Scheduling Independent Tasks in Heterogeneous Environments under Communication Constraints

Size: px
Start display at page:

Download "Scheduling Independent Tasks in Heterogeneous Environments under Communication Constraints"

Transcription

1 Schedulng Independent Tasks n Heterogeneous Envronments under Communcaton Constrants Petros Lampsas 1 Thanass Loukopoulos 2 Fedon Dmopoulos 1 Marouso Athanasou 1 1 Department of Informatcs and Computer Technology Technologcal Educatonal Insttute of Lama 3rd km Old Natonal Road, Lama, Greece {plam, crrus, maryse}@nf.telam.gr 2 Department of Computer and Communcatons Eng. Unversty of Thessaly Glavan 37, Volos, Greece luke@nf.uth.gr Abstract Wth the advent of the Grd, task schedulng n heterogeneous envronments becomes more and more mportant. Of partcular nterest s the fact that especally n scentfc experments a non neglgble amount of data must be transferred to the processng node before a task can commence executon. Gven bandwdth constrants, schedulng both computatons and data transfers s requred. In ths paper we frst develop a sutable model that captures heterogenety n the processng nodes whle mposng communcaton constrants. We proceed by proposng schedulng heurstcs wth the am of mnmzng the total makespan of a set of ndependent tasks. Through a seres of experments we llustrate the potental of a partcular heurstc that s based on backfllng. 1. Introducton The task schedulng problem s of paramount mportance n the felds of parallel systems and computaton Grds [14]. A generc descrpton of t s as follows: gven a set of tasks and a set of processng nodes, map the tasks to the nodes and order ther executon, n order to mnmze the total tme requred to execute all tasks. An avd lterature exsts on the problem nvestgatng dfferent aspects of t. Typcally, n parallel systems processng nodes are assumed to be homogeneous [4] whle n cluster and Grd computng the focus s usually on heterogeneous nodes [2]. Another dstngushng factor s whether tasks are assumed to have nterdependences or not (Drected Acyclc Graph - DAG schedulng [7]). In ths paper we deal wth schedulng ndependent tasks n a heterogeneous dstrbuted envronment wth the man focus beng computatonal Grds. Our model captures heterogenety n two ways. The frst regards the processng power of each node and the dfferent resource levels avalable at them. The second concerns communcaton capabltes. Contrary to most of the exstng works, we place partcular emphass on the latter aspect snce the data needed to be transferred for task executon n a computatonal Grd are often extremely large. Tradtonal formulatons assume that all data transfers may commence at the same tme, whle the transfer delay s always lnear to the avalable bandwdth and the sze of the transferred data. Ths s less than adequate snce t fals to capture congeston ssues [13]. Instead, we follow an alternatve whereby the number of transfers each node may partcpate at s restrcted. Our contrbutons are threefold. Frst, we develop a schedulng model that ncorporates the man parameters of a dstrbuted heterogeneous envronment,.e., node processng power, resource and communcaton constrants. Second, we adapt n our model two popular lst schedulng heurstcs [2] (MaxMn, MnMn). In dong so, we nvestgate dfferent canddates for the task weght functon. Fnally, based on how MaxMn and MnMn work, we develop a new heurstc called BackflledMaxMn (BMaxMn) that combnes the strengths of both algorthms. The rest of the paper s organzed as follows. Secton 2 presents the schedulng model. Secton 3 llustrates the heurstcs whch are expermentally evaluated n Secton 4. Related work and drectons for future research are dscussed n Secton 5. Fnally, Secton 6 concludes the paper.

2 2. Schedulng model Assume there exsts M ndependent tasks (denoted by T k ) that must be scheduled at the N nodes (let N ) of a dstrbuted system. The executon of a task T k at N requres certan data (let d k be the sze), as well as a certan amount of resources (e.g., processors, memory etc.) to be avalable at N. Let r denote the total number of dstnct resources we are nterested at, R x the level of the xth resource at N and r kx the amount of the xth resource requred by T k (meanng that n case T k commences ts executon at N the remanng resource level R x at N wll be decreased by r kx for each dstnct resource x). Task data orgnate from sources whch are connected wth the processng nodes through a communcaton network. Let S be an N M matrx whereby an element S k = 1 ff N s a source for T k s data and 0 otherwse. In case a task s data source s dfferent from the processng node t s assgned, the correspondng data need to be transferred. We assume that a node N can partcpate n at most K smultaneous transfers. Each of the K transfers s performed at a rate a j (where a j s the tme requred to transfer 1 data unt from N j to N and a = a ). Notce that our approach n modelng transfer delays s smlar to the ones n [1] and [11]. Fgure 1. An example of 4 data transfers. In case both the source N j and the destnaton j j N of a data transfer for T k have not reached ther capactes ( K j and K respectvely), the data transfer wll be completed after a jdk tme. Otherwse, we must wat untl one of the ongong transfers s completed before T k s data transfer can commence. Fg. 1 llustrates the concept assumng a node N must perform 4 transfers, from a source N that has unlmted transfer capacty. Fg. 1(a) shows the case where K = 4 and thus all transfers can commence smultaneously. Fg. 1(b) and 1(c) depct two possble j transfer schedules when K = 2, of whch, the one n Fg. 1(c) s optmal. Once data has arrved at the ntended node, task executon wll start, provded the requred resources are avalable. We assume that an estmaton of the executon tme (let e k ) for each T k s readly avalable. The actual runnng tme of a task executon s dependent on the node where t s assgned and s gven by: e k /speedup( N ), where speedup( N ) depcts how much faster the tasks wll execute at a node N as compared to the base system whereby the estmaton of e k was based upon. Smlar assumptons to capture heterogenety were used n [2], [7], [9], [10] to name a few. Our goal s to schedule the M tasks to the avalable nodes n order to mnmze the total makespan. It can be shown that the relevant decson problem s NPcomplete (by reducton to the two-processor schedulng problem [6]). In the sequel we present varous heurstcs to tackle t. 3. Schedulng heurstcs The heurstcs examned here fall nto two categores: () lst-based [2], () backfll-based [5], [16]. Lst based heurstcs order the tasks accordng to a weght functon and schedule them ether n ncreasng or decreasng order. In ths category we examne the MaxMn and MnMn algorthms. The dea behnd backfllng s to fll gaps n the schedule usng small tasks. We dscuss how to mplement ths premse for the partcular case of MaxMn (BackflledMaxMn heurstc). MaxMn heurstc: In the MaxMn heurstc [2] we frst select the maxmum weght task (Max step) and schedule t to the node where ts completon tme wll be mnmum (Mn step). We proceed by greedly schedulng the remanng tasks n decreasng order of weght. Notce that calculatng the completon tme of a task executon nvolves selectng among (possbly) multple sources to download the relevant data. The source that mnmzes the transfer completon tme s selected, whch s not necessarly the one wth mn a due to the fact that t mght already be occuped wth other transfers. The ratonale behnd MaxMn, s that snce the largest tasks domnate the total makespan, schedulng them at the earlest possble tme wll lkely mnmze the overall completon tme. Decdng task weghts requres further attenton. In other related problem domans, e.g., cluster schedulng [2], task weghts (let W k ) usually equal estmated j

