Methodology to Determine Relationships between Performance Factors in Hadoop Cloud Computing Applications

Save this PDF as:
 WORD  PNG  TXT  JPG

Size: px
Start display at page:

Download "Methodology to Determine Relationships between Performance Factors in Hadoop Cloud Computing Applications"

Transcription

1 Methodology to Determne Relatonshps between Performance Factors n Hadoop Cloud Computng Applcatons Lus Eduardo Bautsta Vllalpando 1,2, Alan Aprl 1 and Alan Abran 1 1 Department of Software Engneerng and Informaton Technology - ETS, Unversty of Quebec, Montreal, Canada 2 Department of Electronc Systems, Autonomous Unversty of Aguascalentes, Aguascalentes, Mexco {alan.aprl, Keywords: Abstract: Cloud Computng, Measurement, Performance, Taguch Method, ISO 25010, Mantenance, Hadoop Mapreduce. Cloud Computng s a model for enablng ubqutous, convenent, on-demand network access to a shared pool of confgurable computng resources. Cloud Computng users prefer not to own physcal nfrastructure, but nstead rent Cloud nfrastructure, a Cloud platform or software, from a thrd-party provder. Sometmes, anomales and defects affect a part of the Cloud platform, resultng n degradaton of the Cloud performance. One of the challenges n dentfyng the source of such degradaton s how to determne the type of relatonshp that exsts between the varous performance metrcs whch affect the qualty of the Cloud and more specfcally Cloud applcatons. Ths work uses the Taguch method for the desgn of experments to propose a methodology for dentfyng the relatonshps between the varous confguraton parameters that affect the qualty of Cloud Computng performance n Hadoop envronments. Ths paper s based on a proposed performance measurement framework for Cloud Computng systems, whch ntegrates software qualty concepts from ISO and other nternatonal standards. 1 INTRODUCTION Cloud Computng (CC) s a model for enablng ubqutous, convenent, on-demand network access to a shared pool of confgurable computng resources (e.g., networks, servers, storage, applcatons, and servces) that can be rapdly provsoned and released wth mnmal management effort or servce provder nteracton (Mell and Grance 2011). Some CC users prefer not to own physcal nfrastructure, but nstead rent Cloud nfrastructure, or a Cloud platform or software, from a thrd-party provder. These nfrastructure applcaton optons delvered as a servce are known as Cloud Servces. Servce models for CC are categorzed as: Infrastructure as a Servce (IaaS), Platform as a Servce (PaaS), and Software as a Servce (SaaS) (ISO/IEC 2011). The model that relates the most to the software engneerng communty s the SaaS model. Software engneers focus on software components, and customers use an IT provder s applcatons runnng on a Cloud nfrastructure to process nformaton accordng to ther processng and storage requrements. One of the man characterstcs of ths type of servce s that customers do not manage or control the underlyng Cloud nfrastructure (ncludng network, servers, operatng systems, and storage), except for lmted user-specfc applcaton confguraton settngs. Performance measurement models (PMMo) for CC, and more specfcally for Cloud Computng Applcatons (CCA), should propose a means to dentfy and quantfy "normal applcaton behavour," whch can serve as a baselne for detectng and predctng possble anomales n the software (.e. applcatons n a Cloud envronment) that may mpact n a Cloud applcaton. To be able to desgn such PMMo for CCA, methods are needed to collect the necessary base measures specfc to performance, and analyss models must be desgned to analyze and evaluate the relatonshps that exst among these measures. One of the challenges n desgnng PMMo for CCA s how to determne what type of relatonshp exsts between the varous base measures. For example, what s the extent of the relatonshp between the amount of physcal memory used and the amount of nformaton to process by an applcaton? Thus, ths work proposes the use of a 375

2 CLOSER2014-4thInternatonalConferenceonCloudComputngandServcesScence methodology based on the Taguch method to determne how closely the performance parameters (base measures) nvolved n the performance analyss process are related. The Taguch method combnes ndustral and statstcal experence, and offers a means for mprovng the qualty of manufactured products. It s based on the robust desgn concept, popularzed by Taguch, accordng to whch a well desgned product should cause no problems when used under specfed condtons (Taguch, Chowdhury et al. 2005). Although the experment presented n ths paper was not developed n a CC producton system, the man contrbuton of ths work s to propose the Taguch method as a way to determne relatonshps between performance parameters of CCA. Ths paper s structured as follows. Secton 2 presents background of concepts related to the performance measurement of CCA and ntroduces the MapReduce programmng model, whch s used to develop CCA. In addton, secton 2 presents the PMFCC, whch descrbes the key performance concepts and sub concepts dentfed from nternatonal standards. Secton 3 presents the method for examnng the relatonshps among the performance concepts dentfed n the PMFCC. In ths secton, an expermental methodology based on the Taguch method of expermental desgn, s used and offers a means for mprovng the qualty of product performance. Secton 4 presents the results of the experment and analyzes the relatonshp between the performance factors of CCA. Fnally, secton 5 presents a synthess of the results of ths research and suggests future work. 2 BACKGROUND 2.1 Performance Analyss n Cloud Computng Applcatons Researchers have studed the performance of CCA from varous vewponts. For example, Jackson (Jackson et al., 2010) analyzes hgh performance computng applcatons on the Amazon Web Servces cloud, wth the objectve of examnng the performance of exstng CC nfrastructures and creatng a mechansm to quanttatvely evaluate them. Hs work s focused on the performance of Amazon EC2 as a representatve example of the current manstream of commercal CC servces, and ts potental applcablty to Cloud-based envronments n scentfc computng envronments. He quanttatvely examnes the performance of a set of benchmarks desgned to represent a typcal Hgh Performance Computng (HPC) workload runnng on the Amazon EC2 platform. Tmng results from dfferent applcaton benchmarks are used to compute a Sustaned System Performance (SSP) metrc, whch s a derved measure for measurng the performance delvered by the workload of a computng system. Accordng to the Natonal Energy Research Scentfc Computng Center (NERSC) (Kramer et al., 2005), SSP s useful for evaluatng system performance across any tme frame, and can be appled to any set of systems, any workload, and/or benchmark sute, and for any tme perod. In addton, SSP measures tme to soluton across dfferent applcaton areas, and can be used to evaluate absolute performance and performance relatve to cost (n dollars, energy, or other value propostons). In hs work, Jackson shows that the SSP metrc has a strong correlaton between the percentage of tme an applcaton spends communcatng and ts overall performance on EC2. Also hghlghted, the more communcaton there s, the worse the performance became. Jackson concludes that the communcaton pattern of an applcaton can have a sgnfcant mpact on performance. Other researchers focus on applcatons n vrtualzed Cloud envronments. For nstance, Me (Me et al., 2010) studes the measurement and analyss of the performance of network I/O applcatons (network-ntensve applcatons) n these envronments. The am of hs research s to understand the performance mpact of co-locatng applcatons n a vrtualzed Cloud, n terms of throughput performance and resource sharng effectveness. Me addresses ssues related to managng dle nstances, whch are processes runnng n an operatng system (OS) that are executng dle loops. Results show that when two dentcal I/O applcatons are runnng together, schedulers can approxmately guarantee that each has ts far share of CPU slcng, network bandwdth consumpton, and resultng throughput. It also shows that the duraton of performance degradaton experenced s related to machne capacty, workload level n the runnng doman, and the number of new vrtual machne (VM) nstances to start up. Although these publcatons present nterestng methods for performance measurement of CCA, the approaches used were from an nfrastructure perspectve and dd not consder CCA performance factors from a software engneerng perspectve. Ths work bases the performance evaluaton of CCA on frameworks developed for data ntensve 376

3 MethodologytoDetermneRelatonshpsbetweenPerformanceFactorsnHadoopCloudComputngApplcatons processng.e. lke Hadoop and MapReduce, and by ntegratng software qualty concepts from ISO 25010, as well as frameworks for Cloud Computng Systems (CCS) performance measurement. Ths approach was taken as a novel way to apply concepts of software engneerng to the new paradgm of cloud computng. 2.2 The ISO 5939 Measurement Process Model The purpose of a measurement process, as descrbed n ISO (ISO/IEC 2008), s to collect, analyze, and report data relatng to the products developed and processes mplemented n an organzatonal unt, both to support effectve management of the process and to objectvely demonstrate the qualty of the products. ISO defnes four sequental actvtes n a measurement process: establsh and sustan measurement commtment, plan the measurement process, perform the measurement process, and evaluate the measurement. These actvtes are performed n an teratve cycle that allows for contnuous feedback and mprovement of the measurement process, as shown n Fgure 1. The frst two actvtes recommended by the ISO measurement process, whch are: 1) establsh measurement commtment; and 2) plan the measurement process, were addressed n the work, "Desgn of a Performance Measurement Framework for Cloud Computng (PMFCC) (Bautsta et al., 2012). In ths paper, the bases for the measurement of Cloud Computng concepts that are drectly related to performance are defned. The PMFCC dentfes terms assocated wth the qualty concept of performance, whch have been dentfed from nternatonal standards such as ISO and those of the European Cooperaton on Space Standardzaton. The PMFCC proposes a combnaton of base measures to determne the derved measures of a specfc concept that contrbutes to performance analyss. 2.3 Performance Measurement Framework for Cloud Computng Jan s System Performance Concepts and Sub Concepts A well known perspectve for system performance measurement was proposed by Jan (Jan, 1991), who suggests that a performance study must frst defne a set of performance crtera (or characterstcs) to help carryng out the system measurement process. He notes that system performance s typcally measured usng three sub concepts, f t s performng a servce correctly: 1) responsveness, 2) productvty, and 3) utlzaton, and proposes a measurement process for each. In addton, Jan notes that there are several possble outcomes for each servce request made to a system, whch can be classfed n three categores. The system may: 1) perform the servce correctly, 2) perform the servce ncorrectly, or 3) refuse to perform the servce altogether. Moreover, he defnes three sub concepts assocated wth each of these possble outcomes whch affect system performance: 1) speed, 2) relablty, and 3) avalablty. Fgure 2 presents the possble outcomes of a servce request to a system and the sub concepts assocated wth them. Fgure 1: Sequence of actvtes n a measurement process (adapted from the ISO 5939 measurement process model (ISO/IEC 2008)). 377

