Scheduling Hadoop Jobs to Meet Deadlines

Size: px
Start display at page:

Download "Scheduling Hadoop Jobs to Meet Deadlines"

Transcription

1 Scheduling Hadoop Jobs to Meet Deadlines Kamal Kc, Kemafo Anyanwu Depatment of Compute Science Noth Caolina State Univesity Abstact Use constaints such as deadlines ae impotant equiements that ae not consideed by existing cloud-based data pocessing envionments such as Hadoop. In the cuent implementation, jobs ae scheduled in FIFO ode by default with options fo othe pioity based schedules. In this pape, we extend eal time cluste scheduling appoach to account fo the two-phase computation style of MapReduce. We develop citeia fo scheduling jobs based on use specified deadline constaints and discuss ou implementation and peliminay evaluation of a Deadline Constaint Schedule fo Hadoop which ensues that only jobs whose deadlines can be met ae scheduled fo execution. I. INTRODUCTION Apache Hadoop is an open souce implementation of Google s MapReduce [4] that has gained significant populaity as a platfom fo lage scale data pocessing applications. MapReduce is a paallel data pocessing paadigm tageted at cluste-based computing achitectues. Its advantage is that it allows pogammes to abstact fom the issues of paallelization, scheduling, input patitioning, failove, eplication and focus on designing thei application data flows consisting of filteing and aggegation steps. The MapReduce pogamming model consists of encoding data pocessing in tems of two functions: Map and Reduce. Input data is patitioned into fixed sized blocks and fed into paallel Map tasks which pocess the data chunks and poduce intemediate output as a collection of key-value pai tuples. These tuples ae shuffled acoss diffeent educe nodes based on key values. Each Reduce task pefoms thee steps: copy - the map output is copied to educe nodes, sot - the collected map output is soted based on key values and educe - educe function e.g. aggegation is applied to the data. Vaious effots like Hive [5], Pig[11] offe fiendlie intefaces in the fom of high level quey o dataflow languages a la SQL. This enables uses to encode thei tasks in tems of quey opeatos that ae automatically compiled into Hadoop jobs (MapReduce wokflows) instead of as low-level Map and Reduce functions. The MapReduce achitectue consists of one maste (Jobtacke) and many wokes (Tasktackes). The JobTacke eceives job submitted fom use, beaks it down into map and educe tasks, assigns the tasks to Tasktackes, monitos the pogess of the Tasktackes, and finally when all the tasks ae complete, epots the use about the job completion. Each Tasktacke has a fixed numbe of map and educe task slots that detemine how many map and educe tasks it can un at a time. The Hadoop File System HDFS suppots eliability and fault toleance of MapReduce computation by stoing and eplicating the inputs and outputs of a Hadoop job. Since Hadoop jobs have to shae the cluste esouces, a scheduling policy is used to detemine when a job can execute its tasks. The default scheduling policy of Hadoop is Fist In Fist Out (FIFO). Unde this scheme, the job that was submitted ealie gets pefeence ove jobs submitted late. Recent effots such as Delay Schedule[15], Dynamic Popotional Schedule [13] offe diffeentiated sevice fo Hadoop jobs allowing uses to adjust the pioity levels assigned to thei jobs. Howeve, this does not guaantee that the job will be completed by a specific deadline. [12] comes close to addessing the issue of deadlines but focuses moe on inceasing system utilization. [14], [9] focus on suppoting deadline constaints in taditional paallel computatioodels which diffe fom the two-phase computation and unique dataflow of MapReduce jobs. [1] consides deadline constaints in the context of eal time tansactions in single pocesso envionments. In this pape, we lay the foundation fo dealing with deadline equiements in Hadoop-based data pocessing by (1) poposing a job execution cost model that accounts fo the vaious paametes that affect Hadoop job completion time such as map and educe untimes, map and educe input data sizes, data distibution, etc., (2) pesenting the design of a Constaint-Based Hadoop Schedule that takes use deadlines as pat of its input and detemines the schedulability of a job based on the poposed job execution cost model and does so independent of the numbe of jobs unning in the cluste. Jobs ae only scheduled if specified deadlines can be met. We focus on deadline constaints when MapReduce untime paamete values ae known and leave the issue of estimating job paamete values as futue wok. The est of the pape is oganized as follows: in section II, we discuss the scheduling aspects of the poblem and deive expession fo minimum map/educe task allocation equied to meet deadlines. In section III, we pesent the design and implementation of Constaint Schedule and in section IV, we pesent esults fo task allocation fo diffeent deadlines. A. Poblem Definition II. FOUNDATIONS Poblem Statement: Can a given quey q that tanslates to a MapReduce job J and has to pocess data of size σ be completed within a deadline D, when un in a MapReduce cluste having N nodes with N m map task slots, N educe task slots and possibly k jobs executing at the time. Executing quey q in the MapReduce famewok involves scheduling the coesponding map and educe tasks of job J.

2 While it is possible fo q to tanslate to a sequence of MapReduce jobs, in this pape, we focus on queies that tanslate to a single MapReduce job. Afte a job is submitted, the schedule fist needs to detemine whethe the job can be completed within the specified deadline o not using a schedulability test. Rathe thaake schedulability detemination based on all the jobs unning in the system, we focus on the fee slots availability at the given time o in the futue. Afte it is detemined that the job can be completed within the given deadline, it is enlisted fo scheduling. An impotant issue is then to assign the ight numbe of tasks to the Tasktacke to ensue that the deadline is met. Some stategies include: Assign all map and educe tasks: Assigns all the tasks if the numbe of tasks is less than the total available slots. If not, it assigns the tasks to all the available slots in the cluste. This may esult in jobs submitted late not having enough slots to un. Assiginimum tasks: Assigns only minimum numbe of tasks equied fo a job to meet its deadline. Empty slots may be available fo jobs submitted late. Assign some fixed numbe of tasks. Fo the est of pape, minimum numbe of tasks will be used to efe to the minimum numbe of tasks equied fo a job s schedulability independent of task assignment appoach. A job is schedulable if the minimum numbe of tasks fo both map and educe is less than o equal to the available slots. We show the deivation fo the minimum numbe of map and educe task in the following subsections. B. Deadline Estimation Model We develop an initial estimatioodel based a set of assumptions which will be elaxed late in the discussion: (1) the cluste consists of homogeneous nodes, so that the unit cost of pocessing fo each map o educe node is equal; (2) key distibution of the input data is unifom, so that each educe node gets equal amount of educe data to pocess; (3) educe tasks stats afte all map tasks have completed; (4) the input data is aleady available in HDFS. To deive the expessions fo the minimum numbe of map tasks in m and educe tasks in, we extend the model used in[9] fo Equal Load Patitioning technique by intoducing MapReduce specific notations J, f, c m, c, s m and s as descibed below: q = (A, σ, D): A quey q, whee A is the aival time (time when the quey is submitted), σ is the input data size, D is the elative deadline. J = (t m1, t m2, t m3,..., t mu, t 1, t 2,..., t v ): A Hadoop job that is un to pefom quey q. t mi is the i th map task and t j is the j th educe task whee 1 i u and 1 j v. The Hadoop job J has the aival time, input data and deadline same as the quey q = (A, σ, D). n: total slots assigned to the job in the cluste. n = + n whee, is the map slots and n is the educe slots. α = (α 1, α 2,..., α u ): Map data distibution vecto, whee u is the total numbe of map tasks of job J. α i is the data faction allocated to the i th map task. Map input data is equally distibuted among the map nodes so, α i = 1 u. f : Filte atio. The faction of input that the map pocess poduces as output. Fo most pactical puposes 0 f 1 holds. f σ: Reduce input (map output) = poduct of filteatio and the map input. c m : Cost of pocessing a unit data iap task. c : Cost of pocessing a unit data in educe task. c d : Communication cost of tansfeing unit data. s m : stat time of the fist map task fo the job s : stat time of the fist educe task fo the job To estimate the duation of the job J we conside map completion time, educe completion time and data tansfe duing educe copy phase. The expession fo the cost can be witten as: σc m + fσc n + fσc d Since the job has an aival time A and deadline D, Let, s max then, which gives, theefoe, Similaly, s m + σcm + fσc n + fσc d A + D be the maximum value fo the educe stat time, s max in = = A + D fσc n s m + σcm in m = s max σcm s max s m s max fσc d σc m s m fσc A + D fσc d s Ou constaint schedule uses this citeia to schedule Hadoop jobs. 1) Invalidating Assumptions: Assumption 1. We assumed that nodes ae homogeneous and the value of c m and c was unifom fo all the nodes. To elax this assumption we allow fo the value of c m and c to diffe among the nodes making the job dependent on the slowest node i.e. the value of c m and c is the highest. Theefoe, we modify the above expessions fo in m, in and s max by substituting c m and c with the value of the slowest unning node. Assumption 2: When the key distibution acoss the educe tasks is not unifom, then the wost case is that a single node pocesses all the educe input fσ. In such scenaio, the value of s max will be: s max = A + D fσc fσc d We intend to addess in ou futue wok the effect of vaious data distibutioodels. Stating educe tasks ealy might affect othe jobs ability to obtain educe task slots. In (1) (2)