3 executon tmes ( e k s). Thus, weghts are monotone, n the sense that whenever two tasks T k and T k are consdered for schedulng at the same node and Wk W k, T k wll fnsh earler than T k (regardless of the nspected node). However, when data transfers are consdered, fndng a monotonc weght defnton s mpossble, snce transfer tme s dependent on the source-destnaton par. Thus, t mght be the case that T k wll execute faster than T k at a node N, whle at a node N the opposte can happen, f N s closer to T k s sources than to T k s. The followng weght functons are consdered n the paper: W 1 =,.e., the data sze. k d k W 2 k = d k ( S jk ( aj / N)) / S jk,.e., the j j average transfer tme calculated by takng all possble sources and destnatons of T k. W 3 k = e k,.e., the expected executon tme. W 4k = ek /( speedup( N ) / N),.e., the average executon tme. W 5 k = W 2k + W 4k,.e., the average transfer tme together wth the average executon tme. MnMn heurstc: The MnMn heurstc [2] s symmetrc to MaxMn wth the only dfference beng that tasks are scheduled n ncreasng order of weght (.e., smallest s pcked frst). The ratonale behnd MnMn s that by mnmzng the average start tme of tasks the overall makespan wll be mnmal. BackflledMaxMn (BMaxMn) heurstc: Ths heurstc ams at mprovng the performance of MaxMn, through better resource utlzaton. Recall, that MaxMn selects the maxmum weght task and assgns t to the most benefcal node. Regardless of the weght defnton, tasks nvolvng large data transfers wll most lkely be hgh n the rank. Ths means that the early stages wll be devoted n data transfers leavng the processng nodes otherwse dle. BMaxMn attempts to overcome the above shortcomng, by only usng a porton of the K avalable channels for large data transfers and the rest for relatvely small. Followng s the pseudocode of the algorthm: BMaxMn() (1) whle ( unscheduled tasks) (2) schedule a task wth MaxMn; //let T k, N be the task node par decded by MaxMn (3) for all unscheduled tasks T x //backfll phase (4) f (executon of T x at N does not conflct wth T k s executon) (5) schedule T x at N ; The algorthm starts by assgnng a task usng MaxMn,.e., the task of largest weght (let T k ) to the node where t wll be completed n the mnmum tme (let N ). Lnes 3-5 s the backfll phase, whereby the algorthm checks f any of the remanng tasks can be assgned to N wthout resultng n a resource conflct wth T k. In Sec. 4 we experment wth two varants of BMaxMn. The frst pcks the tasks for backfllng startng from the bottom of the ordered lst (.e., smallest frst). We call ths varant BMaxMn-Up (BMaxMn-U). The other works n a reverse manner by startng from the top of the lst and scannng downwards (BMaxMn-Down, BMaxMn-D for short). Fgure 2. An example of 4 task assgnments. Fg. 2 llustrates an example run of the three heurstcs, assumng one processng node N wth K = 2 that can only execute one task at a tme and one source used s N wth K = 4. The weght functon j j W1 k. Fg. 2(a) shows the tme requred for data transfer and executon by each task. Fg 2(b) and 2(c) depct the schedules produced by MaxMn and MnMn. Notce the performance dfference between the two, whch can be attrbuted to the fact that MnMn utlzes the avalable node resources much earler compared to MaxMn (Sec. 4 further elaborates the concept). Fg. 2(d) presents the schedule of BMaxMn-D. The algorthm frst assgns the largest task ( T 1 ) and enters the backfll phase. Snce the lst s scanned top-down, T s frst consdered for backfll. However, the 2 executon of T 2 conflcts wth T 1 (recall that only one task can be executed each tme). Therefore T 2 s not backflled and the algorthm proceeds by checkng T 3. T 3 causes no conflcts and s thus, scheduled. The same holds true wth T 4 whch s scheduled after T 3. The algorthm then ends the backfll phase and returns to lne 2 n the pseudocode, where T 2 s assgned.

4 Notce, that T 2 s executon must be delayed untl after T 1 fnshes. 4. Experments In ths secton we present some expermental results. Sec. 4.1 descrbes the smulaton setup, whle Sec. 4.2 llustrates our fndngs. Due to space lmtatons only a small set of the experments conducted are shown Smulaton setup We assumed a network consstng of 20 processng nodes and 5 source nodes (.e., nodes that only serve as data repostores and perform no computatons). The average transfer rate between the processng nodes and the sources (.e., aj s) vared unformly between 1 and 3 whle the speedup at the processng nodes between 1 and 2. The number of smultaneous transfers a processng node N could partcpate at, was fxed to K = 4, whle for sources t was set to nfnty. Ths was done n order to smulate the case where the avalable bandwdth at the processng nodes s the restrctve factor for performng the data transfers data sze avg transfer tme estmated executon tme avg executon tme avg transfer+executon MnMn MaxMn BMaxMn-D BMaxMn_U Fgure 3. Effects of dfferent weght metrcs on the total makespan. We generated a total of 500 tasks wth the followng propertes: () data sze ( d k ) vared unformly between 500 and 5,000 data unts, () estmated executon tme ( e k ) vared between 1,500 and 15,000 sec. We selected the above values n order to equalze (on the average case) the data transfer tme and the processng tme of a sngle task (notce the way speedups and transfer rates were defned). Unless otherwse stated we assumed that each node N can execute n parallel at most two tasks due to resource lmtatons (.e., Rx s). Fnally, one, randomly allocated, copy of the data requred to execute a task was avalable Results Frst, we evaluated the performance of the algorthms under the dfferent weght functons presented n Sec. 3. Fg. 3 llustrates the results. The frst thng to notce s that the effect of the weght metrc s not dentcal to all algorthms. MaxMn and BMaxMn-D behave best when the weght metrc s strongly related to the executon tme. MnMn on the other hand acheves a better performance when the transfer cost s factored n the weght, whle the total makespan of BMaxMn-U tends to reman unaffected. In the remanng experments we adopted the data sze as the weght metrc x1 x2 x5 x10 Data sze ncrease MnMn MaxMn BMaxMn-D BMaxMn-U Fgure 4. Total makespan as the sze of data transfers ncreases. Next we evaluated the algorthms as the sze of data transfers ncreases. Fg. 4 dsplays the performance as the average transfer tme changes from beng equal to the average executon tme (x1) towards beng 10 tmes more costly (x10). Notce that the makespan ncreases rather slowly compared to the ncrease n data transfer szes. Ths ndcates that the processng tme stll domnates the makespan whch s expected snce only two tasks can execute n parallel, whle each node can fetch the data of 4 tasks smultaneously ( K = 4 ). MaxMn and BMaxMn-U outperform ther counterparts n most cases (wth the latter havng a small edge over the former). Ths s due to the fact that MaxMn prortzes large transfers. However, n dong so t keeps the processng nodes dle at the begnnng of the schedule. MnMn on the other hand tends to utlze the nodes (for processng) early on, wth the drawback beng that large transfers are left towards the end of the schedule and are thus not sent to the best possble nodes. BMaxMn-U combnes the benefts of both approaches wnnng n most cases. Notce also that BMaxMn-D does not acheve performance comparable to BMaxMn-U, due to the fact that fewer