4 CLOSER2014-4thInternatonalConferenceonCloudComputngandServcesScence Fgure 2: Possble outcomes of a servce request to a system, accordng to Jan. by two man sub concepts: 1) performance effcency, and 2) relablty. We have observed that when a CCS receves a servce request, there are three possble outcomes (the servce s performed correctly, the servce s performed ncorrectly, or the servce cannot be performed). The outcome wll determne the sub concepts that wll be used for performance measurement. For example, suppose that the CCS performs a servce correctly, but, durng executon, the servce faled and was later renstated. Although the servce was ultmately performed successfully, t s clear that the system avalablty (part of the relablty sub concept) was compromsed, and ths affected CCS performance Defnton of Cloud Computng Applcaton Performance The ISO (ISO/IEC 2011) standard defnes software product and computer system qualty from two dstnct perspectves: 1) a qualty n use model, and 2) a product qualty model. The product qualty model s applcable to both systems and software. Accordng to ISO 25010, the propertes of both determne the qualty of the product n a partcular context, based on user requrements. Based on Jan s performance perspectves and the man ISO product qualty characterstcs, we propose the followng defnton of CCA performance measurement: The performance of a Cloud Computng applcaton s determned by analyss of the characterstcs nvolved n performng an effcent and relable servce that meets requrements under stated condtons and wthn the maxmum lmts of the system parameters. Although at frst sght ths defnton may seem complex, t only ncludes the sub concepts necessary to carry out CCA performance analyss Defnton of the Performance Measurement Framework for Cloud Computng Performance measurement concepts and sub concepts have prevously been related usng a proposed relatonshp model whch was descrbed n detal n the PMFCC (Bautsta et al., 2012) (see n Fgure 3). Ths model presents the logcal sequence, from top to bottom, n whch the concepts and sub concepts appear when a performance ssue arses n a Cloud Computng System (CCS). In Fgure 3, system performance s determned Fgure 3: Model of the relatonshps between performance concepts and sub concepts. Thus, PMFCC defnes the base measures related to the performance concepts that represent the system attrbutes, and whch can be measured to assess whether or not the CCA satsfes the stated requrements. These base measures are grouped nto collecton functons, whch are responsble for conductng the measurement process usng a combnaton of base measures through a data collector. They are assocated wth the correspondng ISO qualty derved measures, as presented n Table 1. An example of usng the framework s: how can be measured the CC avalablty concept (presented n Table 1) usng the PMFCC? As a frst step, t needs three collecton functons: 1) the tme 378

5 MethodologytoDetermneRelatonshpsbetweenPerformanceFactorsnHadoopCloudComputngApplcatons functon, 2) the task functon, and 3) the transmsson functon. The tme functon can use several dfferent measurements, such as CPU utlzaton by the user, job duraton, and response tme. These base measures can be obtaned usng a data collector, and then send the measures to a tme functon that calculates a derved measure of the tme concept. An ntermedate servce wll be desgned to combne the results of each functon n order to calculate a derved measure of the avalablty that contrbutes to CC performance, as defned n the framework. Table 1: Functons assocated wth Cloud Computng performance concepts. Base Measures Falures avoded Falures detected Falures predcted Falures resolved Breakdowns Faults corrected Faults detected Faults predcted Tasks entered nto recovery Tasks executed Tasks passed Tasks restarted Tasks restored Tasks successfully restored Contnuous resource utlzaton tme Down tme Maxmum response tme Observaton tme Operaton tme Recovery tme Repar tme Response tme Task tme Tme I/O devces occuped Transmsson response tme Turnaround tme Transmsson errors Transmsson capacty Transmsson rato Collecton Functons for Measures Falure functon Fault functon Task functon Tme functon Transmsson functon Hadoop Mapreduce ISO Derved Measures Maturty Resource utlzaton Fault tolerance Maturty Fault tolerance Avalablty Capacty Maturty Fault tolerance Resource utlzaton Tme behavour Avalablty Capacty Maturty Recoverablty Resource utlzaton Tme behavour Avalablty Capacty Maturty Recoverablty Resource utlzaton Tme behavour Hadoop s an Apache Software Foundaton s project, and encompasses varous Hadoop subprojects. The Hadoop project develops and supports the open source software that supples a framework for the development of hghly scalable dstrbuted computng applcatons desgned to handle processng detals, leavng developers free to focus on applcaton logc (Hadoop, 2012). MapReduce s a Hadoop subproject whch s a programmng model wth an assocated mplementaton for processng and generatng large datasets. Accordng to Dean (Dean and Ghemawat, 2008), programs wrtten n ths functonal style are automatcally parallelzed and executed on a large cluster of commodty machnes. Authors lke Ln (Ln and Dyer, 2010) pont out that today s ssue, whch s the need to tackle large amounts of data, s addressed by a dvde-and-conquer approach, where the basc dea s to partton a large problem nto smaller sub problems. Those sub problems can be processed n parallel by dfferent workers; for example, threads n a processor core, cores n a mult-core processor, multple processors n a machne, or many machnes n a cluster. The ntermedate results of each ndvdual worker are combned to yeld the fnal output. 3 METHODOLOGY 3.1 Defnton of the Problem To desgn the proposed collecton functons proposed n the PMFCC (see n Table 1), t s needed to determne how the varous base measures are related and to what degree. Studyng these relatonshps enables assess the nfluence each of them has on the resultng derved measures. The PMFCC shows many of the relatonshps that exst between the base measures that have a major nfluence on the collecton functons. In CCA, and more specfcally n the MapReduce applcatons, there are over a hundred base measures (ncludng system measures) whch could potentally contrbute to the analyss of CCA performance. A selecton of these measures has to be ncluded n the collecton functons so that the respectve measures can be derved, and from there an ndcaton of the performance of the applcatons can be obtaned. There are two key desgn problems to be solved here: 1) establsh whch base measures are nterrelated, and 2) determne how much the nterrelated measures contrbute to each of the collecton functons. In tradtonal statstcal methods, thrty or more observatons (or data ponts) are typcally needed for 379

6 CLOSER2014-4thInternatonalConferenceonCloudComputngandServcesScence each varable observed, n order to gan meanngful nsght. In addton, a few ndependent varables are needed for the experments desgned to uncover potental relatonshps among them. These experments must be performed under certan predetermned and controlled test condtons. However, ths approach s not approprate here, owng to the large number of varables nvolved and the tme and effort that would be requred, whch s much more than we have allowed for n ths step of the research. Consequently, we have to resort to an analyss method that s better suted to our constrants, specfc problem and study area. A possble canddate approach s Taguch s expermental desgn method, whch nvestgates how dfferent varables affect the mean and varance of a process performance characterstc helpng n determnng how well the process s functonng. Ths method only requres a lmted number of experments, but s more effcent than a factoral desgn n ts ablty to dentfy relatonshps and dependences. The next secton descrbes the method and the concepts to be used. 3.2 Taguch s Method of Expermental Desgn Taguch's Qualty Engneerng Handbook (Taguch et al., 2005) descrbes the Taguch method of expermental desgn, whch was developed by Dr. Gench Taguch, a researcher at the Electronc Control Laboratory n Japan. Ths method combnes ndustral and statstcal experence, and offers a means for mprovng the qualty of manufactured products. It s based on the robust desgn concept, popularzed by Taguch, accordng to whch a well desgned product should cause no problems when used under specfed condtons. Accordng to Chekh (Chekh and Abran 2012), Taguch s two phase qualty strategy s the followng: Phase 1: The onlne phase, whch focuses on the technques and methods used to control qualty durng the producton of the product. Phase 2: The offlne phase, whch focuses on takng those technques and methods nto account before manufacturng the product, that s, durng the desgn phase, the development phase, etc. One of the most mportant actvtes n the offlne phase of the strategy s parameter desgn. Ths s where the parameters are determned that make t possble to satsfy the set qualty objectves (often called the objectve functon) through the use of expermental desgns under set condtons. If the product does not work properly (does not fulfl the objectve functon), then the desgn constants (also called parameters) need to be adjusted so that t wll perform better. Chekh explans that ths actvty ncludes several steps, whch are requred to determne the parameters that satsfy the qualty objectves (output). Accordng to Taguch's Qualty Engneerng Handbook, orthogonal arrays (OA) organzes the parameters affectng the process and the levels at whch they should vary. The OA show the varous experments that wll need to be conducted n order to verfy the effect of the factors studed on the output. Taguch s method tests pars of combnatons, nstead of havng to test all possble combnatons (as n a factoral expermental desgn). Wth ths approach, we can determne whch factors affect product qualty the most n a mnmum number of experments. Taguch s OA arrays can be created manually or they can be derved from determnstc algorthms. They are selected by the number of parameters (varables) and the number of levels (states). An OA array s represented by Ln and Pn, where Ln corresponds to the number of experments to be conducted, and Pn corresponds to the number of parameters to be analyzed. Table 2 presents an example of Taguch s OA L4, meanng that 4 experments are conducted to analyze 3 parameters. Table 2: Taguch s Orthogonal Array L4. No. of Experments (L) P1 P2 P An OA cell contans the factor levels (1 and 2) that determne the types of parameter values for each experment. Once the expermental desgn has been determned and the trals have been carred out, the performance characterstc measurements from each tral can be used to analyze the relatve effect of the varous parameters. Taguch s method s based on the use of the sgnal-to-nose rato (SNR), whch s a measurement scale that has been used n the communcatons ndustry for nearly a century for determnng the extent of the relatonshp between the qualty factors n a measurement model. The SNR approach nvolves the analyss of data for varablty, n whch an nput-to-output relatonshp s studed n the 380