3 this pape, we do not addess the implications of invalidating Assumption 3. We think it is elevant to assume that the input is available in HDFS because MapReduce pocessing stats only afte the data is available. If it becomes impotant to conside the cost of tansfeing the input to HDFS fom use s stoage, the cost can be deived using the values fo tansfe netwok link capacity and the eplication cost, and these costs ae not dependent on the natue of MapReduce job. III. DESIGN AND IMPLEMENTATION Hadoop suppots pluggable schedules and we have implemented Constaint Schedule using the minimum task scheduling citeia developed in the pevious section. It is developed as a contib module using Hadoop vesion souce code. Hadoop config file needs to be modified to use the schedule. We have also implemented a web based inteface that allows the use to specify the deadline fo a given job. A. Design Goals The design goals fo Constaint Schedule wee: (1) To be able to give uses immediate feedback on whethe the job can be completed within the given deadline o not and poceed with execution if deadline can be met. Othewise, uses have the option to esubmit with modified deadline equiements. (2) Maximize the numbe of jobs that can be un in the cluste while satisfying the time equiements of all jobs. When a job is submitted, we pefom its schedulability test simila to that mentioned in [9]. We fist calculate the minimum map tasks fo the submitted job. If the minimum map tasks ae not available at the time when job was submitted based on the total numbe of educe tasks specified fo the job. As mentioned ealie, the numbe of educe tasks fo a job is use defined o is a default value. If the numbe of educe task then the job is ejected. We then calculate s max slots available at s max is not equal to the specified numbe of educe tasks fo the job then also the job is ejected. An altenative to ejecting a job stictly based on s max would be to compute a suitable s value so that the equied numbe of educe tasks slots would be less than the job specified numbe. We intend to exploe this aspect in ou futue wok. Afte a job is scheduled and map tasks ae complete, in is computed to detemine how many educe tasks should be scheduled. Since, s s max, this allows fo cases when in can be less than the educe tasks specified fo the job. This inceases the potential of keeping some educe slots empty. Anothe design goal is to maximize the numbe of jobs while satisfying the deadlines. Thee have been aguments fo and against it when using the minimum tasks scheduling appoach fo multipocesso and cluste computing envionment as mentioned in [8],[9]. Aival ates of the job may play an impotant ole in detemining this. Intuitively, when jobs ae scheduled using only minimum numbe of tasks then it leaves slots empty fo late jobs to execute. We, howeve do not quantitatively evaluate it in this pape. Algoithm 1 scheduletasks(tasktacke t) assignedt asks.initialize() f eemapslots t.f eem apslots f eeeduceslots t.f eereduceslots epeat j nextjob(p ioityqueue) if!j.mapf inished AND j.scheduledm aps < j.minm apt asks then tasks j.getm apt asks(f eemapslots) f eemapslot f eemapslot tasks.size assignedt asks.append(tasks) end if if j.mapf inished AND j.scheduledreduce < j.minreducet asks then tasks j.getreducet asks(f eeeduceslots) f eeeduceslots f eeeduceslots tasks.size assignedt asks.append(tasks) end if until endof Queue(P ioityqueue) OR (feemaptasks == 0 and feeeducetasks == 0) etun assignedt asks B. Task Assignment and Runtime Map and educe tasks untime values ae equied to deive the minimum node citeia. In ou implementation we use static values. A use can supply the map and educe task untime values using the web inteface. In ou futue wok we intend to exploe techniques fo task untime estimation such as cost model fo MapReduce code and estimating cost using a shot sampled job. Recent wok such as Manimal [3] have exploed static code analysis techniques fo MapReduce code. Task assignment is done in eply to the heatbeat obtained fom the Tasktackes. The heatbeat inteval is 3 seconds by default. Constaint Schedule maintains a pioity queue of the jobs odeed by thei deadlines. Tasks scheduling is fist attempted fo jobs that is at the font of pioity queue. If a job s minimum task count is satisfied then we conside next job until thee is no moe jobs emaining in pioity queue o until thee ae no moe unassigned map/educe slots emaining fo the Tasktacke. Thee can be Tasktackes which may not be assigned any tasks if the minimum task count is satisfied fo all the jobs. Algoithm 1 pesents this technique fo scheduling tasks. The scheduling algoithm assigns educe tasks only afte all the map tasks have been completed. This is in accodance with ou assumption in pevious section. IV. EVALUATION To evaluate the Constaint Schedule we an Hadoop job that epesented aggegation opeation, which is one of the common type of opeation pefomed by MapReduce. MapReduce job equivalent to the following quey was used fo the expeiment. SELECT useid, COUNT(actionid) AS num_actions FROM USERACTION; The USERACTION table contains tuples (useid,actionid). The map task of the job

4 pases the input and outputs (useid, actionid) as key-value pais. The educe task then counts the numbe of actionid fo each useid. The filteatio fo this MapReduce job is 1. The geneated data used fo the expeiment had unifom distibution acoss the key - useid, esulting in educes getting equal amount of data to pocess. We have not consideed map and educe task statup ovehead in ou expeiments. We also do not conside the failue ate of the cluste in the expeiments. Data tansfe time was estimated using ipef [6] by pefoming communication between all Jobtacke and Tasktackes. We evaluate the task assignment behavio of Constaint Schedule fo diffeent deadlines. Ou obsevations ae descibed below. A. Expeimental setup Expeiments wee conducted in two diffeent cluste envionments: vitualized cluste and physical cluste. The vitualized cluste consisted of a single physical node with 3 guests as Tasktackes and the host system as Jobtacke. The machine had 4 GB memoy with 64 bit Intel dual coe 2.93GHz pocesso and Ubuntu seve pesent in each vitualized nodes. The physical cluste was a 11 node cluste eseved using VCL [7] whee 10 wee Tasktackes and 1 was Jobtacke. Each node specification was: 4GB maiemoy, Intel 2.33 GHz pocesso unning Redhat Entepise Linux. Hadoop vesion was used in the vitualized cluste and vesion was used in the physical cluste. Both the envionments had Tasktackes with 2 map slots and 2 educe slots. Othe than changing the schedule, all othe configuation paametes wee default values. The HDFS block size was 64 MB. The measued netwok capacity was 12MB/s. B. Results In the vitualized nodes, we obseved that the task execution time vaied depending upon how many nodes wee cuently executing the map tasks. This was due to cpu shaing between the vitualized nodes. Vitualization is one of the scenaios when heteogeneity exists among nodes. We accounted fo heteogeneity in poblem fomulation section by using the execution time of the slowest node in ou minimum task citeia. We obseved the task execution time vaying fom 37 seconds to 120 seconds. We selected the lagest execution time as ou estimate. Figue 1 shows the map task allocation esults fo the same job when submitted with diffeent deadlines. The input size was 975MB which esulted in 16 map tasks. The cluste capacity was 6 map tasks and 6 educe tasks. We obseve that the tasks allocation is done diffeently fo the two deadlines. Fo 600s deadline, 6 map tasks is scheduled at a time. Fo 700s deadline, 5 map tasks ae scheduled. The numbe of map tasks scheduled is the minimum tasks value deived by the schedule. Inteestingly, we also notice that the map tasks fo deadline 700s finishes ealie than the map tasks fo deadline 600s. This is due to the inceased map completion time fo deadline 600s, as it ties to schedule moe map tasks and that esults in inceased pe map completion time though it is within the initial estimate we povide. The incease in pe map #tasks d=600s map count d=700s map count time(ms) Figue 1. Task allocation fo jobs with deadlines 600s and 700s in cluste with vitualized hosts #tasks d=1000s map count d=1000s educe count d=680s map count d=680s educe count e+06 time(ms) Figue 2. Map and educe task allocation fo jobs with deadlines 1000s and 680s in cluste with physical hosts completion time is due to the cpu shaing among vitualized nodes as descibed peviously. Job with deadline 700s ties to schedule less map tasks which esults in bette map completion time and hence it finishes faste. The above esult was obtained by executing the jobs when the cluste had no othe unning jobs. The tansient spikes seen in Figue 1 coespond to the time inteval duing which a task has finished but the task s assigned Tasktacke o any othe Tasktacke with empty slot has not sent heatbeat equest. Similaly, Figue 2 shows the task allocation esults fo the same job unning on the physical cluste with deadlines of 1000s and 680s. The input size was 2.9 GB which esulted in 48 map tasks. The cluste capacity was 20 map tasks and 20 educe tasks. Fo 680s deadline, 20 map tasks and 5 educe tasks ae scheduled. Fo 1000s deadline, 8 map tasks and 4 educe tasks ae scheduled. In both the cases, the deadlines ae met and Constaint Schedule ensues that the minimum task count is met duing the entie job execution. To meet the minimum task citeia, a task may be assigned to Tasktacke even though the data is non local to it. We have not accounted this in ou minimum tasks citeia. Since,

