Clustered Multimedia NOD : Popularity-Based Article Prefetching and Placement

Size: px
Start display at page:

Download "Clustered Multimedia NOD : Popularity-Based Article Prefetching and Placement"

Transcription

1 Clustered Multmeda NOD : Popularty-Based Artcle Prefetchng and Placement Y.J.Km, T.U.Cho, K.O.Jung, Y.K.Kang, S.H.Park, K-Dong Chung Department of Computer Scence, Pusan Natonal Unversty, Korea Abstract Accordng to the current profound development of multmeda and networkng technologes, the way people communcate wth, naturally, has evolved from a textorented nto a multmeda-orented. But, the VOD system technologes can t be easly appled to MNOD system because of the dfference of the basc data and ts propertes. Ths s the reason why we propose the MNOD system. NOD data composng news artcles make a dfference to VOD data n terms of meda type and sze, lfe cycle of artcles and frequency of clents nteracton. Because of NOD data's ntrnsc characterstcs, NOD artcle popularty model may be dfferent from that of VOD vdeos. Hence, we analyze statstcally the log data of one electronc newspaper and show the popularty dstrbuton of artcles s dfferent form Zpf s, whch s a popularty model of VOD data. We propose a new artcle popularty model for NOD data, whch we call Mult-Selecton Zpf dstrbuton. Also, we propose the artcle prefetchng polcy based on the popularty model and lfe cycle model of NOD artcles for ncreasng performance n a MNOD system. And, we consder the data placement on the MNOD system wth prefetch cache. The user requests can be servced from a cache or a dsk. Consequently, n order to place a NOD data, we must consder the correlaton between dsks and cache accordng to data popularty. In ths paper, we propose the data placement polcy that consders both dsk-exstence probablty and cache-exstence probablty of a data usng data popularty. Introducton. Motvaton As a VOD system gets more popularty these days, a NOD(News On Demand) system supportng multmeda servces, we call a MNOD (Multmeda NOD) system n ths paper, wll be popular as a news servce n the near future. Furthermore, the developments of computer network and multmeda technology are makng the multmeda-orented news servces possble. Prevous works generally regarded VOD and MNOD systems as an dentcal category from the vewpont of usng multmeda data. Multmeda data usually need hgh bandwdth and massve storage space and have real-tme crtera. Therefore, we bascally utlze the research results of a VOD system. Clearly, a MNOD system can be categorzed nto the same multmeda applcaton as a VOD system, but t has some ntrnsc aspects that a VOD system doesn't possess[4]. Frst, the artcles of NOD servce are made on and off n a day whle the VOD data s made once a halfmonth or a month. Second, the more recent artcle s made, the more users access the artcle. The good moves are preferred to the others for a long tme, but NOD data aren't. The NOD artcles, whch have the best popularty, don't last more than three days. Thrd, the number of artcles that are requested from each user s vared. Owng to the long length of vdeo data, users servced n VOD system select just one or two vdeo data. But the length of NOD data s short, as a result, users can select several data at a tme. Fnally, NOD data have the temporal and spatal access localty. The user requests burst accordng to the access tme and the knd of artcles. So the VOD system can't be fully appled to a MNOD system because of these dfferent characterstcs. In short, we use the VOD system research results bascally, but wll revse and decorate t. The prevous work showed that the system wth prefetchng has the better performance than the system wthout prefetchng[, 5]. But wthout consderng the other characterzatons such as short-term lfe cycle, the dfference of user s access pattern accordng to tme and etc, we can not expect to the ncreased performance n the MNOD system. The system wth prefetchng only based on data popularty has the problem, whch s data replacement happens too often. Hence, usng the analyss of electronc newspaper logfles, we propose the popularty model and the lfe cycle model of NOD artcles. And we wll suggest the artcle prefetchng polcy based on tme wndow accordng as the models. And then we consder the data placement on system wth prefetch cache. The user requests can be servced from a cache or a dsk. Consequently, n order to place a MNOD data, we must consder the correlaton between dsks and cache accordng to data popularty. In ths paper, we wll propose the data placement polcy that consders both dskexstence probablty and cache-exstence probablty of a data usng data popularty. The remander of ths paper s organzed as follow: In secton 2 the overall archtecture of a proposed MNOD system s presented brefly. Secton 3 shows the popularty model and lfe cycle model of NOD artcles va the statstcal analyss of log fles of electronc newspapers, and n secton 4 we propose the artcle prefetchng polcy based on the popularty and lfe cycle model of NOD artcles. In secton 5 4