7 MethodologytoDetermneRelatonshpsbetweenPerformanceFactorsnHadoopCloudComputngApplcatons measurement system. To determne the effect each parameter has on the output, the SNR (or SN number) s calculated by the formula 1. In ths formula y s the mean value and s s the varance (y s the value of the performance characterstc for a gven experment). where s y SN 10log (1) s 2 1 y N 1 N 1 N u1 2 2 y, u N y, u y u1 =Experment number u=tral number N=Number of trals for experment To mnmze the performance characterstc (objectve functon), the followng defnton of the SNR should be calculated: SN N 2 y u 10 log (2) u 1 N To maxmze the performance characterstc (objectve functon), the followng defnton of the SNR should be calculated: SN N log 2 (3) N u 1 yu Once the SNR values have been calculated for each factor and level, they are tabulated as shown n Table 3, and then the range R (R = hgh SN - low SN) of the SNR for each parameter also s calculated and entered nto Table 3. Table 3: Rank for SNR values. Level P1 P2 P3 1 SN 1,1 SN 2,1 SN 3,1 2 SN 1,2 SN 2,2 SN 3,2 3 SN 1,3 SN 2,3 SN 3,3 Range R P1 R P2 R P3 Rank Accordng to Taguch s method, the larger the R value for a parameter, the greater ts effect on the process. 3.3 Experment Expermental Setup All the experments were conducted on a DELL Studo Workstaton XPS 9100 wth an Intel Core 7 12-core X980 processor runnng at 3.3 GHz, 24 GB DDR3 RAM, a Seagate 1.5 TB 7200 RPM SATA 3Gb/s dsk, and a 1 Gbps network connecton. We used a Lnux CentOS bt dstrbuton and Xen 3.3 as the hypervsor. Ths physcal machne hosts fve vrtual machnes (VM), each wth a dual-core Intel 7 confguraton, 4 GB RAM, 10 GB vrtual storage, and a vrtual network nterface type. In addton, each VM executes the Apache Hadoop dstrbuton verson , whch ncludes the Hadoop Dstrbuted Fle System (HDFS) and MapReduce framework lbrares. One of these VM s the master node, whch executes NameNode (HDFS) and JobTracker (MapReduce), and the rest of the VM are slave nodes runnng DataNodes (HDFS) and JobTrackers (MapReduce). The Apache Hadoop dstrbuton ncludes a set of applcatons for testng the performance of a cluster. Accordng to Hadoop (Hadoop 2012), these applcatons can test varous cluster characterstcs, such as network transfer, storage relablty, cluster avalablty, etc. Four applcatons were selected to obtan performance measures from the Hadoop cluster as for example; the amount of physcal memory used by a Job s a measure that vares accordng to values gven to confguraton parameters, such as the number of fles to process, the amount of nformaton to process, etc. The vewpont taken for the selecton of the above applcatons s that t s possble to use the same type s o parameters to confgure each applcaton as well as cluster machne. Below s a bref descrpton of the applcatons used n the experments: 1. TestDFSIO. Ths s a MapReduce applcaton that reads and wrtes the HDFS test. It executes tasks to test the HDFS to dscover performance bottlenecks n the network, to test the hardware, the OS, and the Hadoop setup of the cluster machnes (partcularly the NameNode and the DataNodes), and to determne the speed of the cluster n terms of I/O. 2. TeraSort. The goal of ths applcaton s to sort large amounts of data as fast as possble. It s a benchmark applcaton that combnes HDFS testng, as well as testng of the MapReduce layers of a Hadoop cluster. 3. MapRedRelablty. Ths program tests the 381

8 CLOSER2014-4thInternatonalConferenceonCloudComputngandServcesScence relablty of the MapReduce framework by njectng faults/falures nto the Map and Reduce stages. 4. MapRedTest. Ths applcaton loops a small job a number of tmes, placng the focus on the MapReduce layer and ts mpact on the HDFS layer. To develop the set of experments, three parameters were selected, whch can be set wth dfferent values for each type of applcaton. These parameters are: 1) the number of fles to process, 2) the total number of bytes to process, and 3) the number of tasks to execute n the cluster. Also, a number of dfferent MapReduce base measures such as Job Duraton, Job Status, Amount of Amount of physcal memory used, etc. were selected as possble qualty objectves (objectve functon). These base measures are related to one or more of the performance derved measures dentfed n the PMFCC Defnton of Factors and Qualty Objectve In a vrtualzed Cloud envronment, Cloud provders mplement clusterng by slcng each physcal machne nto multple vrtual machnes (VM) nterconnected through vrtual nterfaces. So, we establshed a vrtual cluster wth the features mentoned above, n order to obtan representatve results. Ffty experments were performed to test the Hadoop vrtual cluster, varyng the three parameters mentoned prevously. In each experment, four dfferent applcatons were executed, and performance results were recorded for ther analyss. In ths way, the set of experments nvestgates the effect of the followng varables (or control factors, accordng to the Taguch termnology) on the output dependent varable: Number of fles to be processed by the cluster Total number of bytes to be processed by the cluster Number of tasks nto whch to dvde the Job applcaton Accordng to Taguch, qualty s often referred to as conformance to the operatng specfcatons of a system. To hm, the qualty objectve (or dependent varable) determnes the deal functon of the output that the system should show. In our experment, the observed dependent varable s the followng: Amount of physcal memory used by the Job (Mbytes) Experment Development Accordng to the Hadoop documentaton, the number of fles and the amount of data to be processed by a Hadoop cluster wll be determned by the number of processors (cores) avalable and ther storage capacty. Also, the number of tasks to be processed by the cluster wll be determned by the total of number of processng unts (cores) n the cluster. Based on the above premses and the confguraton of the expermental cluster, we have chosen two levels for each parameter n the experment. We determne the dfferent levels of each factor n the followng way: Number of fles to process: o Small set of fles: fewer than 10,000 fles for level 1; o Large set of fles: 10,000 fles or more for level 2. Number of bytes to process, as determned by the storage capacty of the cluster: o fewer than 10,000 Mbytes to process for level 1 (a small amount of data to process); o 10,000 or more Mbytes to process for level 2 (a large amount of data to process). Number of tasks to create, determned, accordng to the MapReduce framework, by the number of processng unts (cores) n the cluster and the number of nput fles to process. Snce the cluster has a total of 10 cores, we decded to perform tests wth: o fewer than 10 tasks for level 1; o 10 or more tasks for level 2. Table 4 Presents a summary of the factors, levels, and values for ths experment. Factor Number Table 4: Factors and Levels. Factor Name Level 1 Level 2 Number of fles to process Number of MB to process Number of tasks to create < 10,000 10,000 < 10,000 10,000 < Usng Taguch s expermental desgn method, the selecton of the approprate OA s determned by the number of factors and levels to be examned. The resultng OA array for ths case study s L4 (presented n Table 2). The assgnment of the varous factors and values of ths OA array s shown n Table

9 MethodologytoDetermneRelatonshpsbetweenPerformanceFactorsnHadoopCloudComputngApplcatons No. of the Experment (L) Table 5: Matrx of Experments. Number of Fles Number of Bytes (MB) Number of Tasks 1 < 10,000 < 10,000 < 10 2 < 10,000 10, ,000 < 10, ,000 10,000 < 10 Table 5 shows the set of experments to be carred out wth dfferent values for each parameter selected. For example, experment 2 nvolves fewer than 10,000 fles, the number of bytes to be processed s greater than or equal to 10,000 Mbytes, and the number of tasks s greater than or equal to 10. A total of 50 experments were carred out by varyng the parameter values. However, only 12 experments met the requrements presented n Table 5. Ths set of 12 experments was dvded nto three groups of four experments each (called trals). The values and results of each experment are presented n Table 6. Taguch s method defned the SNR used to measure robustness, whch s the transformed form of the performance qualty characterstc (output value) used to analyze the results. Snce the objectve of ths experment s to mnmze the qualty characterstc of the output (amount of physcal memory used per Job), the SNR for the qualty characterstc the smaller the better s gven by formula 4, that s: SN log N u 1 2 y u N 10 (4) The SNR result for each experment s shown n Table 7. Accordng to Taguch s method, the factor effect s equal to the dfference between the hghest average SNR and the lowest average SNR for each factor. Ths means that the larger the factor effect for a parameter, the larger the effect the varable has on the process, or, n other words, the more sgnfcant the effect of the factor. Table 8 shows the factor effect for each varable studed n the experment. Table 6: Trals, experments, and resultng values. Tral Experment Number of Fles Mbytes to Process Num. of Tasks Physcal Memory (Mbytes) , , ,000 10, , , ,000,000 10,000, ,000 10,000 1, ,000,000 3, ,000,000 50, Table 7: SNR results. Experment Number of Mbytes to Number of Physcal Memory Physcal Memory Physcal Memory Fles Process Tasks Tral 1 Tral 2 Tral 3 SNR 1 < 10,000 < 10,000 < < 10,000 10, ,000 < 10, ,000 10,000 < Table 8: Factor Effect on the Output Objectve Number of Mbytes to Fles Process Number of Tasks Average SNR at Level Average SNR at Level Factor Effect (dfference) Rank