5 in these expeiments the task count is less and also that we have used the obseved wost case computation time, the data tansfe cost due to non local computation seems to be hidden. Howeve, it s impact in scenaios involving lage numbe of tasks emains to be investigated. Also, ou basic model includes the cost of copy phase of educes but does not sepaately account fo sot phase. Duing the expeiment, we have assumed that the educe cost also includes the sot cost. V. RELATED WORK MapReduce scheduling and time estimation: Until ecently, options fo scheduling tasks in Hadoop wee limited to the default FIFO schedule, FaiSchedule [15] and Capacity Schedule. Latte two and ecent eseach effot Dynamic Popotional Schedule [13] povide moe job shaing and pioitization capability in scheduling esulting in inceased shaing of cluste esouces and moe diffeentiation in sevice levels of diffeent jobs. Time estimation and optimization fo Hadoop jobs has been exploed by [2],[10]. [2] focuses on minimizing the total completion time of a set of MapReduce jobs. [10] estimates the pogess of queies that un as MapReduce DAGs. Most effots on scheduling focus on handling vaious pioity and most time estimation effots ae focused on untime estimation of aleady unning jobs. The closest effot to ou wok is [12] which poposes a schedule that ties to incease system esouce utilization by attempting to hono time constaints simila to a deadline schedule. Howeve, they do not conside the schedulability of a job pio to accepting it fo execution. Also, thei wok emphasizes map tasks and does not model the educe computation. Ou wok attempts to cove these aspects and puts moe emphasis oeeting deadlines in the shaed cluste envionment. Scheduling database tansactions: In [1], the authos evaluate tansactions that have eal time constaints fo single pocesso memoy esident database systems. The authos use estimated execution time fo detemining the Feasible Deadline which detemines whethe a deadline will be met o not. Fo scheduling, they use Ealie Deadline Fist (EDF) and Least Slack method. Ou wok builds aound obtaining the execution estimate (in ou case estimate of map and educe computation). Scheduling unde constaints: A genetic algoithm fo task assignment in gid-based envionments is used in [14]. The scheduling poblem hee is the mapping of tasks onto a suitable sevice level to minimize the execution time of a wokflow and complete it within a given budget. [9] exploes the scheduling of divisible eal time tasks in a cluste envionment. A divisible task efes to a task that can be divided into multiple independent subtasks each of which will pocess a piece of input data. Its data patitioning using Equal Patitioning Rule and task/subtask concept can be consideed to be analogous to Hadoop. The authos deive an expession fo the minimum numbe of nodes equied to meet a deadline. Though similaities with ou wok exist in task assignments and data patitioning, thei wok assumes a single type of computation wheeas Hadoop has two types of computations: map and educe. Map input data size can be detemined when a job is submitted but the educe input size is not known without consideing the key distibution. Also, map computations ae almost unifom in homogeneous envionment but educe computation vay depending upon the size of input a educe task eceives. Consequently, ou wok extends thei basic deadline detemination citeia to account fo the MapReduce style of computation. VI. CONCLUSION In this pape, we extended the eal time cluste scheduling appoach to deive minimum map and educe task count citeia fo pefoming task scheduling with deadline constaints in Hadoop. We pesented the design and implementation of Constaint Schedule fo Hadoop following the poposed appoach. Ou esults show that when deadlines fo job is diffeent, then the schedule assigns diffeent numbe of tasks to Tasktacke and makes sue that the specified deadline is met. In this wok, we have left out aspects of Constaint Schedule such as map/educe task untime estimation, filteatio estimation, data distibution and multiple MapReduce cycle suppot. We plan to addess these topics in ou futue wok. REFERENCES [1] Robet K. Abbott and Hecto Gacia-Molina. Scheduling eal-time tansactions: a pefomance evaluation. ACM Tans. Database Syst., 17(3): , [2] Ashaf Aboulnaga, Ziyu Wang, and Zi Ye Zhang. Packing the most onto you cloud. In CloudDB 09: Poceeding of the fist intenational wokshop on Cloud data management, pages 25 28, New Yok, NY, USA, ACM. [3] Michael J. Cafaella and Chistophe Ré. Manimal: Relational optimization fo data-intensive pogams. In WebDB, [4] Jeffey Dean and Sanjay Ghemawat. Mapeduce: simplified data pocessing on lage clustes. In OSDI 04: Poceedings of the 6th confeence on Symposium on Opeating Systems Design & Implementation, pages 10 10, Bekeley, CA, USA, USENIX Association. [5] Hive. [6] Ipef. [7] Vitual Computing Lab. [8] Wan Yeon Lee, Sung Je Hong, and Jong Kim. On-line scheduling of scalable eal-time tasks oultipocesso systems. J. Paallel Distib. Comput., 63(12): , [9] Xuan Lin, Ying Lu, J. Deogun, and S. Goddad. Real-time divisible load scheduling fo cluste computing. In Real Time and Embedded Technology and Applications Symposium, RTAS th IEEE, pages , [10] Kisti Moton, Magdalena Balazinska, and Dan Gossman. Paatime: a pogess indicato fo mapeduce dags. In SIGMOD 10: Poceedings of the 2010 intenational confeence on Management of data, pages , New Yok, NY, USA, ACM. [11] Pig. [12] Joda Polo, David Caea, Yolanda Becea, Malgozata Steinde, and Ian Whalley. Pefomance-diven task co-scheduling fo mapeduce envionments. In Netwok Opeations and Management Symposium (NOMS), 2010 IEEE, pages , [13] Thomas Sandholm and Kevin Lai. Dynamic popotional shae scheduling in hadoop. In JSSPP 10: 15th Wokshop on Job Scheduling Stategies fo Paallel Pocessing, [14] Jia Yu and Rajkuma Buyya. A budget constained scheduling of wokflow applications on utility gids using genetic algoithms. In Wokshop on Wokflows in Suppot of Lage-Scale Science, Poceedings of the 15th IEEE Intenational Symposium on High Pefomance Distibuted Computing (HPDC. IEEE, IEEE CS Pess, [15] Matei Zahaia, Dhuba Bothaku, Joydeep Sen Sama, Khaled Elmeleegy, Scott Shenke, and Ion Stoica. Delay scheduling: a simple technique fo achieving locality and fainess in cluste scheduling. In EuoSys 10: Poceedings of the 5th Euopean confeence on Compute systems, pages , New Yok, NY, USA, ACM.