5 tasks are backflled. Smlar observatons can be drawn n Fg. 5 whch demonstrates the total makespan as the estmated runnng tmes of the tasks ncrease up to 10 tmes ther basc value. In ths case MnMn together wth BMaxMn-U offer the best performance x1 x2 x5 x10 Estmated executon tme ncrease MnMn MaxMn BMaxMn-D BMaxMn-U Fgure 5. Total makespan as the estmated executon tme ncreases. In the last experment we vared the number of tasks that can be executed n parallel at each node. Fg. 6 llustrates the performance when the degree of parallelsm at each node (due to resource constrants) ncreases from 1 (.e., no parallel executons) to 4. Notce that BMaxMn-U together wth MnMn offers agan the best makespan Number of tasks that can be executed n parallel MnMn MaxMn BMaxMn-D BMaxMn-U Fgure 6. Total makespan as more tasks can be executed n parallel at each node. Overall, from the experments we can conclude that MaxMn acheves better performance when the workload s data-ntensve, whle MnMn outperforms MaxMn f the workload s computaton-ntensve. BMaxMn-U combnes the strengths of both algorthms outperformng them n most cases, whle the BMaxMn-D varant s nferor. In more extensve expermentatons (not shown due to lack of space) we further checked the above conclusons and found that although n most cases they are true, whenever the setup was prmarly bandwdth constraned (and not resource constraned as the one here) the benefts from BMaxMn-U tend to dmnsh. In future work we plan to nvestgate other varants of BMaxMn that do not use all avalable channels for backfll but rather follow a more judcous approach. 5. Related work Task schedulng n heterogeneous envronments has seen sgnfcant research effort (see [4] and [7] for surveys). In [2] the authors evaluated 11 heurstcs for schedulng ndependent tasks requrng no data transfers. MnMn together wth a Genetc Algorthm (GA) were shown to acheve the best performance. For the case of nteractng tasks a heurstc based on the cross-entropy method called MaTCH was proposed n [12] and was found to outperform GA. Mappng was also the goal of [8] where the authors propose a Web servce lke archtecture to dynamcally map ndependent jobs to more than one type of resources. Varous papers focused on data ntensve applcatons, e.g., [3], [9] and [15] to name a few. In [3] an nteger programmng approxmaton was expermentally proven to outperform a greedy algorthm (resemblng MaxMn) whenever data transfers were constraned by the avalable storage space. In [9] a GA was found to outperform greedy algorthms based on FIFO schedulng. For the case where multple data fles must be acqured from varous sources before a task can commence executon, a heurstc based on the set coverng problem that prortzes transfers over computatons acheves a good trade-off between runnng tme and soluton qualty as shown n [15]. The aforementoned works are perhaps the closest to ours snce they too, am at schedulng both computatons and data transfers. However, we dffer sgnfcantly both n the adopted model (e.g., [3] assumes that data transfer delay s due to processng capacty at sources, whle [9] and [15] do not consder any resource constrants on the processng nodes) and n the heurstcs snce we dentfy the potental of schedulng a mx of large and small data transfers and realze t by combnng the backfllng technque [5], [16] wth MaxMn. Two other alternatves to schedulng data and computatons are descrbed n [1] and [11]. In [11] data transfers are completely decoupled from the schedulng process and are performed n a preemptve manner accordng to dataset popularty, whle n [1] the authors provde optmal solutons to dfferent varatons of the schedulng problem under the assumpton that tasks can

6 be arbtrarly dvded. Our work dffers n scope from both these works. As future research we plan to nvestgate the case of DAG schedulng [7] whch has experenced a renewed nterest under the context of schedulng workflowbased applcatons [8], [10]. 6. Conclusons In ths paper we dscussed the schedulng of ndependent tasks n a heterogeneous envronment nvolvng data transfers. The model we adopted captures all the man parameters due to heterogenety whle mposng a constrant on the number of transfers that can be conducted smultaneously by a sngle node. We develop algorthms wth the goal of mnmzng the total makespan. Among them, the backfll varant of MaxMn was found to acheve the best performance n most cases. In an extended verson we plan to nvestgate further mprovements on BMaxMn, as well as take advantage of the strengths of Genetc Algorthms. 7. References [1] O. Beaumont, A. Legrand and Y. Robert, Optmal Algorthms for Schedulng Dvsble Workloads on Heterogeneous Systems, n Proc. 17 th Int. Parallel and Dstrbuted Processng Symp. (IPDPS 2003), Nce, France, [2] T. Braun, H. Segel, N. Beck, L. Bolon, M. Maheswaran, A. Reuther, J. Robertson, M. Theys and B. Yao, A Comparson of Eleven Statc Heurstcs for Mappng a Class of Independent Tasks onto Heterogeneous Dstrbuted Computng Systems, n Journal of Parallel and Dstrbuted Computng (JPDC), Vol. 61(6), pp , June [3] F. Desprez and A. Vernos, Smultaneous Schedulng of Replcaton and Computaton for Data-Intensve Applcatons on the Grd, Research Report RR , INRIA, France, Jan [4] D. Fetelson, L. Rudolph and U. Schwegelshohn, Parallel job schedulng-a status report, n Proc. 10 th Workshop on Job Schedulng Strateges for Parallel Processng (JSSPP 04), pp. 1-16, June [5] E. Frachtenberg, D. Fetelson, F. Petrn and J. Fernandez, Adaptve Parallel Job Schedulng wth Flexble CoSchedulng, n IEEE Trans. on Parallel and Dstrbuted Systems (TPDS), Vol. 16(11), pp , Nov [6] M. R. Garey and D. S. Johnson, Computers and Intractablty, a Gude to the Theory of NP-Completeness. W. H. Freeman and Company, [7] Yu-K. Kwok and I. Ahmad, Statc schedulng algorthms for allocatng drected task graphs to multprocessors, n ACM Computng. Surveys, Vol. 31(4), pp , [8] G. Malewcz, A. Rosenberg and M. Yurkewych, On Schedulng Complex Dags for Internet-Based Computng, n Proc. 19 th Int. Parallel and Dstrbuted Processng Symp. (IPDPS 2005), Denver, Colorado, USA, Aprl [9] T. Phan, K. Ranganathan and R.Son, Evolvng Toward the Perfect Schedule: Co-Schedulng Job Assgnments and Data Replcaton n Wde-Area Systems Usng a Genetc Algorthm, n Proc. 11th Workshop on Job Schedulng Strateges for Parallel Processng (JSSPP 2005), June 19, [10] R. Prodan and T. Fahrnger, Dynamc Schedulng of Scentfc Workflow Applcatons on the Grd : A Case Study, n Proc. ACM Symp. on Appled Computng (SAC 05), pp , Santa Fe, New Mexco, USA, March [11] K.Ranganathan and I. Foster, Decouplng Computaton and Data Schedulng n Dstrbuted Data- Intensve Applcatons, n Proc. 11th IEEE Symposum on Hgh Performance Dstrbuted Computng (HPDC 02), Ednburgh, Scotland, July [12] S. Sanyal and S. K. Das, MaTCH: Mappng Data- Parallel Tasks on a Heterogeneous Computng Platform usng the Cross-Entropy Heurstc, n Proc. 19th Int. Parallel and Dstrbuted Processng Symp. (IPDPS 2005), Denver, Colorado, USA, Aprl [13] O. Snnen, L. Sousa and F. Sandnes, Toward a Realstc Task Schedulng Model, n IEEE Trans. on Parallel and Dstrbuted Systems (TPDS), Vol. 17(3), pp , March [14] Xan-He Sun and Mng Wu, GHS: A Performance System of Grd Computng, n Proc. 19th Int. Parallel and Dstrbuted Processng Symp. (IPDPS 2005), Denver, Colorado, USA, Aprl [15] S. Venugopal and R. Buyya, A Set Coverage-based Mappng Heurstc for Schedulng Dstrbuted Data-Intensve Applcatons on Global Grds, Techncal Report, GRIDS- TR , Grd Computng and Dstrbuted Systems Laboratory, The Unversty of Melbourne, Australa, March 8, [16] Y. Zhang, H. Franke, J. E. Morera, and A. Svasubramanam, Improvng parallel job schedulng by combnng gang schedulng and backfllng technques, n Proc. 14th Int. Parallel and Dstrbuted Processng Symp. (IPDPS 2000), pp , Cancun, May Acknowledgements Ths work was partally supported by the European Socal Fund and Natonal Resources - (EPEAEK-II) ARCHIMIDIS Ι Ι.