10 CLOSER2014-4thInternatonalConferenceonCloudComputngandServcesScence 4 RESULTS 4.1 Analyss and Interpretaton of Results Based on the results presented n Table 8, we can observe that: Number of tasks s the factor that has the most nfluence on the qualty objectve (physcal memory used) of the output observed, at Number of Mbytes to process s the second most nfluental factor, at Number of fles to process s the least nfluental factor n ths case study, at Fgure 4 presents a graphcal representaton of the factor results and ther levels. 0,6 0,4 0,2 0 0,2 0,4 0,6 L1 L2 Fgure 4: Graphcal representaton of factors and ther SNR levels. To represent the optmal condton of the levels, also called the optmal soluton of the levels, an analyss of SNR values s necessary n ths experment. Whether the am s to mnmze or maxmze the qualty characterstc (physcal memory used), t s always necessary to maxmze the SNR parameter values. Consequently, the optmum level of a specfc factor wll be the hghest value of ts SNR. It can be seen that the optmum L1 L2 Number of Fles Number of tasks Mbytes to process L1 L2 level for each factor s represented by the hghest pont n the graph (as presented n Fgure 4); that s, L1, L1, and L2 respectvely. Usng the fndngs presented n Tables 7 and 8 and n Fgure 4, we can conclude that the optmum levels for the factors n ths experment based on the expermental confguraton cluster are: Number of fles to process: The optmum level s fewer than 10,000 fles (level 1). Total number of Mbytes to process: The optmum level s fewer than 10,000 Mbytes (level 1). Number of tasks to be created to dvde the Job: The optmum level s greater than or equal to 10 tasks or more per Job (level 2). 4.2 Statstcal Data Analyss The analyss of varance (ANOVA) s a statstcal technque usually used n the desgn and analyss of experments. Accordng to Trved (Trved, 2002), the purpose of applyng the ANOVA technque to an expermental stuaton s to compare the effect of several factors appled smultaneously to the response varable (qualty characterstc). It allows the effects of the controllable factors to be separated from those of uncontrolled varatons. Table 9 presents the results of ths analyss of the expermental factors. As can be seen n the contrbuton column of Table 9, these results can be nterpreted as follows (represented graphcally n Fgure 5): Number of tasks s the factor that has the most nfluence (43% of the contrbuton) on the physcal memory n ths case study. Total number of bytes to process s the factor that has the second greatest nfluence (34% of the contrbuton) on the processng tme. Number of fles s the factor wth the least nfluence (23% of the contrbuton) on the processng tme n the cluster. Table 9: Analyss of Varance (ANOVA). Factors Degrees of Sum of Squares Varance Contrbuton Varance Freedom (SS) (MS) (%) raton (F) No. of fles Total no. of bytes to process No. of tasks Error Total Error estmate

11 MethodologytoDetermneRelatonshpsbetweenPerformanceFactorsnHadoopCloudComputngApplcatons Fgure 5: Percentage contrbuton of factors. In addton, based on the column related to the varance rato F shown n Table 9, we can conclude that the followng: The factors Number of tasks and Number of Mbytes to process have the most domnant effect and the second most domnant effect on the output varable respectvely. Accordng to Taguch s method, the factor wth the smallest contrbuton s taken as the error estmate. So, the factor Total number of fles to process s taken as the error estmate, snce t corresponds to the smallest sum of squares. The results of ths case study show, based on both the graphcal and statstcal data analyses of the SNR, that the number of tasks nto whch to dvde the Job n a MapReduce applcaton n our cluster has the most nfluence, followed by the number of bytes to process, and, fnally, the number of fles. To summarze, when an applcaton s developed n the MapReduce framework to be executed n ths cluster, the factors mentoned above must be taken nto account n order to mprove the performance of the applcaton, and, more specfcally, the output varable, whch s the amount of physcal memory to be used by a Job. 5 SUMMARY One of the challenges n CC s to delver good performance to ts end users. In ths paper, we present the results of usng a method that determnes the relatonshps among the CCA performance parameters. Ths proposed method s based on a performance measurement framework for Cloud Computng (PMFCC) system, whch defnes a number of terms that are necessary to measure the performance of CCS usng software qualty concepts. The PMFCC defned several collecton functons whch are automated to obtan derved measures and enable analyss of the performance of a CCA. One of the challenges we faced n desgnng these functons was to decde how to determne the extent to whch the base measures are related, and ther nfluence n the analyss of CCA performance. To address ths challenge, we proposed the use of Taguch s method of expermental desgn. Usng ths expermental desgn method, we carred out experments to analyze the relatonshps between the confguraton parameters of several Hadoop applcatons and ther performance qualty measures based on the amount of physcal memory used by a Job. We found that there s a strong relatonshp between the number of tasks executed by a MapReduce applcaton and the amount of physcal memory used by a Job. Our next research actvty wll be to reproduce ths experment n a producton envronment, n order to confrm these tral group results wth greater certanty. Also, ths early research work serves as a bass for a next actvty that wll need to determne the most mportant relatonshps between the performance concepts defned n the PMFCC and enable us to propose a robust model for CCA performance analyss. Further research s also needed on the desgn of measurement models and mechansms to analyze the performance of a real Cloud Computng applcaton, whch could also contrbute to further valdate our proposed method. Such evaluaton work would nclude performance concepts related to software, hardware, and networkng. These concepts would be mapped to the collecton functons dentfed n the PMFCC prevously developed n order to mprove t. We expect that t wll be possble, based on ths work, to propose a robust model n future research that wll be able to analyze Hadoop cluster behavor n a real Cloud Computng envronment. Ths would allow real tme detecton of anomales that affect CCS and CCA performance. REFERENCES Bautsta, L., A. Abran, et al. (2012). "Desgn of a Performance Measurement Framework for Cloud Computng." Journal of Software Engneerng and Applcatons 5(2): Chekh, L. and A. Abran (2012). "Investgaton of the Relatonshps between the Software Qualty Models of ISO 9126 Standard: An Emprcal Study usng the Taguch Method." Software Qualty Professonal Magazne. Dean, J. and S. Ghemawat (2008). "MapReduce: 385

12 CLOSER2014-4thInternatonalConferenceonCloudComputngandServcesScence smplfed data processng on large clusters." Communcatons of the ACM 51(1): Hadoop, A. F. (2012). "What Is Apache Hadoop?", from ISO/IEC (2008). ISO/IEC 15939:2007 Systems and software engneerng Measurement process. Geneva, Swtzerland, Internatonal Organzaton for Standardzaton. ISO/IEC (2011). ISO/IEC 25010: Systems and software engneerng Systems and software product Qualty Requrements and Evaluaton (SQuaRE) System and software qualty models. Geneva, Swtzerland, Internatonal Organzaton for Standardzaton: 43. ISO/IEC (2011). ISO/IEC JTC 1 SC38:Study Group Report on Cloud Computng. Geneva, Swtzerland, Internatonal Organzaton for Standardzaton. Jackson, K. R., L. Ramakrshnan, et al. (2010). Performance Analyss of Hgh Performance Computng Applcatons on the Amazon Web Servces Cloud. IEEE Second Internatonal Conference on Cloud Computng Technology and Scence (CloudCom), Washngton, DC, USA, IEEE Computer Socety. Jan, R. (1991). The Art of Computer Systems Performance Analyss: Technques for Expermental Desgn, Measurement, Smulaton, and Modelng. New York, NY, John Wley & Sons - Interscence. Kramer, W., J. Shalf, et al. (2005). The NERSC Sustaned System Performance (SSP) Metrc. Calforna, USA, Lawrence Berkeley Natonal Laboratory. Ln, J. and C. Dyer (2010). Data-Intensve Text Processng wth MapReduce. Unversty of Maryland, College Park, Manuscrpt of a book n the Morgan & Claypool Synthess Lectures on Human Language Technologes. Me, Y., L. Lu, et al. (2010). Performance Measurements and Analyss of Network I/O Applcatons n Vrtualzed Cloud. IEEE Internatonal Conference on Cloud Computng, CLOUD 2010, Mam, FL, USA, IEEE. Mell, P. and T. Grance (2011). The NIST Defnton of Cloud Computng. Gathersburg, MD, USA, Informaton Technology Laboratory, Natonal Insttute of Standards and Technology: 2-3. Taguch, G., S. Chowdhury, et al. (2005). Taguch's Qualty Engneerng Handbook, John Wley & Sons, New Jersey. Trved, K. S. (2002). Probablty and Statstcs wth Relablty, Queung and Computer Scence Applcatons. New York, U.S.A., John Wley & Sons, Inc. 386

METHODOLOGY TO DETERMINE RELATIONSHIPS BETWEEN PERFORMANCE FACTORS IN HADOOP CLOUD COMPUTING APPLICATIONS

METHODOLOGY TO DETERMINE RELATIONSHIPS BETWEEN PERFORMANCE FACTORS IN HADOOP CLOUD COMPUTING APPLICATIONS METHODOLOGY TO DETERMINE RELATIONSHIPS BETWEEN PERFORMANCE FACTORS IN HADOOP CLOUD COMPUTING APPLICATIONS Lus Eduardo Bautsta Vllalpando 1,2, Alan Aprl 1 and Alan Abran 1 1 Department of Software Engneerng

More information

The Development of Web Log Mining Based on Improve-K-Means Clustering Analysis

The Development of Web Log Mining Based on Improve-K-Means Clustering Analysis The Development of Web Log Mnng Based on Improve-K-Means Clusterng Analyss TngZhong Wang * College of Informaton Technology, Luoyang Normal Unversty, Luoyang, 471022, Chna wangtngzhong2@sna.cn Abstract.

More information

Fault tolerance in cloud technologies presented as a service

Fault tolerance in cloud technologies presented as a service Internatonal Scentfc Conference Computer Scence 2015 Pavel Dzhunev, PhD student Fault tolerance n cloud technologes presented as a servce INTRODUCTION Improvements n technques for vrtualzaton and performance

More information

Survey on Virtual Machine Placement Techniques in Cloud Computing Environment

Survey on Virtual Machine Placement Techniques in Cloud Computing Environment Survey on Vrtual Machne Placement Technques n Cloud Computng Envronment Rajeev Kumar Gupta and R. K. Paterya Department of Computer Scence & Engneerng, MANIT, Bhopal, Inda ABSTRACT In tradtonal data center

More information

Feature selection for intrusion detection. Slobodan Petrović NISlab, Gjøvik University College

Feature selection for intrusion detection. Slobodan Petrović NISlab, Gjøvik University College Feature selecton for ntruson detecton Slobodan Petrovć NISlab, Gjøvk Unversty College Contents The feature selecton problem Intruson detecton Traffc features relevant for IDS The CFS measure The mrmr measure

More information

Data Broadcast on a Multi-System Heterogeneous Overlayed Wireless Network *

Data Broadcast on a Multi-System Heterogeneous Overlayed Wireless Network * JOURNAL OF INFORMATION SCIENCE AND ENGINEERING 24, 819-840 (2008) Data Broadcast on a Mult-System Heterogeneous Overlayed Wreless Network * Department of Computer Scence Natonal Chao Tung Unversty Hsnchu,