An Approach to Optimized Resource Allocation for Cloud Simulation Platform

An Approach to Optimized Resource Allocation for Cloud Simulation Platform An Appoach to Optimized Resouce Allocation fo Cloud Simulation Platfom Haitao Yuan 1, Jing Bi 2, Bo Hu Li 1,3, Xudong Chai 3 1 School of Automation Science and Electical Engineeing, Beihang Univesity,

More information

Software Engineering and Development

Software Engineering and Development I T H E A 67 Softwae Engineeing and Development SOFTWARE DEVELOPMENT PROCESS DYNAMICS MODELING AS STATE MACHINE Leonid Lyubchyk, Vasyl Soloshchuk Abstact: Softwae development pocess modeling is gaining

More information

Cloud Service Reliability: Modeling and Analysis

Cloud Service Reliability: Modeling and Analysis Cloud Sevice eliability: Modeling and Analysis Yuan-Shun Dai * a c, Bo Yang b, Jack Dongaa a, Gewei Zhang c a Innovative Computing Laboatoy, Depatment of Electical Engineeing & Compute Science, Univesity

More information

Chapter 3 Savings, Present Value and Ricardian Equivalence

Chapter 3 Savings, Present Value and Ricardian Equivalence Chapte 3 Savings, Pesent Value and Ricadian Equivalence Chapte Oveview In the pevious chapte we studied the decision of households to supply hous to the labo maket. This decision was a static decision,

More information

HEALTHCARE INTEGRATION BASED ON CLOUD COMPUTING

HEALTHCARE INTEGRATION BASED ON CLOUD COMPUTING U.P.B. Sci. Bull., Seies C, Vol. 77, Iss. 2, 2015 ISSN 2286-3540 HEALTHCARE INTEGRATION BASED ON CLOUD COMPUTING Roxana MARCU 1, Dan POPESCU 2, Iulian DANILĂ 3 A high numbe of infomation systems ae available

More information

Distributed Computing and Big Data: Hadoop and MapReduce

Distributed Computing and Big Data: Hadoop and MapReduce Distibuted Computing and Big Data: Hadoop and Map Bill Keenan, Diecto Tey Heinze, Achitect Thomson Reutes Reseach & Development Agenda R&D Oveview Hadoop and Map Oveview Use Case: Clusteing Legal Documents

More information

Efficient Redundancy Techniques for Latency Reduction in Cloud Systems

Efficient Redundancy Techniques for Latency Reduction in Cloud Systems Efficient Redundancy Techniques fo Latency Reduction in Cloud Systems 1 Gaui Joshi, Emina Soljanin, and Gegoy Wonell Abstact In cloud computing systems, assigning a task to multiple seves and waiting fo

More information

An Efficient Group Key Agreement Protocol for Ad hoc Networks

An Efficient Group Key Agreement Protocol for Ad hoc Networks An Efficient Goup Key Ageement Potocol fo Ad hoc Netwoks Daniel Augot, Raghav haska, Valéie Issany and Daniele Sacchetti INRIA Rocquencout 78153 Le Chesnay Fance {Daniel.Augot, Raghav.haska, Valéie.Issany,

More information

Modeling and Verifying a Price Model for Congestion Control in Computer Networks Using PROMELA/SPIN

Modeling and Verifying a Price Model for Congestion Control in Computer Networks Using PROMELA/SPIN Modeling and Veifying a Pice Model fo Congestion Contol in Compute Netwoks Using PROMELA/SPIN Clement Yuen and Wei Tjioe Depatment of Compute Science Univesity of Toonto 1 King s College Road, Toonto,

More information

Effect of Contention Window on the Performance of IEEE 802.11 WLANs

Effect of Contention Window on the Performance of IEEE 802.11 WLANs Effect of Contention Window on the Pefomance of IEEE 82.11 WLANs Yunli Chen and Dhama P. Agawal Cente fo Distibuted and Mobile Computing, Depatment of ECECS Univesity of Cincinnati, OH 45221-3 {ychen,

More information

Questions & Answers Chapter 10 Software Reliability Prediction, Allocation and Demonstration Testing

Questions & Answers Chapter 10 Software Reliability Prediction, Allocation and Demonstration Testing M13914 Questions & Answes Chapte 10 Softwae Reliability Pediction, Allocation and Demonstation Testing 1. Homewok: How to deive the fomula of failue ate estimate. λ = χ α,+ t When the failue times follow

More information

ON THE (Q, R) POLICY IN PRODUCTION-INVENTORY SYSTEMS

ON THE (Q, R) POLICY IN PRODUCTION-INVENTORY SYSTEMS ON THE R POLICY IN PRODUCTION-INVENTORY SYSTEMS Saifallah Benjaafa and Joon-Seok Kim Depatment of Mechanical Engineeing Univesity of Minnesota Minneapolis MN 55455 Abstact We conside a poduction-inventoy

More information

Memory-Aware Sizing for In-Memory Databases

Memory-Aware Sizing for In-Memory Databases Memoy-Awae Sizing fo In-Memoy Databases Kasten Molka, Giuliano Casale, Thomas Molka, Laua Mooe Depatment of Computing, Impeial College London, United Kingdom {k.molka3, g.casale}@impeial.ac.uk SAP HANA

More information

ENABLING INFORMATION GATHERING PATTERNS FOR EMERGENCY RESPONSE WITH THE OPENKNOWLEDGE SYSTEM

ENABLING INFORMATION GATHERING PATTERNS FOR EMERGENCY RESPONSE WITH THE OPENKNOWLEDGE SYSTEM Computing and Infomatics, Vol. 29, 2010, 537 555 ENABLING INFORMATION GATHERING PATTERNS FOR EMERGENCY RESPONSE WITH THE OPENKNOWLEDGE SYSTEM Gaia Tecaichi, Veonica Rizzi, Mauizio Machese Depatment of

More information

Optimal Peer Selection in a Free-Market Peer-Resource Economy

Optimal Peer Selection in a Free-Market Peer-Resource Economy Optimal Pee Selection in a Fee-Maket Pee-Resouce Economy Micah Adle, Rakesh Kuma, Keith Ross, Dan Rubenstein, David Tune and David D Yao Dept of Compute Science Univesity of Massachusetts Amhest, MA; Email:

More information

Ilona V. Tregub, ScD., Professor

Ilona V. Tregub, ScD., Professor Investment Potfolio Fomation fo the Pension Fund of Russia Ilona V. egub, ScD., Pofesso Mathematical Modeling of Economic Pocesses Depatment he Financial Univesity unde the Govenment of the Russian Fedeation

More information

Optimizing Content Retrieval Delay for LT-based Distributed Cloud Storage Systems

Optimizing Content Retrieval Delay for LT-based Distributed Cloud Storage Systems Optimizing Content Retieval Delay fo LT-based Distibuted Cloud Stoage Systems Haifeng Lu, Chuan Heng Foh, Yonggang Wen, and Jianfei Cai School of Compute Engineeing, Nanyang Technological Univesity, Singapoe

More information

Energy Efficient Cache Invalidation in a Mobile Environment

Energy Efficient Cache Invalidation in a Mobile Environment Enegy Efficient Cache Invalidation in a Mobile Envionment Naottam Chand, Ramesh Chanda Joshi, Manoj Misa Electonics & Compute Engineeing Depatment Indian Institute of Technology, Rookee - 247 667. INDIA

More information

THE DISTRIBUTED LOCATION RESOLUTION PROBLEM AND ITS EFFICIENT SOLUTION

THE DISTRIBUTED LOCATION RESOLUTION PROBLEM AND ITS EFFICIENT SOLUTION IADIS Intenational Confeence Applied Computing 2006 THE DISTRIBUTED LOCATION RESOLUTION PROBLEM AND ITS EFFICIENT SOLUTION Jög Roth Univesity of Hagen 58084 Hagen, Gemany Joeg.Roth@Fenuni-hagen.de ABSTRACT

More information

Concept and Experiences on using a Wiki-based System for Software-related Seminar Papers

Concept and Experiences on using a Wiki-based System for Software-related Seminar Papers Concept and Expeiences on using a Wiki-based System fo Softwae-elated Semina Papes Dominik Fanke and Stefan Kowalewski RWTH Aachen Univesity, 52074 Aachen, Gemany, {fanke, kowalewski}@embedded.wth-aachen.de,

