphs Authors: Peter Desnoyers, Deepak Ganesan, Prashant Shenoy ( University of Massachusetts, Amherst, MA) Presented by: Nikhil Raghavan

More information

Enhanced Paxos Commit for Transactions on DHTs

Enhanced Paxos Commit for Transactions on DHTs Konrad-Zuse-Zentrum für Informationstechnik Berlin Takustraße 7 D-14195 Berlin-Dahlem Germany FLORIAN SCHINTKE, ALEXANDER REINEFELD, SEIF HARIDI, THORSTEN SCHÜTT Enhanced Paxos Commit for Transactions

More information

HASH TABLES. Hash Tables Page 1

HASH TABLES. Hash Tables Page 1 HASH TABLES TABLE OF CONTENTS 1. Introduction to Hashing 2. Java Implementation of Linear Probing 3. Maurer s Quadratic Probing 4. Double Hashing 5. Separate Chaining 6. Hash Functions 7. Alphanumeric

More information

Building a low-latency, proximity-aware DHT-based P2P network

Building a low-latency, proximity-aware DHT-based P2P network Building a low-latency, proximity-aware DHT-based P2P network Ngoc Ben DANG, Son Tung VU, Hoai Son NGUYEN Department of Computer network College of Technology, Vietnam National University, Hanoi 144 Xuan

More information

Research Article Apriori Association Rule Algorithms using VMware Environment

Research Article Apriori Association Rule Algorithms using VMware Environment Research Journal of Applied Sciences, Engineering and Technology 8(2): 16-166, 214 DOI:1.1926/rjaset.8.955 ISSN: 24-7459; e-issn: 24-7467 214 Maxwell Scientific Publication Corp. Submitted: January 2,

More information

Cassandra - A Decentralized Structured Storage System. Avinash Lakshman and Prashant Malik Facebook

Cassandra - A Decentralized Structured Storage System. Avinash Lakshman and Prashant Malik Facebook Cassandra - A Decentralized Structured Storage System Avinash Lakshman and Prashant Malik Facebook Agenda Outline Data Model System Architecture Implementation Experiments Outline Extension of Bigtable

More information

A Directed-multicast Routing Approach with Path Replication in Content Addressable Network

A Directed-multicast Routing Approach with Path Replication in Content Addressable Network 2010 Second International Conference on Communication Software and Networks A Directed-multicast Routing Approach with Path Replication in Content Addressable Network Wenbo Shen, Weizhe Zhang, Hongli Zhang,

More information

3D Digitization of Human Foot Based on Computer Stereo Vision Combined with KINECT Sensor Hai-Qing YANG a,*, Li HE b, Geng-Xin GUO c and Yong-Jun XU d

3D Digitization of Human Foot Based on Computer Stereo Vision Combined with KINECT Sensor Hai-Qing YANG a,*, Li HE b, Geng-Xin GUO c and Yong-Jun XU d 2017 International Conference on Mechanical Engineering and Control Automation (ICMECA 2017) ISBN: 978-1-60595-449-3 3D Digitization of Human Foot Based on Computer Stereo Vision Combined with KINECT Sensor

More information

Design of Liquid Level Control System Based on Simulink and PLC

Design of Liquid Level Control System Based on Simulink and PLC 2017 3rd International Symposium on Mechatronics and Industrial Informatics (ISMII 2017) ISBN: 978-1-60595-501-8 Design of Liquid Level Control System Based on Simulink and PLC Xiuwei FU 1,a, Ying QI 2,b

More information

A Novel Image Super-resolution Reconstruction Algorithm based on Modified Sparse Representation

A Novel Image Super-resolution Reconstruction Algorithm based on Modified Sparse Representation , pp.162-167 http://dx.doi.org/10.14257/astl.2016.138.33 A Novel Image Super-resolution Reconstruction Algorithm based on Modified Sparse Representation Liqiang Hu, Chaofeng He Shijiazhuang Tiedao University,

More information

Open Access Robust Design for Generalized Point Extract CNN with Application in Image Processing

Open Access Robust Design for Generalized Point Extract CNN with Application in Image Processing Send Orders for Reprints to reprints@benthamscience.ae The Open Automation and Control Systems Journal, 205, 7, 33-337 33 Open Access Robust Design for Generalized Point Extract CNN with Application in

More information

Energy Efficient in Cloud Computing

Energy Efficient in Cloud Computing Energy Efficient in Cloud Computing Christoph Aschberger Franziska Halbrainer May 24, 2013 1 of 25 Introduction Energy consumption by Google 2011: 2,675,898 MWh. We found that we use roughly as much electricity