More information

A METHODOLOGY FOR IDENTIFYING THE RELATIONSHIP BETWEEN PERFORMANCE FACTORS FOR CLOUD COMPUTING APPLICATIONS

A METHODOLOGY FOR IDENTIFYING THE RELATIONSHIP BETWEEN PERFORMANCE FACTORS FOR CLOUD COMPUTING APPLICATIONS A METHODOLOGY FOR IDENTIFYING THE RELATIONSHIP BETWEEN PERFORMANCE FACTORS FOR CLOUD COMPUTING APPLICATIONS Luis Bautista 1, 2, Alain April 2, Alain Abran 2 1 Department of Electronic Systems, Autonomous

More information

Forecasting the Direction and Strength of Stock Market Movement

Forecasting the Direction and Strength of Stock Market Movement Forecastng the Drecton and Strength of Stock Market Movement Jngwe Chen Mng Chen Nan Ye cjngwe@stanford.edu mchen5@stanford.edu nanye@stanford.edu Abstract - Stock market s one of the most complcated systems

More information

A Replication-Based and Fault Tolerant Allocation Algorithm for Cloud Computing

A Replication-Based and Fault Tolerant Allocation Algorithm for Cloud Computing A Replcaton-Based and Fault Tolerant Allocaton Algorthm for Cloud Computng Tork Altameem Dept of Computer Scence, RCC, Kng Saud Unversty, PO Box: 28095 11437 Ryadh-Saud Araba Abstract The very large nfrastructure

More information

ANALYZING THE RELATIONSHIPS BETWEEN QUALITY, TIME, AND COST IN PROJECT MANAGEMENT DECISION MAKING

ANALYZING THE RELATIONSHIPS BETWEEN QUALITY, TIME, AND COST IN PROJECT MANAGEMENT DECISION MAKING ANALYZING THE RELATIONSHIPS BETWEEN QUALITY, TIME, AND COST IN PROJECT MANAGEMENT DECISION MAKING Matthew J. Lberatore, Department of Management and Operatons, Vllanova Unversty, Vllanova, PA 19085, 610-519-4390,

More information

Activity Scheduling for Cost-Time Investment Optimization in Project Management

Activity Scheduling for Cost-Time Investment Optimization in Project Management PROJECT MANAGEMENT 4 th Internatonal Conference on Industral Engneerng and Industral Management XIV Congreso de Ingenería de Organzacón Donosta- San Sebastán, September 8 th -10 th 010 Actvty Schedulng

More information

Politecnico di Torino. Porto Institutional Repository

Politecnico di Torino. Porto Institutional Repository Poltecnco d Torno Porto Insttutonal Repostory [Artcle] A cost-effectve cloud computng framework for acceleratng multmeda communcaton smulatons Orgnal Ctaton: D. Angel, E. Masala (2012). A cost-effectve

More information

IMPACT ANALYSIS OF A CELLULAR PHONE

IMPACT ANALYSIS OF A CELLULAR PHONE 4 th ASA & μeta Internatonal Conference IMPACT AALYSIS OF A CELLULAR PHOE We Lu, 2 Hongy L Bejng FEAonlne Engneerng Co.,Ltd. Bejng, Chna ABSTRACT Drop test smulaton plays an mportant role n nvestgatng

More information

An Interest-Oriented Network Evolution Mechanism for Online Communities

An Interest-Oriented Network Evolution Mechanism for Online Communities An Interest-Orented Network Evoluton Mechansm for Onlne Communtes Cahong Sun and Xaopng Yang School of Informaton, Renmn Unversty of Chna, Bejng 100872, P.R. Chna {chsun,yang}@ruc.edu.cn Abstract. Onlne

More information

benefit is 2, paid if the policyholder dies within the year, and probability of death within the year is ).

benefit is 2, paid if the policyholder dies within the year, and probability of death within the year is ). REVIEW OF RISK MANAGEMENT CONCEPTS LOSS DISTRIBUTIONS AND INSURANCE Loss and nsurance: When someone s subject to the rsk of ncurrng a fnancal loss, the loss s generally modeled usng a random varable or

More information

Network Security Situation Evaluation Method for Distributed Denial of Service

Network Security Situation Evaluation Method for Distributed Denial of Service Network Securty Stuaton Evaluaton Method for Dstrbuted Denal of Servce Jn Q,2, Cu YMn,2, Huang MnHuan,2, Kuang XaoHu,2, TangHong,2 ) Scence and Technology on Informaton System Securty Laboratory, Bejng,

More information

Communication Networks II Contents

Communication Networks II Contents 8 / 1 -- Communcaton Networs II (Görg) -- www.comnets.un-bremen.de Communcaton Networs II Contents 1 Fundamentals of probablty theory 2 Traffc n communcaton networs 3 Stochastc & Marovan Processes (SP

More information

Project Networks With Mixed-Time Constraints

Project Networks With Mixed-Time Constraints Project Networs Wth Mxed-Tme Constrants L Caccetta and B Wattananon Western Australan Centre of Excellence n Industral Optmsaton (WACEIO) Curtn Unversty of Technology GPO Box U1987 Perth Western Australa

More information

DEFINING %COMPLETE IN MICROSOFT PROJECT

DEFINING %COMPLETE IN MICROSOFT PROJECT CelersSystems DEFINING %COMPLETE IN MICROSOFT PROJECT PREPARED BY James E Aksel, PMP, PMI-SP, MVP For Addtonal Informaton about Earned Value Management Systems and reportng, please contact: CelersSystems,

More information

Efficient Striping Techniques for Variable Bit Rate Continuous Media File Servers æ

Efficient Striping Techniques for Variable Bit Rate Continuous Media File Servers æ Effcent Strpng Technques for Varable Bt Rate Contnuous Meda Fle Servers æ Prashant J. Shenoy Harrck M. Vn Department of Computer Scence, Department of Computer Scences, Unversty of Massachusetts at Amherst

More information

CHOLESTEROL REFERENCE METHOD LABORATORY NETWORK. Sample Stability Protocol

CHOLESTEROL REFERENCE METHOD LABORATORY NETWORK. Sample Stability Protocol CHOLESTEROL REFERENCE METHOD LABORATORY NETWORK Sample Stablty Protocol Background The Cholesterol Reference Method Laboratory Network (CRMLN) developed certfcaton protocols for total cholesterol, HDL

More information

HP Mission-Critical Services

HP Mission-Critical Services HP Msson-Crtcal Servces Delverng busness value to IT Jelena Bratc Zarko Subotc TS Support tm Mart 2012, Podgorca 2010 Hewlett-Packard Development Company, L.P. The nformaton contaned heren s subject to

More information

MAPP. MERIS level 3 cloud and water vapour products. Issue: 1. Revision: 0. Date: 9.12.1998. Function Name Organisation Signature Date

MAPP. MERIS level 3 cloud and water vapour products. Issue: 1. Revision: 0. Date: 9.12.1998. Function Name Organisation Signature Date Ttel: Project: Doc. No.: MERIS level 3 cloud and water vapour products MAPP MAPP-ATBD-ClWVL3 Issue: 1 Revson: 0 Date: 9.12.1998 Functon Name Organsaton Sgnature Date Author: Bennartz FUB Preusker FUB Schüller

More information

Overview of monitoring and evaluation

Overview of monitoring and evaluation 540 Toolkt to Combat Traffckng n Persons Tool 10.1 Overvew of montorng and evaluaton Overvew Ths tool brefly descrbes both montorng and evaluaton, and the dstncton between the two. What s montorng? Montorng

More information

Open Access A Load Balancing Strategy with Bandwidth Constraint in Cloud Computing. Jing Deng 1,*, Ping Guo 2, Qi Li 3, Haizhu Chen 1

Open Access A Load Balancing Strategy with Bandwidth Constraint in Cloud Computing. Jing Deng 1,*, Ping Guo 2, Qi Li 3, Haizhu Chen 1 Send Orders for Reprnts to reprnts@benthamscence.ae The Open Cybernetcs & Systemcs Journal, 2014, 8, 115-121 115 Open Access A Load Balancng Strategy wth Bandwdth Constrant n Cloud Computng Jng Deng 1,*,

More information

Traffic State Estimation in the Traffic Management Center of Berlin

Traffic State Estimation in the Traffic Management Center of Berlin Traffc State Estmaton n the Traffc Management Center of Berln Authors: Peter Vortsch, PTV AG, Stumpfstrasse, D-763 Karlsruhe, Germany phone ++49/72/965/35, emal peter.vortsch@ptv.de Peter Möhl, PTV AG,

More information

A DYNAMIC CRASHING METHOD FOR PROJECT MANAGEMENT USING SIMULATION-BASED OPTIMIZATION. Michael E. Kuhl Radhamés A. Tolentino-Peña

A DYNAMIC CRASHING METHOD FOR PROJECT MANAGEMENT USING SIMULATION-BASED OPTIMIZATION. Michael E. Kuhl Radhamés A. Tolentino-Peña Proceedngs of the 2008 Wnter Smulaton Conference S. J. Mason, R. R. Hll, L. Mönch, O. Rose, T. Jefferson, J. W. Fowler eds. A DYNAMIC CRASHING METHOD FOR PROJECT MANAGEMENT USING SIMULATION-BASED OPTIMIZATION

More information

On the Optimal Control of a Cascade of Hydro-Electric Power Stations

On the Optimal Control of a Cascade of Hydro-Electric Power Stations On the Optmal Control of a Cascade of Hydro-Electrc Power Statons M.C.M. Guedes a, A.F. Rbero a, G.V. Smrnov b and S. Vlela c a Department of Mathematcs, School of Scences, Unversty of Porto, Portugal;

More information

The Analysis of Outliers in Statistical Data

The Analysis of Outliers in Statistical Data THALES Project No. xxxx The Analyss of Outlers n Statstcal Data Research Team Chrysses Caron, Assocate Professor (P.I.) Vaslk Karot, Doctoral canddate Polychrons Economou, Chrstna Perrakou, Postgraduate