2 the artcle placement polcy n the MNOD system wth prefetch cache s proposed and evaluated. Lastly, n secton 6 we menton the concluson and future works..2 Related work Data cachng and Prefetchng In order to establsh cachng polcy for NOD data, the popularty and lfe cycle model are mportant condtons. [] showed that the popularty dstrbuton of real vdeo was a double exponental dstrbuton rather than a Zpf s and proposed cache polcy based on the lfe cycle of vdeos. In [], the long-term model of move popularty was proposed for a herarchcal VOD system, whch was the modfed exponental dstrbuton wth three parameters. The strateges of these papers are not acceptable for MNOD system because these analyzed only a VOD data, whose lfe cycle s longer than that of NOD artcle. [] made a psychologcal analyss of access pattern of users, based on a model on human memory and nssted that the recency of data was more mportant crteron n data cachng algorthm than the frequency. But because of varous characterstcs of NOD artcles, consderng only access recency s nsuffcent for cachng or placement of artcles. [4] proposed a news artcle prefetchng polcy but had not the correct crteron for the data replacement. In ths paper, we wll show the measure of short-term and long-term of artcle popularty. And we wll propose artcle prefetchng polcy consderng that measures. Data placement In Multmeda system, data placement polces have been studed so as to ncrease the success rato of user requests. Therefore, most studes were contnued to get the hgh dsk bandwdth. Data Strpng method s the representatve placement polcy. The famous strpng methods are round-robn, random[3], staggered[7], permutaton strpng[8] and etc. And then, the polces to prevent dsk bottleneck, such as dynamc replca polcy[5] and data placement polcy based on popularty were proposed. But these polces consdered only dsk condtons based on data popularty. But n these days, snce almost all systems have data cache, the user requests are servced from a cache or a dsk. Consequently, n order to place data, we must consder the correlaton between dsks and cache accordng as data popularty. 2. MNOD system overvew Though a NOD(News On Demand) system has many dfferences from the well-known VOD(Vdeo On Demand) system, a NOD system s stll, n many aspects, smlar to a VOD system n terms of usng multmeda data. Hence, we can draw out the structure of a MNOD system usng that of a VOD system. And the scalablty of an MNOD system has to be nvestgated to support hundreds of thousands of smultaneous users. As shown n Fgure, we suggest the two-layered clustered MNOD server system by consderng the scalablty. The suggested system structure contans two man modules: a control node and a storage node. Many studes on the VOD system archtectures, theoretcally, assume that there are two logcal parts n the server[, 2, 5]. Roughly speakng, the control node deals wth the admsson control for the storage node and the storage node schedules the admtted requests from the control node and treats storng and retrevng data. The control node s entrusted wth the control part of the whole MNOD server. It controls the permsson nto enterng the storage node. If a NOD stream arrves at the control node, the admsson manager checks f the requested stream may be admtted or not and whenever a request s permtted, t s passed from the control node to the storage node. The storage node s responsble for both storng and retrevng NOD data. Now that the storage node only schedules the permtted streams from the control node, no &RQWUROÃRGH 6WRUDJHÃRGV Fgure. Proposed clustered MNOD system specal admsson control s requred. 3. Artcle popularty model &RQWUROÃRGH NOD data composng news artcles make a dfference to vdeo data of VOD n terms of meda type and sze, lfe cycle of artcle and frequent clents nteracton etc. And, NOD data may be created on and off at any tmes, and t s popular and recent artcles that most of the clents access manly. Because of these characterstcs, NOD artcle 5