Comparison of Heuristics for Scheduling Independent Tasks on Heterogeneous Distributed Environments

Comparison of Heuristics for Scheduling Independent Tasks on Heterogeneous Distributed Environments Comparson of Heurstcs for Schedulng Independent Tasks on Heterogeneous Dstrbuted Envronments Hesam Izakan¹, Ath Abraham², Senor Member, IEEE, Václav Snášel³ ¹ Islamc Azad Unversty, Ramsar Branch, Ramsar,

More information

Compiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz

Compiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz Compler Desgn Sprng 2014 Regster Allocaton Sample Exercses and Solutons Prof. Pedro C. Dnz USC / Informaton Scences Insttute 4676 Admralty Way, Sute 1001 Marna del Rey, Calforna 90292 pedro@s.edu Regster

More information

Parallelism for Nested Loops with Non-uniform and Flow Dependences

Parallelism for Nested Loops with Non-uniform and Flow Dependences Parallelsm for Nested Loops wth Non-unform and Flow Dependences Sam-Jn Jeong Dept. of Informaton & Communcaton Engneerng, Cheonan Unversty, 5, Anseo-dong, Cheonan, Chungnam, 330-80, Korea. seong@cheonan.ac.kr

More information

The Greedy Method. Outline and Reading. Change Money Problem. Greedy Algorithms. Applications of the Greedy Strategy. The Greedy Method Technique

The Greedy Method. Outline and Reading. Change Money Problem. Greedy Algorithms. Applications of the Greedy Strategy. The Greedy Method Technique //00 :0 AM Outlne and Readng The Greedy Method The Greedy Method Technque (secton.) Fractonal Knapsack Problem (secton..) Task Schedulng (secton..) Mnmum Spannng Trees (secton.) Change Money Problem Greedy

More information

An Optimal Algorithm for Prufer Codes *

An Optimal Algorithm for Prufer Codes * J. Software Engneerng & Applcatons, 2009, 2: 111-115 do:10.4236/jsea.2009.22016 Publshed Onlne July 2009 (www.scrp.org/journal/jsea) An Optmal Algorthm for Prufer Codes * Xaodong Wang 1, 2, Le Wang 3,

More information

Load Balancing for Hex-Cell Interconnection Network

Load Balancing for Hex-Cell Interconnection Network Int. J. Communcatons, Network and System Scences,,, - Publshed Onlne Aprl n ScRes. http://www.scrp.org/journal/jcns http://dx.do.org/./jcns.. Load Balancng for Hex-Cell Interconnecton Network Saher Manaseer,

More information

Distributed Resource Scheduling in Grid Computing Using Fuzzy Approach

Distributed Resource Scheduling in Grid Computing Using Fuzzy Approach Dstrbuted Resource Schedulng n Grd Computng Usng Fuzzy Approach Shahram Amn, Mohammad Ahmad Computer Engneerng Department Islamc Azad Unversty branch Mahallat, Iran Islamc Azad Unversty branch khomen,

More information

CHAPTER 2 PROPOSED IMPROVED PARTICLE SWARM OPTIMIZATION

CHAPTER 2 PROPOSED IMPROVED PARTICLE SWARM OPTIMIZATION 24 CHAPTER 2 PROPOSED IMPROVED PARTICLE SWARM OPTIMIZATION The present chapter proposes an IPSO approach for multprocessor task schedulng problem wth two classfcatons, namely, statc ndependent tasks and

More information

Subspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points;

Subspace clustering. Clustering. Fundamental to all clustering techniques is the choice of distance measure between data points; Subspace clusterng Clusterng Fundamental to all clusterng technques s the choce of dstance measure between data ponts; D q ( ) ( ) 2 x x = x x, j k = 1 k jk Squared Eucldean dstance Assumpton: All features

More information

Chapter 1. Introduction

Chapter 1. Introduction Chapter 1 Introducton 1.1 Parallel Processng There s a contnual demand for greater computatonal speed from a computer system than s currently possble (.e. sequental systems). Areas need great computatonal

More information

A Binarization Algorithm specialized on Document Images and Photos

A Binarization Algorithm specialized on Document Images and Photos A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a

More information

A Genetic Algorithm Based Dynamic Load Balancing Scheme for Heterogeneous Distributed Systems

A Genetic Algorithm Based Dynamic Load Balancing Scheme for Heterogeneous Distributed Systems Proceedngs of the Internatonal Conference on Parallel and Dstrbuted Processng Technques and Applcatons, PDPTA 2008, Las Vegas, Nevada, USA, July 14-17, 2008, 2 Volumes. CSREA Press 2008, ISBN 1-60132-084-1

More information

AADL : about scheduling analysis

AADL : about scheduling analysis AADL : about schedulng analyss Schedulng analyss, what s t? Embedded real-tme crtcal systems have temporal constrants to meet (e.g. deadlne). Many systems are bult wth operatng systems provdng multtaskng

More information

Real-time Fault-tolerant Scheduling Algorithm for Distributed Computing Systems

Real-time Fault-tolerant Scheduling Algorithm for Distributed Computing Systems Real-tme Fault-tolerant Schedulng Algorthm for Dstrbuted Computng Systems Yun Lng, Y Ouyang College of Computer Scence and Informaton Engneerng Zheang Gongshang Unversty Postal code: 310018 P.R.CHINA {ylng,

More information

Module Management Tool in Software Development Organizations

Module Management Tool in Software Development Organizations Journal of Computer Scence (5): 8-, 7 ISSN 59-66 7 Scence Publcatons Management Tool n Software Development Organzatons Ahmad A. Al-Rababah and Mohammad A. Al-Rababah Faculty of IT, Al-Ahlyyah Amman Unversty,

More information

Helsinki University Of Technology, Systems Analysis Laboratory Mat Independent research projects in applied mathematics (3 cr)

Helsinki University Of Technology, Systems Analysis Laboratory Mat Independent research projects in applied mathematics (3 cr) Helsnk Unversty Of Technology, Systems Analyss Laboratory Mat-2.08 Independent research projects n appled mathematcs (3 cr) "! #$&% Antt Laukkanen 506 R ajlaukka@cc.hut.f 2 Introducton...3 2 Multattrbute

More information

S1 Note. Basis functions.

S1 Note. Basis functions. S1 Note. Bass functons. Contents Types of bass functons...1 The Fourer bass...2 B-splne bass...3 Power and type I error rates wth dfferent numbers of bass functons...4 Table S1. Smulaton results of type

More information

Parallel matrix-vector multiplication

Parallel matrix-vector multiplication Appendx A Parallel matrx-vector multplcaton The reduced transton matrx of the three-dmensonal cage model for gel electrophoress, descrbed n secton 3.2, becomes excessvely large for polymer lengths more

More information

A mathematical programming approach to the analysis, design and scheduling of offshore oilfields

A mathematical programming approach to the analysis, design and scheduling of offshore oilfields 17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 A mathematcal programmng approach to the analyss, desgn and

More information

Wishing you all a Total Quality New Year!

Wishing you all a Total Quality New Year! Total Qualty Management and Sx Sgma Post Graduate Program 214-15 Sesson 4 Vnay Kumar Kalakband Assstant Professor Operatons & Systems Area 1 Wshng you all a Total Qualty New Year! Hope you acheve Sx sgma

More information

Assignment # 2. Farrukh Jabeen Algorithms 510 Assignment #2 Due Date: June 15, 2009.