More information

Study on Model of Risks Assessment of Standard Operation in Rural Power Network

Study on Model of Risks Assessment of Standard Operation in Rural Power Network Study on Model of Rsks Assessment of Standard Operaton n Rural Power Network Qngj L 1, Tao Yang 2 1 Qngj L, College of Informaton and Electrcal Engneerng, Shenyang Agrculture Unversty, Shenyang 110866,

More information

Design and Development of a Security Evaluation Platform Based on International Standards

Design and Development of a Security Evaluation Platform Based on International Standards Internatonal Journal of Informatcs Socety, VOL.5, NO.2 (203) 7-80 7 Desgn and Development of a Securty Evaluaton Platform Based on Internatonal Standards Yuj Takahash and Yoshm Teshgawara Graduate School

More information

A hybrid global optimization algorithm based on parallel chaos optimization and outlook algorithm

A hybrid global optimization algorithm based on parallel chaos optimization and outlook algorithm Avalable onlne www.ocpr.com Journal of Chemcal and Pharmaceutcal Research, 2014, 6(7):1884-1889 Research Artcle ISSN : 0975-7384 CODEN(USA) : JCPRC5 A hybrd global optmzaton algorthm based on parallel

More information

APPLICATION OF PROBE DATA COLLECTED VIA INFRARED BEACONS TO TRAFFIC MANEGEMENT

APPLICATION OF PROBE DATA COLLECTED VIA INFRARED BEACONS TO TRAFFIC MANEGEMENT APPLICATION OF PROBE DATA COLLECTED VIA INFRARED BEACONS TO TRAFFIC MANEGEMENT Toshhko Oda (1), Kochro Iwaoka (2) (1), (2) Infrastructure Systems Busness Unt, Panasonc System Networks Co., Ltd. Saedo-cho

More information

Research of Network System Reconfigurable Model Based on the Finite State Automation

Research of Network System Reconfigurable Model Based on the Finite State Automation JOURNAL OF NETWORKS, VOL., NO. 5, MAY 24 237 Research of Network System Reconfgurable Model Based on the Fnte State Automaton Shenghan Zhou and Wenbng Chang School of Relablty and System Engneerng, Behang

More information

To manage leave, meeting institutional requirements and treating individual staff members fairly and consistently.

To manage leave, meeting institutional requirements and treating individual staff members fairly and consistently. Corporate Polces & Procedures Human Resources - Document CPP216 Leave Management Frst Produced: Current Verson: Past Revsons: Revew Cycle: Apples From: 09/09/09 26/10/12 09/09/09 3 years Immedately Authorsaton:

More information

Can Auto Liability Insurance Purchases Signal Risk Attitude?

Can Auto Liability Insurance Purchases Signal Risk Attitude? Internatonal Journal of Busness and Economcs, 2011, Vol. 10, No. 2, 159-164 Can Auto Lablty Insurance Purchases Sgnal Rsk Atttude? Chu-Shu L Department of Internatonal Busness, Asa Unversty, Tawan Sheng-Chang

More information

An Analysis of Central Processor Scheduling in Multiprogrammed Computer Systems

An Analysis of Central Processor Scheduling in Multiprogrammed Computer Systems STAN-CS-73-355 I SU-SE-73-013 An Analyss of Central Processor Schedulng n Multprogrammed Computer Systems (Dgest Edton) by Thomas G. Prce October 1972 Techncal Report No. 57 Reproducton n whole or n part

More information

Master s Thesis. Configuring robust virtual wireless sensor networks for Internet of Things inspired by brain functional networks

Master s Thesis. Configuring robust virtual wireless sensor networks for Internet of Things inspired by brain functional networks Master s Thess Ttle Confgurng robust vrtual wreless sensor networks for Internet of Thngs nspred by bran functonal networks Supervsor Professor Masayuk Murata Author Shnya Toyonaga February 10th, 2014

More information

IWFMS: An Internal Workflow Management System/Optimizer for Hadoop

IWFMS: An Internal Workflow Management System/Optimizer for Hadoop IWFMS: An Internal Workflow Management System/Optmzer for Hadoop Lan Lu, Yao Shen Department of Computer Scence and Engneerng Shangha JaoTong Unversty Shangha, Chna lustrve@gmal.com, yshen@cs.sjtu.edu.cn

More information

Forecasting the Demand of Emergency Supplies: Based on the CBR Theory and BP Neural Network

Forecasting the Demand of Emergency Supplies: Based on the CBR Theory and BP Neural Network 700 Proceedngs of the 8th Internatonal Conference on Innovaton & Management Forecastng the Demand of Emergency Supples: Based on the CBR Theory and BP Neural Network Fu Deqang, Lu Yun, L Changbng School

More information

DBA-VM: Dynamic Bandwidth Allocator for Virtual Machines

DBA-VM: Dynamic Bandwidth Allocator for Virtual Machines DBA-VM: Dynamc Bandwdth Allocator for Vrtual Machnes Ahmed Amamou, Manel Bourguba, Kamel Haddadou and Guy Pujolle LIP6, Perre & Mare Cure Unversty, 4 Place Jusseu 755 Pars, France Gand SAS, 65 Boulevard

More information

Vision Mouse. Saurabh Sarkar a* University of Cincinnati, Cincinnati, USA ABSTRACT 1. INTRODUCTION

Vision Mouse. Saurabh Sarkar a* University of Cincinnati, Cincinnati, USA ABSTRACT 1. INTRODUCTION Vson Mouse Saurabh Sarkar a* a Unversty of Cncnnat, Cncnnat, USA ABSTRACT The report dscusses a vson based approach towards trackng of eyes and fngers. The report descrbes the process of locatng the possble

More information

A Cost-Effective Strategy for Intermediate Data Storage in Scientific Cloud Workflow Systems

A Cost-Effective Strategy for Intermediate Data Storage in Scientific Cloud Workflow Systems A Cost-Effectve Strategy for Intermedate Data Storage n Scentfc Cloud Workflow Systems Dong Yuan, Yun Yang, Xao Lu, Jnjun Chen Faculty of Informaton and Communcaton Technologes, Swnburne Unversty of Technology

More information

AN APPOINTMENT ORDER OUTPATIENT SCHEDULING SYSTEM THAT IMPROVES OUTPATIENT EXPERIENCE

AN APPOINTMENT ORDER OUTPATIENT SCHEDULING SYSTEM THAT IMPROVES OUTPATIENT EXPERIENCE AN APPOINTMENT ORDER OUTPATIENT SCHEDULING SYSTEM THAT IMPROVES OUTPATIENT EXPERIENCE Yu-L Huang Industral Engneerng Department New Mexco State Unversty Las Cruces, New Mexco 88003, U.S.A. Abstract Patent

More information

Cloud-based Social Application Deployment using Local Processing and Global Distribution

Cloud-based Social Application Deployment using Local Processing and Global Distribution Cloud-based Socal Applcaton Deployment usng Local Processng and Global Dstrbuton Zh Wang *, Baochun L, Lfeng Sun *, and Shqang Yang * * Bejng Key Laboratory of Networked Multmeda Department of Computer

More information

Optimization Model of Reliable Data Storage in Cloud Environment Using Genetic Algorithm

Optimization Model of Reliable Data Storage in Cloud Environment Using Genetic Algorithm Internatonal Journal of Grd Dstrbuton Computng, pp.175-190 http://dx.do.org/10.14257/gdc.2014.7.6.14 Optmzaton odel of Relable Data Storage n Cloud Envronment Usng Genetc Algorthm Feng Lu 1,2,3, Hatao

More information

INVESTIGATION OF VEHICULAR USERS FAIRNESS IN CDMA-HDR NETWORKS

INVESTIGATION OF VEHICULAR USERS FAIRNESS IN CDMA-HDR NETWORKS 21 22 September 2007, BULGARIA 119 Proceedngs of the Internatonal Conference on Informaton Technologes (InfoTech-2007) 21 st 22 nd September 2007, Bulgara vol. 2 INVESTIGATION OF VEHICULAR USERS FAIRNESS

More information

Effective Network Defense Strategies against Malicious Attacks with Various Defense Mechanisms under Quality of Service Constraints

Effective Network Defense Strategies against Malicious Attacks with Various Defense Mechanisms under Quality of Service Constraints Effectve Network Defense Strateges aganst Malcous Attacks wth Varous Defense Mechansms under Qualty of Servce Constrants Frank Yeong-Sung Ln Department of Informaton Natonal Tawan Unversty Tape, Tawan,

More information

Brigid Mullany, Ph.D University of North Carolina, Charlotte

Brigid Mullany, Ph.D University of North Carolina, Charlotte Evaluaton And Comparson Of The Dfferent Standards Used To Defne The Postonal Accuracy And Repeatablty Of Numercally Controlled Machnng Center Axes Brgd Mullany, Ph.D Unversty of North Carolna, Charlotte

More information

An Evaluation of the Extended Logistic, Simple Logistic, and Gompertz Models for Forecasting Short Lifecycle Products and Services

An Evaluation of the Extended Logistic, Simple Logistic, and Gompertz Models for Forecasting Short Lifecycle Products and Services An Evaluaton of the Extended Logstc, Smple Logstc, and Gompertz Models for Forecastng Short Lfecycle Products and Servces Charles V. Trappey a,1, Hsn-yng Wu b a Professor (Management Scence), Natonal Chao

More information

What is Candidate Sampling

What is Candidate Sampling What s Canddate Samplng Say we have a multclass or mult label problem where each tranng example ( x, T ) conssts of a context x a small (mult)set of target classes T out of a large unverse L of possble

More information

BUSINESS PROCESS PERFORMANCE MANAGEMENT USING BAYESIAN BELIEF NETWORK. 0688, dskim@ssu.ac.kr

BUSINESS PROCESS PERFORMANCE MANAGEMENT USING BAYESIAN BELIEF NETWORK. 0688, dskim@ssu.ac.kr Proceedngs of the 41st Internatonal Conference on Computers & Industral Engneerng BUSINESS PROCESS PERFORMANCE MANAGEMENT USING BAYESIAN BELIEF NETWORK Yeong-bn Mn 1, Yongwoo Shn 2, Km Jeehong 1, Dongsoo