More information

Towards Automatic Update of Access Control Policy

Towards Automatic Update of Access Control Policy Towads Automatic Update of Access Contol Policy Jinwei Hu, Yan Zhang, and Ruixuan Li Intelligent Systems Laboatoy, School of Computing and Mathematics Univesity of Westen Sydney, Sydney 1797, Austalia

More information

Reduced Pattern Training Based on Task Decomposition Using Pattern Distributor

Reduced Pattern Training Based on Task Decomposition Using Pattern Distributor > PNN05-P762 < Reduced Patten Taining Based on Task Decomposition Using Patten Distibuto Sheng-Uei Guan, Chunyu Bao, and TseNgee Neo Abstact Task Decomposition with Patten Distibuto (PD) is a new task

More information

Research on Risk Assessment of the Transformer Based on Life Cycle Cost

Research on Risk Assessment of the Transformer Based on Life Cycle Cost ntenational Jounal of Smat Gid and lean Enegy eseach on isk Assessment of the Tansfome Based on Life ycle ost Hui Zhou a, Guowei Wu a, Weiwei Pan a, Yunhe Hou b, hong Wang b * a Zhejiang Electic Powe opoation,

More information

High Availability Replication Strategy for Deduplication Storage System

High Availability Replication Strategy for Deduplication Storage System Zhengda Zhou, Jingli Zhou College of Compute Science and Technology, Huazhong Univesity of Science and Technology, *, zhouzd@smail.hust.edu.cn jlzhou@mail.hust.edu.cn Abstact As the amount of digital data

More information

who supply the system vectors for their JVM products. 1 HBench:Java will work best with support from JVM vendors

who supply the system vectors for their JVM products. 1 HBench:Java will work best with support from JVM vendors Appeaed in the ACM Java Gande 2000 Confeence, San Fancisco, Califonia, June 3-5, 2000 HBench:Java: An Application-Specific Benchmaking Famewok fo Java Vitual Machines Xiaolan Zhang Mago Seltze Division

More information

AN IMPLEMENTATION OF BINARY AND FLOATING POINT CHROMOSOME REPRESENTATION IN GENETIC ALGORITHM

AN IMPLEMENTATION OF BINARY AND FLOATING POINT CHROMOSOME REPRESENTATION IN GENETIC ALGORITHM AN IMPLEMENTATION OF BINARY AND FLOATING POINT CHROMOSOME REPRESENTATION IN GENETIC ALGORITHM Main Golub Faculty of Electical Engineeing and Computing, Univesity of Zageb Depatment of Electonics, Micoelectonics,

More information

Comparing Availability of Various Rack Power Redundancy Configurations

Comparing Availability of Various Rack Power Redundancy Configurations Compaing Availability of Vaious Rack Powe Redundancy Configuations By Victo Avela White Pape #48 Executive Summay Tansfe switches and dual-path powe distibution to IT equipment ae used to enhance the availability

More information

A framework for the selection of enterprise resource planning (ERP) system based on fuzzy decision making methods

A framework for the selection of enterprise resource planning (ERP) system based on fuzzy decision making methods A famewok fo the selection of entepise esouce planning (ERP) system based on fuzzy decision making methods Omid Golshan Tafti M.s student in Industial Management, Univesity of Yazd Omidgolshan87@yahoo.com

More information

Evaluating the impact of Blade Server and Virtualization Software Technologies on the RIT Datacenter

Evaluating the impact of Blade Server and Virtualization Software Technologies on the RIT Datacenter Evaluating the impact of and Vitualization Softwae Technologies on the RIT Datacente Chistophe M Butle Vitual Infastuctue Administato Rocheste Institute of Technology s Datacente Contact: chis.butle@it.edu

More information

STUDENT RESPONSE TO ANNUITY FORMULA DERIVATION

STUDENT RESPONSE TO ANNUITY FORMULA DERIVATION Page 1 STUDENT RESPONSE TO ANNUITY FORMULA DERIVATION C. Alan Blaylock, Hendeson State Univesity ABSTRACT This pape pesents an intuitive appoach to deiving annuity fomulas fo classoom use and attempts

More information

Peer-to-Peer File Sharing Game using Correlated Equilibrium

Peer-to-Peer File Sharing Game using Correlated Equilibrium Pee-to-Pee File Shaing Game using Coelated Equilibium Beibei Wang, Zhu Han, and K. J. Ray Liu Depatment of Electical and Compute Engineeing and Institute fo Systems Reseach, Univesity of Mayland, College

More information

An Analysis of Manufacturer Benefits under Vendor Managed Systems

An Analysis of Manufacturer Benefits under Vendor Managed Systems An Analysis of Manufactue Benefits unde Vendo Managed Systems Seçil Savaşaneil Depatment of Industial Engineeing, Middle East Technical Univesity, 06531, Ankaa, TURKEY secil@ie.metu.edu.t Nesim Ekip 1

More information

Self-Adaptive and Resource-Efficient SLA Enactment for Cloud Computing Infrastructures

Self-Adaptive and Resource-Efficient SLA Enactment for Cloud Computing Infrastructures 2012 IEEE Fifth Intenational Confeence on Cloud Computing Self-Adaptive and Resouce-Efficient SLA Enactment fo Cloud Computing Infastuctues Michael Maue, Ivona Bandic Distibuted Systems Goup Vienna Univesity

More information

The transport performance evaluation system building of logistics enterprises

The transport performance evaluation system building of logistics enterprises Jounal of Industial Engineeing and Management JIEM, 213 6(4): 194-114 Online ISSN: 213-953 Pint ISSN: 213-8423 http://dx.doi.og/1.3926/jiem.784 The tanspot pefomance evaluation system building of logistics

More information

Over-encryption: Management of Access Control Evolution on Outsourced Data

Over-encryption: Management of Access Control Evolution on Outsourced Data Ove-encyption: Management of Access Contol Evolution on Outsouced Data Sabina De Capitani di Vimecati DTI - Univesità di Milano 26013 Cema - Italy decapita@dti.unimi.it Stefano Paaboschi DIIMM - Univesità

More information

Uncertain Version Control in Open Collaborative Editing of Tree-Structured Documents

Uncertain Version Control in Open Collaborative Editing of Tree-Structured Documents Uncetain Vesion Contol in Open Collaboative Editing of Tee-Stuctued Documents M. Lamine Ba Institut Mines Télécom; Télécom PaisTech; LTCI Pais, Fance mouhamadou.ba@ telecom-paistech.f Talel Abdessalem

More information

Adaptive Queue Management with Restraint on Non-Responsive Flows

Adaptive Queue Management with Restraint on Non-Responsive Flows Adaptive Queue Management wi Restaint on Non-Responsive Flows Lan Li and Gyungho Lee Depatment of Electical and Compute Engineeing Univesity of Illinois at Chicago 85 S. Mogan Steet Chicago, IL 667 {lli,

More information

Automatic Testing of Neighbor Discovery Protocol Based on FSM and TTCN*

Automatic Testing of Neighbor Discovery Protocol Based on FSM and TTCN* Automatic Testing of Neighbo Discovey Potocol Based on FSM and TTCN* Zhiliang Wang, Xia Yin, Haibin Wang, and Jianping Wu Depatment of Compute Science, Tsinghua Univesity Beijing, P. R. China, 100084 Email:

More information

How to create RAID 1 mirroring with a hard disk that already has data or an operating system on it

How to create RAID 1 mirroring with a hard disk that already has data or an operating system on it AnswesThatWok TM How to set up a RAID1 mio with a dive which aleady has Windows installed How to ceate RAID 1 mioing with a had disk that aleady has data o an opeating system on it Date Company PC / Seve

More information

Comparing Availability of Various Rack Power Redundancy Configurations

Comparing Availability of Various Rack Power Redundancy Configurations Compaing Availability of Vaious Rack Powe Redundancy Configuations White Pape 48 Revision by Victo Avela > Executive summay Tansfe switches and dual-path powe distibution to IT equipment ae used to enhance

More information

Tracking/Fusion and Deghosting with Doppler Frequency from Two Passive Acoustic Sensors

Tracking/Fusion and Deghosting with Doppler Frequency from Two Passive Acoustic Sensors Tacking/Fusion and Deghosting with Dopple Fequency fom Two Passive Acoustic Sensos Rong Yang, Gee Wah Ng DSO National Laboatoies 2 Science Pak Dive Singapoe 11823 Emails: yong@dso.og.sg, ngeewah@dso.og.sg