Assignment # 2. Farrukh Jabeen Algorithms 510 Assignment #2 Due Date: June 15, 2009. Farrukh Jabeen Algorthms 51 Assgnment #2 Due Date: June 15, 29. Assgnment # 2 Chapter 3 Dscrete Fourer Transforms Implement the FFT for the DFT. Descrbed n sectons 3.1 and 3.2. Delverables: 1. Concse descrpton

More information

A New Token Allocation Algorithm for TCP Traffic in Diffserv Network

A New Token Allocation Algorithm for TCP Traffic in Diffserv Network A New Token Allocaton Algorthm for TCP Traffc n Dffserv Network A New Token Allocaton Algorthm for TCP Traffc n Dffserv Network S. Sudha and N. Ammasagounden Natonal Insttute of Technology, Truchrappall,

More information

CMPS 10 Introduction to Computer Science Lecture Notes

CMPS 10 Introduction to Computer Science Lecture Notes CPS 0 Introducton to Computer Scence Lecture Notes Chapter : Algorthm Desgn How should we present algorthms? Natural languages lke Englsh, Spansh, or French whch are rch n nterpretaton and meanng are not

More information

6.854 Advanced Algorithms Petar Maymounkov Problem Set 11 (November 23, 2005) With: Benjamin Rossman, Oren Weimann, and Pouya Kheradpour

6.854 Advanced Algorithms Petar Maymounkov Problem Set 11 (November 23, 2005) With: Benjamin Rossman, Oren Weimann, and Pouya Kheradpour 6.854 Advanced Algorthms Petar Maymounkov Problem Set 11 (November 23, 2005) Wth: Benjamn Rossman, Oren Wemann, and Pouya Kheradpour Problem 1. We reduce vertex cover to MAX-SAT wth weghts, such that the

More information

Routing in Degree-constrained FSO Mesh Networks

Routing in Degree-constrained FSO Mesh Networks Internatonal Journal of Hybrd Informaton Technology Vol., No., Aprl, 009 Routng n Degree-constraned FSO Mesh Networks Zpng Hu, Pramode Verma, and James Sluss Jr. School of Electrcal & Computer Engneerng

More information

Simulation Based Analysis of FAST TCP using OMNET++

Simulation Based Analysis of FAST TCP using OMNET++ Smulaton Based Analyss of FAST TCP usng OMNET++ Umar ul Hassan 04030038@lums.edu.pk Md Term Report CS678 Topcs n Internet Research Sprng, 2006 Introducton Internet traffc s doublng roughly every 3 months

More information

Meta-heuristics for Multidimensional Knapsack Problems

Meta-heuristics for Multidimensional Knapsack Problems 2012 4th Internatonal Conference on Computer Research and Development IPCSIT vol.39 (2012) (2012) IACSIT Press, Sngapore Meta-heurstcs for Multdmensonal Knapsack Problems Zhbao Man + Computer Scence Department,

More information

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization Problem efntons and Evaluaton Crtera for Computatonal Expensve Optmzaton B. Lu 1, Q. Chen and Q. Zhang 3, J. J. Lang 4, P. N. Suganthan, B. Y. Qu 6 1 epartment of Computng, Glyndwr Unversty, UK Faclty

More information

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data A Fast Content-Based Multmeda Retreval Technque Usng Compressed Data Borko Furht and Pornvt Saksobhavvat NSF Multmeda Laboratory Florda Atlantc Unversty, Boca Raton, Florda 3343 ABSTRACT In ths paper,

More information

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration Improvement of Spatal Resoluton Usng BlockMatchng Based Moton Estmaton and Frame Integraton Danya Suga and Takayuk Hamamoto Graduate School of Engneerng, Tokyo Unversty of Scence, 6-3-1, Nuku, Katsuska-ku,

More information

Brave New World Pseudocode Reference

Brave New World Pseudocode Reference Brave New World Pseudocode Reference Pseudocode s a way to descrbe how to accomplsh tasks usng basc steps lke those a computer mght perform. In ths week s lab, you'll see how a form of pseudocode can be

More information

Load-Balanced Anycast Routing

Load-Balanced Anycast Routing Load-Balanced Anycast Routng Chng-Yu Ln, Jung-Hua Lo, and Sy-Yen Kuo Department of Electrcal Engneerng atonal Tawan Unversty, Tape, Tawan sykuo@cc.ee.ntu.edu.tw Abstract For fault-tolerance and load-balance

More information

A Hybrid Genetic Algorithm for Routing Optimization in IP Networks Utilizing Bandwidth and Delay Metrics

A Hybrid Genetic Algorithm for Routing Optimization in IP Networks Utilizing Bandwidth and Delay Metrics A Hybrd Genetc Algorthm for Routng Optmzaton n IP Networks Utlzng Bandwdth and Delay Metrcs Anton Redl Insttute of Communcaton Networks, Munch Unversty of Technology, Arcsstr. 21, 80290 Munch, Germany

More information

Term Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task

Term Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task Proceedngs of NTCIR-6 Workshop Meetng, May 15-18, 2007, Tokyo, Japan Term Weghtng Classfcaton System Usng the Ch-square Statstc for the Classfcaton Subtask at NTCIR-6 Patent Retreval Task Kotaro Hashmoto

More information

The Codesign Challenge

The Codesign Challenge ECE 4530 Codesgn Challenge Fall 2007 Hardware/Software Codesgn The Codesgn Challenge Objectves In the codesgn challenge, your task s to accelerate a gven software reference mplementaton as fast as possble.

More information

CHAPTER 2 DECOMPOSITION OF GRAPHS

CHAPTER 2 DECOMPOSITION OF GRAPHS CHAPTER DECOMPOSITION OF GRAPHS. INTRODUCTION A graph H s called a Supersubdvson of a graph G f H s obtaned from G by replacng every edge uv of G by a bpartte graph,m (m may vary for each edge by dentfyng

More information

Fault tolerant workflow scheduling based on replication and resubmission of tasks in Cloud Computing

Fault tolerant workflow scheduling based on replication and resubmission of tasks in Cloud Computing Fault tolerant workflow schedulng based on replcaton and resubmsson of tasks n Cloud Computng Jayadvya S K* Department of Computer Scence & Engneerng Natonal Insttute of Technology Truchrappall, Taml Nadu,

More information

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS ARPN Journal of Engneerng and Appled Scences 006-017 Asan Research Publshng Network (ARPN). All rghts reserved. NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS Igor Grgoryev, Svetlana

More information

Performance Study of Parallel Programming on Cloud Computing Environments Using MapReduce

Performance Study of Parallel Programming on Cloud Computing Environments Using MapReduce Performance Study of Parallel Programmng on Cloud Computng Envronments Usng MapReduce Wen-Chung Shh, Shan-Shyong Tseng Department of Informaton Scence and Applcatons Asa Unversty Tachung, 41354, Tawan

More information

A SYSTOLIC APPROACH TO LOOP PARTITIONING AND MAPPING INTO FIXED SIZE DISTRIBUTED MEMORY ARCHITECTURES

A SYSTOLIC APPROACH TO LOOP PARTITIONING AND MAPPING INTO FIXED SIZE DISTRIBUTED MEMORY ARCHITECTURES A SYSOLIC APPROACH O LOOP PARIIONING AND MAPPING INO FIXED SIZE DISRIBUED MEMORY ARCHIECURES Ioanns Drosts, Nektaros Kozrs, George Papakonstantnou and Panayots sanakas Natonal echncal Unversty of Athens

More information

X- Chart Using ANOM Approach

X- Chart Using ANOM Approach ISSN 1684-8403 Journal of Statstcs Volume 17, 010, pp. 3-3 Abstract X- Chart Usng ANOM Approach Gullapall Chakravarth 1 and Chaluvad Venkateswara Rao Control lmts for ndvdual measurements (X) chart are

More information

An efficient iterative source routing algorithm

An efficient iterative source routing algorithm An effcent teratve source routng algorthm Gang Cheng Ye Tan Nrwan Ansar Advanced Networng Lab Department of Electrcal Computer Engneerng New Jersey Insttute of Technology Newar NJ 7 {gc yt Ansar}@ntedu