More information

Optimization of network mesh topologies and link capacities for congestion relief

Optimization of network mesh topologies and link capacities for congestion relief Optmzaton of networ mesh topologes and ln capactes for congeston relef D. de Vllers * J.M. Hattngh School of Computer-, Statstcal- and Mathematcal Scences Potchefstroom Unversty for CHE * E-mal: rwddv@pu.ac.za

More information

Efficient Project Portfolio as a tool for Enterprise Risk Management

Efficient Project Portfolio as a tool for Enterprise Risk Management Effcent Proect Portfolo as a tool for Enterprse Rsk Management Valentn O. Nkonov Ural State Techncal Unversty Growth Traectory Consultng Company January 5, 27 Effcent Proect Portfolo as a tool for Enterprse

More information

A Secure Password-Authenticated Key Agreement Using Smart Cards

A Secure Password-Authenticated Key Agreement Using Smart Cards A Secure Password-Authentcated Key Agreement Usng Smart Cards Ka Chan 1, Wen-Chung Kuo 2 and Jn-Chou Cheng 3 1 Department of Computer and Informaton Scence, R.O.C. Mltary Academy, Kaohsung 83059, Tawan,

More information

A Design Method of High-availability and Low-optical-loss Optical Aggregation Network Architecture

A Design Method of High-availability and Low-optical-loss Optical Aggregation Network Architecture A Desgn Method of Hgh-avalablty and Low-optcal-loss Optcal Aggregaton Network Archtecture Takehro Sato, Kuntaka Ashzawa, Kazumasa Tokuhash, Dasuke Ish, Satoru Okamoto and Naoak Yamanaka Dept. of Informaton

More information

Risk Model of Long-Term Production Scheduling in Open Pit Gold Mining

Risk Model of Long-Term Production Scheduling in Open Pit Gold Mining Rsk Model of Long-Term Producton Schedulng n Open Pt Gold Mnng R Halatchev 1 and P Lever 2 ABSTRACT Open pt gold mnng s an mportant sector of the Australan mnng ndustry. It uses large amounts of nvestments,

More information

On-Line Fault Detection in Wind Turbine Transmission System using Adaptive Filter and Robust Statistical Features

On-Line Fault Detection in Wind Turbine Transmission System using Adaptive Filter and Robust Statistical Features On-Lne Fault Detecton n Wnd Turbne Transmsson System usng Adaptve Flter and Robust Statstcal Features Ruoyu L Remote Dagnostcs Center SKF USA Inc. 3443 N. Sam Houston Pkwy., Houston TX 77086 Emal: ruoyu.l@skf.com

More information

iavenue iavenue i i i iavenue iavenue iavenue

iavenue iavenue i i i iavenue iavenue iavenue Saratoga Systems' enterprse-wde Avenue CRM system s a comprehensve web-enabled software soluton. Ths next generaton system enables you to effectvely manage and enhance your customer relatonshps n both

More information

Improved SVM in Cloud Computing Information Mining

Improved SVM in Cloud Computing Information Mining Internatonal Journal of Grd Dstrbuton Computng Vol.8, No.1 (015), pp.33-40 http://dx.do.org/10.1457/jgdc.015.8.1.04 Improved n Cloud Computng Informaton Mnng Lvshuhong (ZhengDe polytechnc college JangSu

More information

Cloud Auto-Scaling with Deadline and Budget Constraints

Cloud Auto-Scaling with Deadline and Budget Constraints Prelmnary verson. Fnal verson appears In Proceedngs of 11th ACM/IEEE Internatonal Conference on Grd Computng (Grd 21). Oct 25-28, 21. Brussels, Belgum. Cloud Auto-Scalng wth Deadlne and Budget Constrants

More information

Stochastic Protocol Modeling for Anomaly Based Network Intrusion Detection

Stochastic Protocol Modeling for Anomaly Based Network Intrusion Detection Stochastc Protocol Modelng for Anomaly Based Network Intruson Detecton Juan M. Estevez-Tapador, Pedro Garca-Teodoro, and Jesus E. Daz-Verdejo Department of Electroncs and Computer Technology Unversty of

More information

Business Process Improvement using Multi-objective Optimisation K. Vergidis 1, A. Tiwari 1 and B. Majeed 2

Business Process Improvement using Multi-objective Optimisation K. Vergidis 1, A. Tiwari 1 and B. Majeed 2 Busness Process Improvement usng Mult-objectve Optmsaton K. Vergds 1, A. Twar 1 and B. Majeed 2 1 Manufacturng Department, School of Industral and Manufacturng Scence, Cranfeld Unversty, Cranfeld, MK43

More information

The OC Curve of Attribute Acceptance Plans

The OC Curve of Attribute Acceptance Plans The OC Curve of Attrbute Acceptance Plans The Operatng Characterstc (OC) curve descrbes the probablty of acceptng a lot as a functon of the lot s qualty. Fgure 1 shows a typcal OC Curve. 10 8 6 4 1 3 4

More information

Daily Mood Assessment based on Mobile Phone Sensing

Daily Mood Assessment based on Mobile Phone Sensing 2012 Nnth Internatonal Conference on Wearable and Implantable Body Sensor Networks Daly Mood Assessment based on Moble Phone Sensng Yuanchao Ma Bn Xu Yn Ba Guodong Sun Department of Computer Scence and

More information

A General Simulation Framework for Supply Chain Modeling: State of the Art and Case Study

A General Simulation Framework for Supply Chain Modeling: State of the Art and Case Study IJCSI Internatonal Journal of Computer Scence Issues, Vol. 7, Issue 2, No 3, March 2010 ISSN (Onlne): 1694-0784 ISSN (Prnt): 1694-0814 1 A General Smulaton Framework for Supply Chan Modelng: State of the

More information

"Research Note" APPLICATION OF CHARGE SIMULATION METHOD TO ELECTRIC FIELD CALCULATION IN THE POWER CABLES *

Research Note APPLICATION OF CHARGE SIMULATION METHOD TO ELECTRIC FIELD CALCULATION IN THE POWER CABLES * Iranan Journal of Scence & Technology, Transacton B, Engneerng, ol. 30, No. B6, 789-794 rnted n The Islamc Republc of Iran, 006 Shraz Unversty "Research Note" ALICATION OF CHARGE SIMULATION METHOD TO ELECTRIC

More information

Traditional versus Online Courses, Efforts, and Learning Performance

Traditional versus Online Courses, Efforts, and Learning Performance Tradtonal versus Onlne Courses, Efforts, and Learnng Performance Kuang-Cheng Tseng, Department of Internatonal Trade, Chung-Yuan Chrstan Unversty, Tawan Shan-Yng Chu, Department of Internatonal Trade,

More information

Performance Analysis of Energy Consumption of Smartphone Running Mobile Hotspot Application

Performance Analysis of Energy Consumption of Smartphone Running Mobile Hotspot Application Internatonal Journal of mart Grd and lean Energy Performance Analyss of Energy onsumpton of martphone Runnng Moble Hotspot Applcaton Yun on hung a chool of Electronc Engneerng, oongsl Unversty, 511 angdo-dong,

More information

Bayesian Network Based Causal Relationship Identification and Funding Success Prediction in P2P Lending

Bayesian Network Based Causal Relationship Identification and Funding Success Prediction in P2P Lending Proceedngs of 2012 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 25 (2012) (2012) IACSIT Press, Sngapore Bayesan Network Based Causal Relatonshp Identfcaton and Fundng Success

More information

Risk-based Fatigue Estimate of Deep Water Risers -- Course Project for EM388F: Fracture Mechanics, Spring 2008

Risk-based Fatigue Estimate of Deep Water Risers -- Course Project for EM388F: Fracture Mechanics, Spring 2008 Rsk-based Fatgue Estmate of Deep Water Rsers -- Course Project for EM388F: Fracture Mechancs, Sprng 2008 Chen Sh Department of Cvl, Archtectural, and Envronmental Engneerng The Unversty of Texas at Austn

More information

Module 2 LOSSLESS IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur

Module 2 LOSSLESS IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur Module LOSSLESS IMAGE COMPRESSION SYSTEMS Lesson 3 Lossless Compresson: Huffman Codng Instructonal Objectves At the end of ths lesson, the students should be able to:. Defne and measure source entropy..

More information

Calculation of Sampling Weights

Calculation of Sampling Weights Perre Foy Statstcs Canada 4 Calculaton of Samplng Weghts 4.1 OVERVIEW The basc sample desgn used n TIMSS Populatons 1 and 2 was a two-stage stratfed cluster desgn. 1 The frst stage conssted of a sample

More information

The Analysis of Covariance. ERSH 8310 Keppel and Wickens Chapter 15

The Analysis of Covariance. ERSH 8310 Keppel and Wickens Chapter 15 The Analyss of Covarance ERSH 830 Keppel and Wckens Chapter 5 Today s Class Intal Consderatons Covarance and Lnear Regresson The Lnear Regresson Equaton TheAnalyss of Covarance Assumptons Underlyng the

More information

Data Mining from the Information Systems: Performance Indicators at Masaryk University in Brno

Data Mining from the Information Systems: Performance Indicators at Masaryk University in Brno Data Mnng from the Informaton Systems: Performance Indcators at Masaryk Unversty n Brno Mkuláš Bek EUA Workshop Strasbourg, 1-2 December 2006 1 Locaton of Brno Brno EUA Workshop Strasbourg, 1-2 December

More information

Causal, Explanatory Forecasting. Analysis. Regression Analysis. Simple Linear Regression. Which is Independent? Forecasting

Causal, Explanatory Forecasting. Analysis. Regression Analysis. Simple Linear Regression. Which is Independent? Forecasting Causal, Explanatory Forecastng Assumes cause-and-effect relatonshp between system nputs and ts output Forecastng wth Regresson Analyss Rchard S. Barr Inputs System Cause + Effect Relatonshp The job of

More information

Damage detection in composite laminates using coin-tap method

Damage detection in composite laminates using coin-tap method Damage detecton n composte lamnates usng con-tap method S.J. Km Korea Aerospace Research Insttute, 45 Eoeun-Dong, Youseong-Gu, 35-333 Daejeon, Republc of Korea yaeln@kar.re.kr 45 The con-tap test has the

More information

Calculating the high frequency transmission line parameters of power cables

Calculating the high frequency transmission line parameters of power cables < ' Calculatng the hgh frequency transmsson lne parameters of power cables Authors: Dr. John Dcknson, Laboratory Servces Manager, N 0 RW E B Communcatons Mr. Peter J. Ncholson, Project Assgnment Manager,

More information

RESEARCH ON DUAL-SHAKER SINE VIBRATION CONTROL. Yaoqi FENG 1, Hanping QIU 1. China Academy of Space Technology (CAST) yaoqi.feng@yahoo.