More information

Towards Realizing a Low Cost and Highly Available Datacenter Power Infrastructure

Towards Realizing a Low Cost and Highly Available Datacenter Power Infrastructure Towads Realizing a Low Cost and Highly Available Datacente Powe Infastuctue Siam Govindan, Di Wang, Lydia Chen, Anand Sivasubamaniam, and Bhuvan Ugaonka The Pennsylvania State Univesity. IBM Reseach Zuich

More information

The Role of Gravity in Orbital Motion

The Role of Gravity in Orbital Motion ! The Role of Gavity in Obital Motion Pat of: Inquiy Science with Datmouth Developed by: Chistophe Caoll, Depatment of Physics & Astonomy, Datmouth College Adapted fom: How Gavity Affects Obits (Ohio State

More information

Multicriteria analysis in telecommunications

Multicriteria analysis in telecommunications Poceedings of the 37th Hawaii Intenational Confeence on System Sciences - 2004 Multiciteia analysis in telecommunications Janusz Ganat and Andze P. Wiezbicki National Institute of Telecommunications Szachowa

More information

Converting knowledge Into Practice

Converting knowledge Into Practice Conveting knowledge Into Pactice Boke Nightmae srs Tend Ride By Vladimi Ribakov Ceato of Pips Caie 20 of June 2010 2 0 1 0 C o p y i g h t s V l a d i m i R i b a k o v 1 Disclaime and Risk Wanings Tading

More information

MATHEMATICAL SIMULATION OF MASS SPECTRUM

MATHEMATICAL SIMULATION OF MASS SPECTRUM MATHEMATICA SIMUATION OF MASS SPECTUM.Beánek, J.Knížek, Z. Pulpán 3, M. Hubálek 4, V. Novák Univesity of South Bohemia, Ceske Budejovice, Chales Univesity, Hadec Kalove, 3 Univesity of Hadec Kalove, Hadec

More information

A Comparative Analysis of Data Center Network Architectures

A Comparative Analysis of Data Center Network Architectures A Compaative Analysis of Data Cente Netwok Achitectues Fan Yao, Jingxin Wu, Guu Venkataamani, Suesh Subamaniam Depatment of Electical and Compute Engineeing, The Geoge Washington Univesity, Washington,

More information

Chapter 2 Valiant Load-Balancing: Building Networks That Can Support All Traffic Matrices

Chapter 2 Valiant Load-Balancing: Building Networks That Can Support All Traffic Matrices Chapte 2 Valiant Load-Balancing: Building etwoks That Can Suppot All Taffic Matices Rui Zhang-Shen Abstact This pape is a bief suvey on how Valiant load-balancing (VLB) can be used to build netwoks that

More information

A formalism of ontology to support a software maintenance knowledge-based system

A formalism of ontology to support a software maintenance knowledge-based system A fomalism of ontology to suppot a softwae maintenance knowledge-based system Alain Apil 1, Jean-Mac Deshanais 1, and Reine Dumke 2 1 École de Technologie Supéieue, 1100 Note-Dame West, Monteal, Canada

More information

An Infrastructure Cost Evaluation of Single- and Multi-Access Networks with Heterogeneous Traffic Density

An Infrastructure Cost Evaluation of Single- and Multi-Access Networks with Heterogeneous Traffic Density An Infastuctue Cost Evaluation of Single- and Multi-Access Netwoks with Heteogeneous Taffic Density Andes Fuuskä and Magnus Almgen Wieless Access Netwoks Eicsson Reseach Kista, Sweden [andes.fuuska, magnus.almgen]@eicsson.com

More information

Things to Remember. r Complete all of the sections on the Retirement Benefit Options form that apply to your request.

Things to Remember. r Complete all of the sections on the Retirement Benefit Options form that apply to your request. Retiement Benefit 1 Things to Remembe Complete all of the sections on the Retiement Benefit fom that apply to you equest. If this is an initial equest, and not a change in a cuent distibution, emembe to

More information

How To Use A Network On A Network With A Powerline (Lan) On A Pcode (Lan On Alan) (Lan For Acedo) (Moe) (Omo) On An Ipo) Or Ipo (

How To Use A Network On A Network With A Powerline (Lan) On A Pcode (Lan On Alan) (Lan For Acedo) (Moe) (Omo) On An Ipo) Or Ipo ( Hubs, Bidges, and Switches Used fo extending LANs in tems of geogaphical coveage, numbe of nodes, administation capabilities, etc. Diffe in egads to: m collision domain isolation m laye at which they opeate

More information

Give me all I pay for Execution Guarantees in Electronic Commerce Payment Processes

Give me all I pay for Execution Guarantees in Electronic Commerce Payment Processes Give me all I pay fo Execution Guaantees in Electonic Commece Payment Pocesses Heiko Schuldt Andei Popovici Hans-Jög Schek Email: Database Reseach Goup Institute of Infomation Systems ETH Zentum, 8092

More information

INITIAL MARGIN CALCULATION ON DERIVATIVE MARKETS OPTION VALUATION FORMULAS

INITIAL MARGIN CALCULATION ON DERIVATIVE MARKETS OPTION VALUATION FORMULAS INITIAL MARGIN CALCULATION ON DERIVATIVE MARKETS OPTION VALUATION FORMULAS Vesion:.0 Date: June 0 Disclaime This document is solely intended as infomation fo cleaing membes and othes who ae inteested in

More information

An Epidemic Model of Mobile Phone Virus

An Epidemic Model of Mobile Phone Virus An Epidemic Model of Mobile Phone Vius Hui Zheng, Dong Li, Zhuo Gao 3 Netwok Reseach Cente, Tsinghua Univesity, P. R. China zh@tsinghua.edu.cn School of Compute Science and Technology, Huazhong Univesity

More information

Data Center Demand Response: Avoiding the Coincident Peak via Workload Shifting and Local Generation

Data Center Demand Response: Avoiding the Coincident Peak via Workload Shifting and Local Generation (213) 1 28 Data Cente Demand Response: Avoiding the Coincident Peak via Wokload Shifting and Local Geneation Zhenhua Liu 1, Adam Wieman 1, Yuan Chen 2, Benjamin Razon 1, Niangjun Chen 1 1 Califonia Institute

More information

How to recover your Exchange 2003/2007 mailboxes and emails if all you have available are your PRIV1.EDB and PRIV1.STM Information Store database

How to recover your Exchange 2003/2007 mailboxes and emails if all you have available are your PRIV1.EDB and PRIV1.STM Information Store database AnswesThatWok TM Recoveing Emails and Mailboxes fom a PRIV1.EDB Exchange 2003 IS database How to ecove you Exchange 2003/2007 mailboxes and emails if all you have available ae you PRIV1.EDB and PRIV1.STM

More information

Financial Derivatives for Computer Network Capacity Markets with Quality-of-Service Guarantees

Financial Derivatives for Computer Network Capacity Markets with Quality-of-Service Guarantees Financial Deivatives fo Compute Netwok Capacity Makets with Quality-of-Sevice Guaantees Pette Pettesson pp@kth.se Febuay 2003 SICS Technical Repot T2003:03 Keywods Netwoking and Intenet Achitectue. Abstact

More information

METHODOLOGICAL APPROACH TO STRATEGIC PERFORMANCE OPTIMIZATION

METHODOLOGICAL APPROACH TO STRATEGIC PERFORMANCE OPTIMIZATION ETHODOOGICA APPOACH TO STATEGIC PEFOANCE OPTIIZATION ao Hell * Stjepan Vidačić ** Željo Gaača *** eceived: 4. 07. 2009 Peliminay communication Accepted: 5. 0. 2009 UDC 65.02.4 This pape pesents a matix

More information

They aim to select the best services that satisfy the user s. other providers infrastructures and utility services to run

They aim to select the best services that satisfy the user s. other providers infrastructures and utility services to run End-to-End Qo Mapping and Aggegation fo electing Cloud evices Raed Kaim, Chen Ding, Ali Mii Depatment of Compute cience Ryeson Univesity, Toonto, Canada 2kaim@yeson.ca, cding@scs.yeson.ca, ali.mii@yeson.ca