More information

Burst Round Robin as a Proportional-Share Scheduling Algorithm

Burst Round Robin as a Proportional-Share Scheduling Algorithm Burst Round Robn as a Proportonal-Share Schedulng Algorthm Tarek Helmy * Abdelkader Dekdouk ** * College of Computer Scence & Engneerng, Kng Fahd Unversty of Petroleum and Mnerals, Dhahran 31261, Saud

More information

Scheduling Remote Access to Scientific Instruments in Cyberinfrastructure for Education and Research

Scheduling Remote Access to Scientific Instruments in Cyberinfrastructure for Education and Research Schedulng Remote Access to Scentfc Instruments n Cybernfrastructure for Educaton and Research Je Yn 1, Junwe Cao 2,3,*, Yuexuan Wang 4, Lanchen Lu 1,3 and Cheng Wu 1,3 1 Natonal CIMS Engneerng and Research

More information

Efficient Content Distribution in Wireless P2P Networks

Efficient Content Distribution in Wireless P2P Networks Effcent Content Dstrbuton n Wreless P2P Networs Qong Sun, Vctor O. K. L, and Ka-Cheong Leung Department of Electrcal and Electronc Engneerng The Unversty of Hong Kong Pofulam Road, Hong Kong, Chna {oansun,

More information

Complex Numbers. Now we also saw that if a and b were both positive then ab = a b. For a second let s forget that restriction and do the following.

Complex Numbers. Now we also saw that if a and b were both positive then ab = a b. For a second let s forget that restriction and do the following. Complex Numbers The last topc n ths secton s not really related to most of what we ve done n ths chapter, although t s somewhat related to the radcals secton as we wll see. We also won t need the materal

More information

Video Proxy System for a Large-scale VOD System (DINA)

Video Proxy System for a Large-scale VOD System (DINA) Vdeo Proxy System for a Large-scale VOD System (DINA) KWUN-CHUNG CHAN #, KWOK-WAI CHEUNG *# #Department of Informaton Engneerng *Centre of Innovaton and Technology The Chnese Unversty of Hong Kong SHATIN,

More information

Channel 0. Channel 1 Channel 2. Channel 3 Channel 4. Channel 5 Channel 6 Channel 7

Channel 0. Channel 1 Channel 2. Channel 3 Channel 4. Channel 5 Channel 6 Channel 7 Optmzed Regonal Cachng for On-Demand Data Delvery Derek L. Eager Mchael C. Ferrs Mary K. Vernon Unversty of Saskatchewan Unversty of Wsconsn Madson Saskatoon, SK Canada S7N 5A9 Madson, WI 5376 eager@cs.usask.ca

More information

NAG Fortran Library Chapter Introduction. G10 Smoothing in Statistics

NAG Fortran Library Chapter Introduction. G10 Smoothing in Statistics Introducton G10 NAG Fortran Lbrary Chapter Introducton G10 Smoothng n Statstcs Contents 1 Scope of the Chapter... 2 2 Background to the Problems... 2 2.1 Smoothng Methods... 2 2.2 Smoothng Splnes and Regresson

More information

Kent State University CS 4/ Design and Analysis of Algorithms. Dept. of Math & Computer Science LECT-16. Dynamic Programming

Kent State University CS 4/ Design and Analysis of Algorithms. Dept. of Math & Computer Science LECT-16. Dynamic Programming CS 4/560 Desgn and Analyss of Algorthms Kent State Unversty Dept. of Math & Computer Scence LECT-6 Dynamc Programmng 2 Dynamc Programmng Dynamc Programmng, lke the dvde-and-conquer method, solves problems

More information

An Entropy-Based Approach to Integrated Information Needs Assessment

An Entropy-Based Approach to Integrated Information Needs Assessment Dstrbuton Statement A: Approved for publc release; dstrbuton s unlmted. An Entropy-Based Approach to ntegrated nformaton Needs Assessment June 8, 2004 Wllam J. Farrell Lockheed Martn Advanced Technology

More information

Related-Mode Attacks on CTR Encryption Mode

Related-Mode Attacks on CTR Encryption Mode Internatonal Journal of Network Securty, Vol.4, No.3, PP.282 287, May 2007 282 Related-Mode Attacks on CTR Encrypton Mode Dayn Wang, Dongda Ln, and Wenlng Wu (Correspondng author: Dayn Wang) Key Laboratory

More information

Sum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints

Sum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints Australan Journal of Basc and Appled Scences, 2(4): 1204-1208, 2008 ISSN 1991-8178 Sum of Lnear and Fractonal Multobjectve Programmng Problem under Fuzzy Rules Constrants 1 2 Sanjay Jan and Kalash Lachhwan

More information

Dynamic Voltage Scaling of Supply and Body Bias Exploiting Software Runtime Distribution

Dynamic Voltage Scaling of Supply and Body Bias Exploiting Software Runtime Distribution Dynamc Voltage Scalng of Supply and Body Bas Explotng Software Runtme Dstrbuton Sungpack Hong EE Department Stanford Unversty Sungjoo Yoo, Byeong Bn, Kyu-Myung Cho, Soo-Kwan Eo Samsung Electroncs Taehwan

More information

Empirical Distributions of Parameter Estimates. in Binary Logistic Regression Using Bootstrap

Empirical Distributions of Parameter Estimates. in Binary Logistic Regression Using Bootstrap Int. Journal of Math. Analyss, Vol. 8, 4, no. 5, 7-7 HIKARI Ltd, www.m-hkar.com http://dx.do.org/.988/jma.4.494 Emprcal Dstrbutons of Parameter Estmates n Bnary Logstc Regresson Usng Bootstrap Anwar Ftranto*

More information

TECHNIQUE OF FORMATION HOMOGENEOUS SAMPLE SAME OBJECTS. Muradaliyev A.Z.

TECHNIQUE OF FORMATION HOMOGENEOUS SAMPLE SAME OBJECTS. Muradaliyev A.Z. TECHNIQUE OF FORMATION HOMOGENEOUS SAMPLE SAME OBJECTS Muradalyev AZ Azerbajan Scentfc-Research and Desgn-Prospectng Insttute of Energetc AZ1012, Ave HZardab-94 E-mal:aydn_murad@yahoocom Importance of

More information

Course Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms

Course Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms Course Introducton Course Topcs Exams, abs, Proects A quc loo at a few algorthms 1 Advanced Data Structures and Algorthms Descrpton: We are gong to dscuss algorthm complexty analyss, algorthm desgn technques

More information

Steps for Computing the Dissimilarity, Entropy, Herfindahl-Hirschman and. Accessibility (Gravity with Competition) Indices

Steps for Computing the Dissimilarity, Entropy, Herfindahl-Hirschman and. Accessibility (Gravity with Competition) Indices Steps for Computng the Dssmlarty, Entropy, Herfndahl-Hrschman and Accessblty (Gravty wth Competton) Indces I. Dssmlarty Index Measurement: The followng formula can be used to measure the evenness between

More information

5 The Primal-Dual Method

5 The Primal-Dual Method 5 The Prmal-Dual Method Orgnally desgned as a method for solvng lnear programs, where t reduces weghted optmzaton problems to smpler combnatoral ones, the prmal-dual method (PDM) has receved much attenton

More information

CS 534: Computer Vision Model Fitting

CS 534: Computer Vision Model Fitting CS 534: Computer Vson Model Fttng Sprng 004 Ahmed Elgammal Dept of Computer Scence CS 534 Model Fttng - 1 Outlnes Model fttng s mportant Least-squares fttng Maxmum lkelhood estmaton MAP estmaton Robust

More information

Programming in Fortran 90 : 2017/2018

Programming in Fortran 90 : 2017/2018 Programmng n Fortran 90 : 2017/2018 Programmng n Fortran 90 : 2017/2018 Exercse 1 : Evaluaton of functon dependng on nput Wrte a program who evaluate the functon f (x,y) for any two user specfed values

More information

Classifier Selection Based on Data Complexity Measures *

Classifier Selection Based on Data Complexity Measures * Classfer Selecton Based on Data Complexty Measures * Edth Hernández-Reyes, J.A. Carrasco-Ochoa, and J.Fco. Martínez-Trndad Natonal Insttute for Astrophyscs, Optcs and Electroncs, Lus Enrque Erro No.1 Sta.