More information

SCALABLE TRANSACTIONS USING MONGODB CLOUD DATA STORE

SCALABLE TRANSACTIONS USING MONGODB CLOUD DATA STORE SCALABLE TRANSACTIONS USING MONGODB CLOUD DATA STORE 1 LALITA CHAPLE, 2 DR. SWATI AHIRRAO 1,2 Computer Science Symbiosis Institute of Technology (SIT), Symbiosis International University (SIU), Lavale,

More information

BROMS: Best Ratio of MLC to SLC

BROMS: Best Ratio of MLC to SLC BROMS: Best Ratio of MLC to SLC Wei Wang 1, Tao Xie 2, Deng Zhou 1 1 Computational Science Research Center, San Diego State University 2 Computer Science Department, San Diego State University Partitioned

More information

Adaptive replica consistency policy for Kafka

Adaptive replica consistency policy for Kafka Adaptive replica consistency policy for Kafka Zonghuai Guo 1,2,*, Shiwang Ding 1,2 1 Chongqing University of Posts and Telecommunications, 400065, Nan'an District, Chongqing, P.R.China 2 Chongqing Mobile

More information

INTERNATIONAL JOURNAL OF PURE AND APPLIED RESEARCH IN ENGINEERING AND TECHNOLOGY

INTERNATIONAL JOURNAL OF PURE AND APPLIED RESEARCH IN ENGINEERING AND TECHNOLOGY INTERNATIONAL JOURNAL OF PURE AND APPLIED RESEARCH IN ENGINEERING AND TECHNOLOGY A PATH FOR HORIZING YOUR INNOVATIVE WORK EFFICIENT DATA PROCESSING IN PEER NETWORK USING CLOUD COMPUTING SHILPA VASANTRAO

More information

NoSQL and Database as a Service

NoSQL and Database as a Service NoSQL and Database as a Service Prof. Dr. Marcel Graf TSM-ClComp-EN Cloud Computing (C) 2017 HEIG-VD Databases Historical overview Prehistory : Hierarchical or network databases 1980 Ascent of relational

More information

Predicting Response Times for the Spotify Backend

Predicting Response Times for the Spotify Backend Predicting Response Times for the Spotify Backend Rerngvit Yanggratoke, Gunnar Kreitz, Mikael Goldmann,andRolfStadler ACCESS Linnaeus Center, KTH Royal Institute of Technology, Sweden Spotify, Sweden Email:

More information

Open Access The Three-dimensional Coding Based on the Cone for XML Under Weaving Multi-documents

Open Access The Three-dimensional Coding Based on the Cone for XML Under Weaving Multi-documents Send Orders for Reprints to reprints@benthamscience.ae 676 The Open Automation and Control Systems Journal, 2014, 6, 676-683 Open Access The Three-dimensional Coding Based on the Cone for XML Under Weaving

More information

A Distributed Approach to Fast Map Overlay

A Distributed Approach to Fast Map Overlay A Distributed Approach to Fast Map Overlay Peter Y. Wu Robert Morris University Abstract Map overlay is the core operation in many GIS applications. We briefly survey the different approaches, and describe

More information

HVM: A METHOD FOR IMPROVING THE PERFORMANCE OF EXECUTING SQL-QUERY OVER ENCRYPTED DATABASE

HVM: A METHOD FOR IMPROVING THE PERFORMANCE OF EXECUTING SQL-QUERY OVER ENCRYPTED DATABASE HVM: A METHOD FOR IMPROVING THE PERFORMANCE OF EXECUTING SQL-QUERY OVER ENCRYPTED DATABASE JA AFER AL-SARAIREH King Hussein School of Computing Sciences, Princess Sumaya University for Technology, Jordan

More information

Implementation of RS-485 Communication between PLC and PC of Distributed Control System Based on VB

Implementation of RS-485 Communication between PLC and PC of Distributed Control System Based on VB IOP Conference Series: Materials Science and Engineering PAPER OPEN ACCESS Implementation of RS-485 Communication between PLC and PC of Distributed Control System Based on VB To cite this article: Chuan

More information

x + 1 = 2 Write x as a fraction in the form a, where a and b are positive integers with no common factor other than 1. Pass on the value of b a.

x + 1 = 2 Write x as a fraction in the form a, where a and b are positive integers with no common factor other than 1. Pass on the value of b a. A1 Solve for x the equation x + 1 = 2 x. Write x as a fraction in the form a, where a and b are positive b integers with no common factor other than 1. Pass on the value of b a. A3 ABCD is a square that

More information