More information

Channel selection in e-commerce age: A strategic analysis of co-op advertising models

Channel selection in e-commerce age: A strategic analysis of co-op advertising models Jounal of Industial Engineeing and Management JIEM, 013 6(1):89-103 Online ISSN: 013-0953 Pint ISSN: 013-843 http://dx.doi.og/10.396/jiem.664 Channel selection in e-commece age: A stategic analysis of

More information

Database Management Systems

Database Management Systems Contents Database Management Systems (COP 5725) D. Makus Schneide Depatment of Compute & Infomation Science & Engineeing (CISE) Database Systems Reseach & Development Cente Couse Syllabus 1 Sping 2012

More information

A Two-Step Tabu Search Heuristic for Multi-Period Multi-Site Assignment Problem with Joint Requirement of Multiple Resource Types

A Two-Step Tabu Search Heuristic for Multi-Period Multi-Site Assignment Problem with Joint Requirement of Multiple Resource Types Aticle A Two-Step Tabu Seach Heuistic fo Multi-Peiod Multi-Site Assignment Poblem with Joint Requiement of Multiple Resouce Types Siavit Swangnop and Paveena Chaovalitwongse* Depatment of Industial Engineeing,

More information

An Introduction to Omega

An Introduction to Omega An Intoduction to Omega Con Keating and William F. Shadwick These distibutions have the same mean and vaiance. Ae you indiffeent to thei isk-ewad chaacteistics? The Finance Development Cente 2002 1 Fom

More information

An application of stochastic programming in solving capacity allocation and migration planning problem under uncertainty

An application of stochastic programming in solving capacity allocation and migration planning problem under uncertainty An application of stochastic pogamming in solving capacity allocation and migation planning poblem unde uncetainty Yin-Yann Chen * and Hsiao-Yao Fan Depatment of Industial Management, National Fomosa Univesity,

More information

Mining Relatedness Graphs for Data Integration

Mining Relatedness Graphs for Data Integration Mining Relatedness Gaphs fo Data Integation Jeemy T. Engle (jtengle@indiana.edu) Ying Feng (yingfeng@indiana.edu) Robet L. Goldstone (goldsto@indiana.edu) Indiana Univesity Bloomington, IN. 47405 USA Abstact

More information

Timing Synchronization in High Mobility OFDM Systems

Timing Synchronization in High Mobility OFDM Systems Timing Synchonization in High Mobility OFDM Systems Yasamin Mostofi Depatment of Electical Engineeing Stanfod Univesity Stanfod, CA 94305, USA Email: yasi@wieless.stanfod.edu Donald C. Cox Depatment of

More information

Approximation Algorithms for Data Management in Networks

Approximation Algorithms for Data Management in Networks Appoximation Algoithms fo Data Management in Netwoks Chistof Kick Heinz Nixdof Institute and Depatment of Mathematics & Compute Science adebon Univesity Gemany kueke@upb.de Haald Räcke Heinz Nixdof Institute

More information

Strength Analysis and Optimization Design about the key parts of the Robot

Strength Analysis and Optimization Design about the key parts of the Robot Intenational Jounal of Reseach in Engineeing and Science (IJRES) ISSN (Online): 2320-9364, ISSN (Pint): 2320-9356 www.ijes.og Volume 3 Issue 3 ǁ Mach 2015 ǁ PP.25-29 Stength Analysis and Optimization Design

More information

Development of Mathematical Model for Market-Oriented Cloud Computing

Development of Mathematical Model for Market-Oriented Cloud Computing Intenational Jounal of Compute Applications (0975 8887) Volume 9 No.11, Novembe 2010 Development of Mathematical Model fo Maket-Oiented Cloud Computing K.Mukhejee Depatment of Compute Science & Engineeing.

More information

Load Balancing in Processor Sharing Systems

Load Balancing in Processor Sharing Systems Load Balancing in ocesso Shaing Systems Eitan Altman INRIA Sophia Antipolis 2004, oute des Lucioles 06902 Sophia Antipolis, Fance altman@sophia.inia.f Utzi Ayesta LAAS-CNRS Univesité de Toulouse 7, Avenue

More information

Load Balancing in Processor Sharing Systems

Load Balancing in Processor Sharing Systems Load Balancing in ocesso Shaing Systems Eitan Altman INRIA Sophia Antipolis 2004, oute des Lucioles 06902 Sophia Antipolis, Fance altman@sophia.inia.f Utzi Ayesta LAAS-CNRS Univesité de Toulouse 7, Avenue

More information

Nice Resource Reservations in Linux

Nice Resource Reservations in Linux Nice Resouce Resevations in Linux Matin Ohlin Depatment of Automatic Contol Lund Univesity Box 118, SE-221 Lund, Sweden Email: Matin.Ohlin@contol.lth.se Matin A. Kjæ Depatment of Automatic Contol Lund

More information

SUPPORT VECTOR MACHINE FOR BANDWIDTH ANALYSIS OF SLOTTED MICROSTRIP ANTENNA

SUPPORT VECTOR MACHINE FOR BANDWIDTH ANALYSIS OF SLOTTED MICROSTRIP ANTENNA Intenational Jounal of Compute Science, Systems Engineeing and Infomation Technology, 4(), 20, pp. 67-7 SUPPORT VECTOR MACHIE FOR BADWIDTH AALYSIS OF SLOTTED MICROSTRIP ATEA Venmathi A.R. & Vanitha L.

More information

9:6.4 Sample Questions/Requests for Managing Underwriter Candidates

9:6.4 Sample Questions/Requests for Managing Underwriter Candidates 9:6.4 INITIAL PUBLIC OFFERINGS 9:6.4 Sample Questions/Requests fo Managing Undewite Candidates Recent IPO Expeience Please povide a list of all completed o withdawn IPOs in which you fim has paticipated

More information

The impact of migration on the provision. of UK public services (SRG.10.039.4) Final Report. December 2011

The impact of migration on the provision. of UK public services (SRG.10.039.4) Final Report. December 2011 The impact of migation on the povision of UK public sevices (SRG.10.039.4) Final Repot Decembe 2011 The obustness The obustness of the analysis of the is analysis the esponsibility is the esponsibility

More information

Power Monitoring and Control for Electric Home Appliances Based on Power Line Communication

Power Monitoring and Control for Electric Home Appliances Based on Power Line Communication I²MTC 2008 IEEE Intenational Instumentation and Measuement Technology Confeence Victoia, Vancouve Island, Canada, May 12 15, 2008 Powe Monitoing and Contol fo Electic Home Appliances Based on Powe Line

More information

Financing Terms in the EOQ Model

Financing Terms in the EOQ Model Financing Tems in the EOQ Model Habone W. Stuat, J. Columbia Business School New Yok, NY 1007 hws7@columbia.edu August 6, 004 1 Intoduction This note discusses two tems that ae often omitted fom the standad

More information

est using the formula I = Prt, where I is the interest earned, P is the principal, r is the interest rate, and t is the time in years.

est using the formula I = Prt, where I is the interest earned, P is the principal, r is the interest rate, and t is the time in years. 9.2 Inteest Objectives 1. Undestand the simple inteest fomula. 2. Use the compound inteest fomula to find futue value. 3. Solve the compound inteest fomula fo diffeent unknowns, such as the pesent value,

More information

HIGH AVAILABILITY SOLUTION: RESOURCE USAGE MANAGEMENT IN VIRTUALIZED SOFTWARE AGING

HIGH AVAILABILITY SOLUTION: RESOURCE USAGE MANAGEMENT IN VIRTUALIZED SOFTWARE AGING Intenational Jounal o Compute Science & Inomation Technology (IJCSIT) Vol 4, No, June 0 HIGH AVAILAILITY SOLUTION: ESOUCE USAGE MANAGEMENT IN VITUALIZED SOFTWAE AGING Aye Myat Myat aing and Ni La Thein

More information

Model-Driven Engineering of Adaptation Engines for Self-Adaptive Software: Executable Runtime Megamodels

Model-Driven Engineering of Adaptation Engines for Self-Adaptive Software: Executable Runtime Megamodels Model-Diven Engineeing of Adaptation Engines fo Self-Adaptive Softwae: Executable Runtime Megamodels Thomas Vogel, Holge Giese Technische Beichte N. 66 des Hasso-Plattne-Instituts fü Softwaesystemtechnik