More information

DESIGNING TRANSMISSION SCHEDULES FOR WIRELESS AD HOC NETWORKS TO MAXIMIZE NETWORK THROUGHPUT

DESIGNING TRANSMISSION SCHEDULES FOR WIRELESS AD HOC NETWORKS TO MAXIMIZE NETWORK THROUGHPUT DESIGNING TRANSMISSION SCHEDULES FOR WIRELESS AD HOC NETWORKS TO MAXIMIZE NETWORK THROUGHPUT Bran J. Wolf, Joseph L. Hammond, and Harlan B. Russell Dept. of Electrcal and Computer Engneerng, Clemson Unversty,

More information

Maintaining temporal validity of real-time data on non-continuously executing resources

Maintaining temporal validity of real-time data on non-continuously executing resources Mantanng temporal valdty of real-tme data on non-contnuously executng resources Tan Ba, Hong Lu and Juan Yang Hunan Insttute of Scence and Technology, College of Computer Scence, 44, Yueyang, Chna Wuhan

More information

Shared Running Buffer Based Proxy Caching of Streaming Sessions

Shared Running Buffer Based Proxy Caching of Streaming Sessions Shared Runnng Buffer Based Proxy Cachng of Streamng Sessons Songqng Chen, Bo Shen, Yong Yan, Sujoy Basu Moble and Meda Systems Laboratory HP Laboratores Palo Alto HPL-23-47 March th, 23* E-mal: sqchen@cs.wm.edu,

More information

Support Vector Machines

Support Vector Machines /9/207 MIST.6060 Busness Intellgence and Data Mnng What are Support Vector Machnes? Support Vector Machnes Support Vector Machnes (SVMs) are supervsed learnng technques that analyze data and recognze patterns.

More information

Learning the Kernel Parameters in Kernel Minimum Distance Classifier

Learning the Kernel Parameters in Kernel Minimum Distance Classifier Learnng the Kernel Parameters n Kernel Mnmum Dstance Classfer Daoqang Zhang 1,, Songcan Chen and Zh-Hua Zhou 1* 1 Natonal Laboratory for Novel Software Technology Nanjng Unversty, Nanjng 193, Chna Department

More information

Machine Learning: Algorithms and Applications

Machine Learning: Algorithms and Applications 14/05/1 Machne Learnng: Algorthms and Applcatons Florano Zn Free Unversty of Bozen-Bolzano Faculty of Computer Scence Academc Year 011-01 Lecture 10: 14 May 01 Unsupervsed Learnng cont Sldes courtesy of

More information

Smoothing Spline ANOVA for variable screening

Smoothing Spline ANOVA for variable screening Smoothng Splne ANOVA for varable screenng a useful tool for metamodels tranng and mult-objectve optmzaton L. Rcco, E. Rgon, A. Turco Outlne RSM Introducton Possble couplng Test case MOO MOO wth Game Theory

More information

SAO: A Stream Index for Answering Linear Optimization Queries

SAO: A Stream Index for Answering Linear Optimization Queries SAO: A Stream Index for Answerng near Optmzaton Queres Gang uo Kun-ung Wu Phlp S. Yu IBM T.J. Watson Research Center {luog, klwu, psyu}@us.bm.com Abstract near optmzaton queres retreve the top-k tuples

More information

WIRELESS communication technology has gained widespread

WIRELESS communication technology has gained widespread 616 IEEE TRANSACTIONS ON MOBILE COMPUTING, VOL. 4, NO. 6, NOVEMBER/DECEMBER 2005 Dstrbuted Far Schedulng n a Wreless LAN Ntn Vadya, Senor Member, IEEE, Anurag Dugar, Seema Gupta, and Paramvr Bahl, Senor

More information

Sequential search. Building Java Programs Chapter 13. Sequential search. Sequential search

Sequential search. Building Java Programs Chapter 13. Sequential search. Sequential search Sequental search Buldng Java Programs Chapter 13 Searchng and Sortng sequental search: Locates a target value n an array/lst by examnng each element from start to fnsh. How many elements wll t need to

More information

Some Advanced SPC Tools 1. Cumulative Sum Control (Cusum) Chart For the data shown in Table 9-1, the x chart can be generated.

Some Advanced SPC Tools 1. Cumulative Sum Control (Cusum) Chart For the data shown in Table 9-1, the x chart can be generated. Some Advanced SP Tools 1. umulatve Sum ontrol (usum) hart For the data shown n Table 9-1, the x chart can be generated. However, the shft taken place at sample #21 s not apparent. 92 For ths set samples,

More information

A GENETIC ALGORITHM FOR PROCESS SCHEDULING IN DISTRIBUTED OPERATING SYSTEMS CONSIDERING LOAD BALANCING

A GENETIC ALGORITHM FOR PROCESS SCHEDULING IN DISTRIBUTED OPERATING SYSTEMS CONSIDERING LOAD BALANCING A GENETIC ALGORITHM FOR PROCESS SCHEDULING IN DISTRIBUTED OPERATING SYSTEMS CONSIDERING LOAD BALANCING M. Nkravan and M. H. Kashan Department of Electrcal Computer Islamc Azad Unversty, Shahrar Shahreqods

More information

A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS

A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS Proceedngs of the Wnter Smulaton Conference M E Kuhl, N M Steger, F B Armstrong, and J A Jones, eds A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS Mark W Brantley Chun-Hung

More information

Active Contours/Snakes

Active Contours/Snakes Actve Contours/Snakes Erkut Erdem Acknowledgement: The sldes are adapted from the sldes prepared by K. Grauman of Unversty of Texas at Austn Fttng: Edges vs. boundares Edges useful sgnal to ndcate occludng

More information

Motivation. EE 457 Unit 4. Throughput vs. Latency. Performance Depends on View Point?! Computer System Performance. An individual user wants to:

Motivation. EE 457 Unit 4. Throughput vs. Latency. Performance Depends on View Point?! Computer System Performance. An individual user wants to: 4.1 4.2 Motvaton EE 457 Unt 4 Computer System Performance An ndvdual user wants to: Mnmze sngle program executon tme A datacenter owner wants to: Maxmze number of Mnmze ( ) http://e-tellgentnternetmarketng.com/webste/frustrated-computer-user-2/

More information

TPL-Aware Displacement-driven Detailed Placement Refinement with Coloring Constraints

TPL-Aware Displacement-driven Detailed Placement Refinement with Coloring Constraints TPL-ware Dsplacement-drven Detaled Placement Refnement wth Colorng Constrants Tao Ln Iowa State Unversty tln@astate.edu Chrs Chu Iowa State Unversty cnchu@astate.edu BSTRCT To mnmze the effect of process

More information

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching A Fast Vsual Trackng Algorthm Based on Crcle Pxels Matchng Zhqang Hou hou_zhq@sohu.com Chongzhao Han czhan@mal.xjtu.edu.cn Ln Zheng Abstract: A fast vsual trackng algorthm based on crcle pxels matchng