RESEARCH ON DUAL-SHAKER SINE VIBRATION CONTROL. Yaoqi FENG 1, Hanping QIU 1. China Academy of Space Technology (CAST) yaoqi.feng@yahoo. ICSV4 Carns Australa 9- July, 007 RESEARCH ON DUAL-SHAKER SINE VIBRATION CONTROL Yaoq FENG, Hanpng QIU Dynamc Test Laboratory, BISEE Chna Academy of Space Technology (CAST) yaoq.feng@yahoo.com Abstract

More information

A Programming Model for the Cloud Platform

A Programming Model for the Cloud Platform Internatonal Journal of Advanced Scence and Technology A Programmng Model for the Cloud Platform Xaodong Lu School of Computer Engneerng and Scence Shangha Unversty, Shangha 200072, Chna luxaodongxht@qq.com

More information

Software project management with GAs

Software project management with GAs Informaton Scences 177 (27) 238 241 www.elsever.com/locate/ns Software project management wth GAs Enrque Alba *, J. Francsco Chcano Unversty of Málaga, Grupo GISUM, Departamento de Lenguajes y Cencas de

More information

行 政 院 國 家 科 學 委 員 會 補 助 專 題 研 究 計 畫 成 果 報 告 期 中 進 度 報 告

行 政 院 國 家 科 學 委 員 會 補 助 專 題 研 究 計 畫 成 果 報 告 期 中 進 度 報 告 行 政 院 國 家 科 學 委 員 會 補 助 專 題 研 究 計 畫 成 果 報 告 期 中 進 度 報 告 畫 類 別 : 個 別 型 計 畫 半 導 體 產 業 大 型 廠 房 之 設 施 規 劃 計 畫 編 號 :NSC 96-2628-E-009-026-MY3 執 行 期 間 : 2007 年 8 月 1 日 至 2010 年 7 月 31 日 計 畫 主 持 人 : 巫 木 誠 共 同

More information

An MILP model for planning of batch plants operating in a campaign-mode

An MILP model for planning of batch plants operating in a campaign-mode An MILP model for plannng of batch plants operatng n a campagn-mode Yanna Fumero Insttuto de Desarrollo y Dseño CONICET UTN yfumero@santafe-concet.gov.ar Gabrela Corsano Insttuto de Desarrollo y Dseño

More information

LITERATURE REVIEW: VARIOUS PRIORITY BASED TASK SCHEDULING ALGORITHMS IN CLOUD COMPUTING

LITERATURE REVIEW: VARIOUS PRIORITY BASED TASK SCHEDULING ALGORITHMS IN CLOUD COMPUTING LITERATURE REVIEW: VARIOUS PRIORITY BASED TASK SCHEDULING ALGORITHMS IN CLOUD COMPUTING 1 MS. POOJA.P.VASANI, 2 MR. NISHANT.S. SANGHANI 1 M.Tech. [Software Systems] Student, Patel College of Scence and

More information

Course outline. Financial Time Series Analysis. Overview. Data analysis. Predictive signal. Trading strategy

Course outline. Financial Time Series Analysis. Overview. Data analysis. Predictive signal. Trading strategy Fnancal Tme Seres Analyss Patrck McSharry patrck@mcsharry.net www.mcsharry.net Trnty Term 2014 Mathematcal Insttute Unversty of Oxford Course outlne 1. Data analyss, probablty, correlatons, vsualsaton

More information

2008/8. An integrated model for warehouse and inventory planning. Géraldine Strack and Yves Pochet

2008/8. An integrated model for warehouse and inventory planning. Géraldine Strack and Yves Pochet 2008/8 An ntegrated model for warehouse and nventory plannng Géraldne Strack and Yves Pochet CORE Voe du Roman Pays 34 B-1348 Louvan-la-Neuve, Belgum. Tel (32 10) 47 43 04 Fax (32 10) 47 43 01 E-mal: corestat-lbrary@uclouvan.be

More information

Load Balancing By Max-Min Algorithm in Private Cloud Environment

Load Balancing By Max-Min Algorithm in Private Cloud Environment Internatonal Journal of Scence and Research (IJSR ISSN (Onlne: 2319-7064 Index Coperncus Value (2013: 6.14 Impact Factor (2013: 4.438 Load Balancng By Max-Mn Algorthm n Prvate Cloud Envronment S M S Suntharam

More information

Multiple-Period Attribution: Residuals and Compounding

Multiple-Period Attribution: Residuals and Compounding Multple-Perod Attrbuton: Resduals and Compoundng Our revewer gave these authors full marks for dealng wth an ssue that performance measurers and vendors often regard as propretary nformaton. In 1994, Dens

More information

The Application of Fractional Brownian Motion in Option Pricing

The Application of Fractional Brownian Motion in Option Pricing Vol. 0, No. (05), pp. 73-8 http://dx.do.org/0.457/jmue.05.0..6 The Applcaton of Fractonal Brownan Moton n Opton Prcng Qng-xn Zhou School of Basc Scence,arbn Unversty of Commerce,arbn zhouqngxn98@6.com

More information

Network Aware Load-Balancing via Parallel VM Migration for Data Centers

Network Aware Load-Balancing via Parallel VM Migration for Data Centers Network Aware Load-Balancng va Parallel VM Mgraton for Data Centers Kun-Tng Chen 2, Chen Chen 12, Po-Hsang Wang 2 1 Informaton Technology Servce Center, 2 Department of Computer Scence Natonal Chao Tung

More information

Demographic and Health Surveys Methodology

Demographic and Health Surveys Methodology samplng and household lstng manual Demographc and Health Surveys Methodology Ths document s part of the Demographc and Health Survey s DHS Toolkt of methodology for the MEASURE DHS Phase III project, mplemented

More information

A New Task Scheduling Algorithm Based on Improved Genetic Algorithm

A New Task Scheduling Algorithm Based on Improved Genetic Algorithm A New Task Schedulng Algorthm Based on Improved Genetc Algorthm n Cloud Computng Envronment Congcong Xong, Long Feng, Lxan Chen A New Task Schedulng Algorthm Based on Improved Genetc Algorthm n Cloud Computng

More information

FORMAL ANALYSIS FOR REAL-TIME SCHEDULING

FORMAL ANALYSIS FOR REAL-TIME SCHEDULING FORMAL ANALYSIS FOR REAL-TIME SCHEDULING Bruno Dutertre and Vctora Stavrdou, SRI Internatonal, Menlo Park, CA Introducton In modern avoncs archtectures, applcaton software ncreasngly reles on servces provded

More information

M3S MULTIMEDIA MOBILITY MANAGEMENT AND LOAD BALANCING IN WIRELESS BROADCAST NETWORKS

M3S MULTIMEDIA MOBILITY MANAGEMENT AND LOAD BALANCING IN WIRELESS BROADCAST NETWORKS M3S MULTIMEDIA MOBILITY MANAGEMENT AND LOAD BALANCING IN WIRELESS BROADCAST NETWORKS Bogdan Cubotaru, Gabrel-Mro Muntean Performance Engneerng Laboratory, RINCE School of Electronc Engneerng Dubln Cty

More information

) of the Cell class is created containing information about events associated with the cell. Events are added to the Cell instance

) of the Cell class is created containing information about events associated with the cell. Events are added to the Cell instance Calbraton Method Instances of the Cell class (one nstance for each FMS cell) contan ADC raw data and methods assocated wth each partcular FMS cell. The calbraton method ncludes event selecton (Class Cell

More information

The Greedy Method. Introduction. 0/1 Knapsack Problem

The Greedy Method. Introduction. 0/1 Knapsack Problem The Greedy Method Introducton We have completed data structures. We now are gong to look at algorthm desgn methods. Often we are lookng at optmzaton problems whose performance s exponental. For an optmzaton

More information

Enabling P2P One-view Multi-party Video Conferencing

Enabling P2P One-view Multi-party Video Conferencing Enablng P2P One-vew Mult-party Vdeo Conferencng Yongxang Zhao, Yong Lu, Changja Chen, and JanYn Zhang Abstract Mult-Party Vdeo Conferencng (MPVC) facltates realtme group nteracton between users. Whle P2P

More information

An Alternative Way to Measure Private Equity Performance

An Alternative Way to Measure Private Equity Performance An Alternatve Way to Measure Prvate Equty Performance Peter Todd Parlux Investment Technology LLC Summary Internal Rate of Return (IRR) s probably the most common way to measure the performance of prvate

More information

Selecting Best Employee of the Year Using Analytical Hierarchy Process

Selecting Best Employee of the Year Using Analytical Hierarchy Process J. Basc. Appl. Sc. Res., 5(11)72-76, 2015 2015, TextRoad Publcaton ISSN 2090-4304 Journal of Basc and Appled Scentfc Research www.textroad.com Selectng Best Employee of the Year Usng Analytcal Herarchy

More information

Dynamic Resource Allocation and Power Management in Virtualized Data Centers

Dynamic Resource Allocation and Power Management in Virtualized Data Centers Dynamc Resource Allocaton and Power Management n Vrtualzed Data Centers Rahul Urgaonkar, Ulas C. Kozat, Ken Igarash, Mchael J. Neely urgaonka@usc.edu, {kozat, garash}@docomolabs-usa.com, mjneely@usc.edu

More information