More information

Review Graph based Online Store Review Spammer Detection

Review Graph based Online Store Review Spammer Detection Review Gaph based Online Stoe Review Spamme Detection Guan Wang, Sihong Xie, Bing Liu, Philip S. Yu Univesity of Illinois at Chicago Chicago, USA gwang26@uic.edu sxie6@uic.edu liub@uic.edu psyu@uic.edu

More information

Chris J. Skinner The probability of identification: applying ideas from forensic statistics to disclosure risk assessment

Chris J. Skinner The probability of identification: applying ideas from forensic statistics to disclosure risk assessment Chis J. Skinne The pobability of identification: applying ideas fom foensic statistics to disclosue isk assessment Aticle (Accepted vesion) (Refeeed) Oiginal citation: Skinne, Chis J. (2007) The pobability

More information

Analyzing Ballistic Missile Defense System Effectiveness Based on Functional Dependency Network Analysis

Analyzing Ballistic Missile Defense System Effectiveness Based on Functional Dependency Network Analysis Send Odes fo Repints to epints@benthamscience.ae 678 The Open Cybenetics & Systemics Jounal, 2015, 9, 678-682 Open Access Analyzing Ballistic Missile Defense System Effectiveness Based on Functional Dependency

More information

How To Find The Optimal Stategy For Buying Life Insuance

How To Find The Optimal Stategy For Buying Life Insuance Life Insuance Puchasing to Reach a Bequest Ehan Bayakta Depatment of Mathematics, Univesity of Michigan Ann Abo, Michigan, USA, 48109 S. David Pomislow Depatment of Mathematics, Yok Univesity Toonto, Ontaio,

More information

Dual channel closed-loop supply chain coordination with a reward-driven remanufacturing policy

Dual channel closed-loop supply chain coordination with a reward-driven remanufacturing policy Intenational Jounal of Poduction Reseach ISSN: -753 Pint 1366-588X Online Jounal homepage: http://www.tandfonline.com/loi/tps Dual channel closed-loop supply chain coodination with a ewad-diven emanufactuing

More information

Top K Nearest Keyword Search on Large Graphs

Top K Nearest Keyword Search on Large Graphs Top K Neaest Keywod Seach on Lage Gaphs Miao Qiao, Lu Qin, Hong Cheng, Jeffey Xu Yu, Wentao Tian The Chinese Univesity of Hong Kong, Hong Kong, China {mqiao,lqin,hcheng,yu,wttian}@se.cuhk.edu.hk ABSTRACT

More information

INVESTIGATION OF FLOW INSIDE AN AXIAL-FLOW PUMP OF GV IMP TYPE

INVESTIGATION OF FLOW INSIDE AN AXIAL-FLOW PUMP OF GV IMP TYPE 1 INVESTIGATION OF FLOW INSIDE AN AXIAL-FLOW PUMP OF GV IMP TYPE ANATOLIY A. YEVTUSHENKO 1, ALEXEY N. KOCHEVSKY 1, NATALYA A. FEDOTOVA 1, ALEXANDER Y. SCHELYAEV 2, VLADIMIR N. KONSHIN 2 1 Depatment of

More information

AgentTime: A Distributed Multi-agent Software System for University s Timetabling

AgentTime: A Distributed Multi-agent Software System for University s Timetabling AgentTime: A Distibuted Multi-agent Softwae System fo Univesity s Timetabling Eduad Babkin 1,2, Habib Adbulab 1, Tatiana Babkina 2 1 LITIS laboatoy, INSA de Rouen. Rouen, Fance. 2 State Univesity Highe

More information

Electricity transmission network optimization model of supply and demand the case in Taiwan electricity transmission system

Electricity transmission network optimization model of supply and demand the case in Taiwan electricity transmission system Electicity tansmission netwok optimization model of supply and demand the case in Taiwan electicity tansmission system Miao-Sheng Chen a Chien-Liang Wang b,c, Sheng-Chuan Wang d,e a Taichung Banch Gaduate

More information

arxiv:1110.2612v1 [q-fin.st] 12 Oct 2011

arxiv:1110.2612v1 [q-fin.st] 12 Oct 2011 Maket inefficiency identified by both single and multiple cuency tends T.Toká 1, and D. Hováth 1, 1 Sos Reseach a.s., Stojáenská 3, 040 01 Košice, Slovak Republic Abstact axiv:1110.2612v1 [q-fin.st] 12

More information

Saving and Investing for Early Retirement: A Theoretical Analysis

Saving and Investing for Early Retirement: A Theoretical Analysis Saving and Investing fo Ealy Retiement: A Theoetical Analysis Emmanuel Fahi MIT Stavos Panageas Whaton Fist Vesion: Mach, 23 This Vesion: Januay, 25 E. Fahi: MIT Depatment of Economics, 5 Memoial Dive,

More information

How To Schedule A Cloud Comuting On A Computer (I.E. A Computer)

How To Schedule A Cloud Comuting On A Computer (I.E. A Computer) Intenational Jounal on Cloud Comuting: Sevices and Achitectue(IJCCSA,Vol., No.3,Novembe 20 STOCHASTIC MARKOV MODEL APPROACH FOR EFFICIENT VIRTUAL MACHINES SCHEDULING ON PRIVATE CLOUD Hsu Mon Kyi and Thinn

More information

Questions for Review. By buying bonds This period you save s, next period you get s(1+r)

Questions for Review. By buying bonds This period you save s, next period you get s(1+r) MACROECONOMICS 2006 Week 5 Semina Questions Questions fo Review 1. How do consumes save in the two-peiod model? By buying bonds This peiod you save s, next peiod you get s() 2. What is the slope of a consume

More information

PLANNING THE CAPACITY OF A WEB SERVER: AN EXPERIENCE REPORT

PLANNING THE CAPACITY OF A WEB SERVER: AN EXPERIENCE REPORT PLANNING THE CAPACITY OF A WEB SERVER: AN EXPERIENCE REPORT Daniel A. Menascé Robet Peaino Depatment of Compute Science, MS 4A5 Univesity Computing Geoge Mason Univesity Geoge Mason Univesity Faifax, VA

More information

Magnetic Bearing with Radial Magnetized Permanent Magnets

Magnetic Bearing with Radial Magnetized Permanent Magnets Wold Applied Sciences Jounal 23 (4): 495-499, 2013 ISSN 1818-4952 IDOSI Publications, 2013 DOI: 10.5829/idosi.wasj.2013.23.04.23080 Magnetic eaing with Radial Magnetized Pemanent Magnets Vyacheslav Evgenevich

More information

The Binomial Distribution

The Binomial Distribution The Binomial Distibution A. It would be vey tedious if, evey time we had a slightly diffeent poblem, we had to detemine the pobability distibutions fom scatch. Luckily, thee ae enough similaities between

More information

Office of Family Assistance. Evaluation Resource Guide for Responsible Fatherhood Programs

Office of Family Assistance. Evaluation Resource Guide for Responsible Fatherhood Programs Office of Family Assistance Evaluation Resouce Guide fo Responsible Fathehood Pogams Contents Intoduction........................................................ 4 Backgound..........................................................

More information

Risk Sensitive Portfolio Management With Cox-Ingersoll-Ross Interest Rates: the HJB Equation

Risk Sensitive Portfolio Management With Cox-Ingersoll-Ross Interest Rates: the HJB Equation Risk Sensitive Potfolio Management With Cox-Ingesoll-Ross Inteest Rates: the HJB Equation Tomasz R. Bielecki Depatment of Mathematics, The Notheasten Illinois Univesity 55 Noth St. Louis Avenue, Chicago,

More information

California s Duals Demonstration: A Transparent. Process. Margaret Tatar Chief, Medi-Cal Managed Care Division. CA Coo 8/21/12

California s Duals Demonstration: A Transparent. Process. Margaret Tatar Chief, Medi-Cal Managed Care Division. CA Coo 8/21/12 Califonia s Duals Demonstation: A Tanspaent and Inclusive Stakeholde Pocess Magaet Tata Chief, Medi-Cal Managed Cae Division Depatment of Health Cae Sevices 1 Stakeholde Engagement 1. 2. Inclusive Building

More information