More information

Solving two-person zero-sum game by Matlab

Solving two-person zero-sum game by Matlab Appled Mechancs and Materals Onlne: 2011-02-02 ISSN: 1662-7482, Vols. 50-51, pp 262-265 do:10.4028/www.scentfc.net/amm.50-51.262 2011 Trans Tech Publcatons, Swtzerland Solvng two-person zero-sum game by

More information

Some material adapted from Mohamed Younis, UMBC CMSC 611 Spr 2003 course slides Some material adapted from Hennessy & Patterson / 2003 Elsevier

Some material adapted from Mohamed Younis, UMBC CMSC 611 Spr 2003 course slides Some material adapted from Hennessy & Patterson / 2003 Elsevier Some materal adapted from Mohamed Youns, UMBC CMSC 611 Spr 2003 course sldes Some materal adapted from Hennessy & Patterson / 2003 Elsever Scence Performance = 1 Executon tme Speedup = Performance (B)

More information

MULTIHOP wireless networks are a paradigm in wireless

MULTIHOP wireless networks are a paradigm in wireless 400 IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, VOL. 17, NO. 1, JANUARY 2018 Toward Optmal Dstrbuted Node Schedulng n a Multhop Wreless Network Through Local Votng Dmtros J. Vergados, Member, IEEE, Natala

More information

Reducing Frame Rate for Object Tracking

Reducing Frame Rate for Object Tracking Reducng Frame Rate for Object Trackng Pavel Korshunov 1 and We Tsang Oo 2 1 Natonal Unversty of Sngapore, Sngapore 11977, pavelkor@comp.nus.edu.sg 2 Natonal Unversty of Sngapore, Sngapore 11977, oowt@comp.nus.edu.sg

More information

Performance and Cost Optimization for Multiple Large-scale Grid Workflow Applications

Performance and Cost Optimization for Multiple Large-scale Grid Workflow Applications Performance and Cost Optmzaton for Multple Large-scale Grd Worflow Applcatons Rubng Duan, Radu Prodan, Thomas Fahrnger Insttute of Computer cence, Unversty of Innsbruc Emal: rubng.duan@ub.ac.at ABTRACT

More information

On the Fairness-Efficiency Tradeoff for Packet Processing with Multiple Resources

On the Fairness-Efficiency Tradeoff for Packet Processing with Multiple Resources On the Farness-Effcency Tradeoff for Packet Processng wth Multple Resources We Wang, Chen Feng, Baochun L, and Ben Lang Department of Electrcal and Computer Engneerng, Unversty of Toronto {wewang, cfeng,

More information

A fair buffer allocation scheme

A fair buffer allocation scheme A far buffer allocaton scheme Juha Henanen and Kalev Klkk Telecom Fnland P.O. Box 228, SF-330 Tampere, Fnland E-mal: juha.henanen@tele.f Abstract An approprate servce for data traffc n ATM networks requres

More information

Range images. Range image registration. Examples of sampling patterns. Range images and range surfaces

Range images. Range image registration. Examples of sampling patterns. Range images and range surfaces Range mages For many structured lght scanners, the range data forms a hghly regular pattern known as a range mage. he samplng pattern s determned by the specfc scanner. Range mage regstraton 1 Examples

More information

Constructing Minimum Connected Dominating Set: Algorithmic approach

Constructing Minimum Connected Dominating Set: Algorithmic approach Constructng Mnmum Connected Domnatng Set: Algorthmc approach G.N. Puroht and Usha Sharma Centre for Mathematcal Scences, Banasthal Unversty, Rajasthan 304022 usha.sharma94@yahoo.com Abstract: Connected

More information

Improved Resource Allocation Algorithms for Practical Image Encoding in a Ubiquitous Computing Environment

Improved Resource Allocation Algorithms for Practical Image Encoding in a Ubiquitous Computing Environment JOURNAL OF COMPUTERS, VOL. 4, NO. 9, SEPTEMBER 2009 873 Improved Resource Allocaton Algorthms for Practcal Image Encodng n a Ubqutous Computng Envronment Manxong Dong, Long Zheng, Kaoru Ota, Song Guo School

More information

Optimal Workload-based Weighted Wavelet Synopses

Optimal Workload-based Weighted Wavelet Synopses Optmal Workload-based Weghted Wavelet Synopses Yoss Matas School of Computer Scence Tel Avv Unversty Tel Avv 69978, Israel matas@tau.ac.l Danel Urel School of Computer Scence Tel Avv Unversty Tel Avv 69978,

More information

Array transposition in CUDA shared memory

Array transposition in CUDA shared memory Array transposton n CUDA shared memory Mke Gles February 19, 2014 Abstract Ths short note s nspred by some code wrtten by Jeremy Appleyard for the transposton of data through shared memory. I had some

More information

On the Network Partitioning of Large Urban Transportation Networks

On the Network Partitioning of Large Urban Transportation Networks On the etwor Parttonng of Large Urban Transportaton etwors Hamdeh Etemadna and Khaled Abdelghany Abstract Ths paper ams at developng a traffc networ parttonng mechansm for dstrbuted traffc management applcatons.

More information

An Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation

An Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation 17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 An Iteratve Soluton Approach to Process Plant Layout usng Mxed

More information

GSLM Operations Research II Fall 13/14

GSLM Operations Research II Fall 13/14 GSLM 58 Operatons Research II Fall /4 6. Separable Programmng Consder a general NLP mn f(x) s.t. g j (x) b j j =. m. Defnton 6.. The NLP s a separable program f ts objectve functon and all constrants are

More information

Virtual Memory. Background. No. 10. Virtual Memory: concept. Logical Memory Space (review) Demand Paging(1) Virtual Memory

Virtual Memory. Background. No. 10. Virtual Memory: concept. Logical Memory Space (review) Demand Paging(1) Virtual Memory Background EECS. Operatng System Fundamentals No. Vrtual Memory Prof. Hu Jang Department of Electrcal Engneerng and Computer Scence, York Unversty Memory-management methods normally requres the entre process

More information

Fibre-Optic AWG-based Real-Time Networks

Fibre-Optic AWG-based Real-Time Networks Fbre-Optc AWG-based Real-Tme Networks Krstna Kunert, Annette Böhm, Magnus Jonsson, School of Informaton Scence, Computer and Electrcal Engneerng, Halmstad Unversty {Magnus.Jonsson, Krstna.Kunert}@de.hh.se

More information

Problem Set 3 Solutions

Problem Set 3 Solutions Introducton to Algorthms October 4, 2002 Massachusetts Insttute of Technology 6046J/18410J Professors Erk Demane and Shaf Goldwasser Handout 14 Problem Set 3 Solutons (Exercses were not to be turned n,

More information

A comparative study of scheduling algorithms for the multiple deadline-constrained workflows in heterogeneous computing systems with time windows

A comparative study of scheduling algorithms for the multiple deadline-constrained workflows in heterogeneous computing systems with time windows Proceda Computer Scence Volume 29, 2014, Pages 509 522 ICCS 2014 14th Internatonal Conference on Computatonal Scence A comparatve study of schedulng algorthms for the multple deadlne-constraned workflows

More information

Performance Evaluation of Information Retrieval Systems

Performance Evaluation of Information Retrieval Systems Why System Evaluaton? Performance Evaluaton of Informaton Retreval Systems Many sldes n ths secton are adapted from Prof. Joydeep Ghosh (UT ECE) who n turn adapted them from Prof. Dk Lee (Unv. of Scence

More information

Content Classification for Caching under CDNs

Content Classification for Caching under CDNs Content Classfcaton for Cachng under CDNs George Palls 1,2, Charlaos homos 2, Konstantnos Stamos 2, Athena Vakal 2, George Andreads 3 1 Department of Computer Scence, Unversty of Cyprus 2 Department of

More information