3 popularty model may be dfferent from that of VOD move. In ths secton we gve a statstcal analyss of log data of electronc newspaper runnng currently, and show that the popularty dstrbuton of artcle s dfferent from Zpf s, whch s a popularty model of VOD. In partcular, we propose a new artcle popularty model, whch we call Mult-Selecton Zpf dstrbuton, and the algorthm generatng ts probablty dstrbuton. Ths artcle popularty model s a short-term popularty model that denotes popularty dstrbuton among artcles n an nstant tme. Whle the short-term popularty model notfes what artcles are prefetched, long-term popularty models, denotng the lfe cycle of artcles, notfy when the artcles are prefetched. In addton, these models can be used for cachng and placement of artcles. Thus, n ths secton we also present a long-term popularty model of NOD artcles. 3. Mult-Selecton Zpf dstrbuton (a) Average access popularty NOD data Zpf Exponental (b) Comparson wth other dstrbutons Fgure 2. The real access dstrbuton of NOD artcle The typcal popularty model of VOD data s Zpf dstrbuton; t denotes access probablty of the -th popular move when the moves are sorted n the order of decreasng popularty. Then, do the popularty of NOD data follow a Zpf dstrbuton? We analyzed statstcally the log fles of electronc newspapers runnng currently and found that news artcle s popularty model s dfferent from Zpf s. In Fgure2, (a) shows the mean access popularty of news artcles for a month, resulted from makng a statstcal analyss of log data of June 8. (b) shows the comparson of NOD artcle popularty wth Zpf s and Exponental s; The upper porton of NOD artcle dstrbuton decrease less drastcally than Zpf s but more fast than Exponental s. When a user connects wth a NOD server, the user requests several artcles rather than only one artcle. Ths mply that popularty does not concentrate on one artcle but spread over several artcles. Consequently the curve of popularty dstrbuton of NOD artcles s less steep than Zpf s. For comparson, we use a pure Zpf dstrbuton wthout skew and an exponental dstrbuton wth λ =.. A user does not watch only one artcle but several artcles whle he s connected wth a NOD server. If a user selects k artcles on average n NOD server wth N artcles, the possble number of k artcle selectons s as follow: N N! Μ = = k k!( N k)! In other words t may be thought that he selects a group composed of k artcles n Μ artcle groups. Suppose that the popularty of these Μ artcle groups follow a Zpf dstrbuton, the probablty on whch each artcle s selected s the normalzed sum of probabltes of groups contanng the artcle. The defnton of Mult-Selecton Zpf dstrbuton and the algorthm generatng the probablty dstrbuton are as follows: Defnton: Suppose that a user selects k artcles on average n NOD server wth N artcles and the popularty of the Μ artcle groups follows a Zpf dstrbuton, we defne artcle access probablty as the normalzed sum of probabltes of artcle groups contanng each artcle. We call the dstrbuton of artcle access probablty Mult- Selecton Zpf dstrbuton Input: k (the number of selectons), N (the number of artcle) Algorthm: Generate Μ artcle groups; Decde popularty of the groups and sort n Popularty order; Calculate the populartes of groups usng Zpf dstrbuton; whle (all artcles) { addng the probablty of groups contanng the artcle; normalze such that the sum of calculated probabltes s ; Fgure 3. Mult-Selecton Zpf probablty generaton algorthm Fgure4 compares Mult-Selecton Zpf dstrbuton wth Zpf s; (a) shows that the curve of Mult-Selecton Zpf s s less steep than Zpf s. In (b), as the number of 6

4 access popularty (a)comparson of k=5 wth Zpf Mzpf(K3) Mzpf(K5) (b)comparson of k=5 wth k=3 Fgure 4. Mult-selecton Zpf dstrbuton and Zpf s averg Mzpf(K5) hour (a) Access Popularty as artcle lfe cycle.2 access popularty.5..5 atrcle A artcle B hour (b) Lfe cycle of two artcles Fgure 6. Artcle lfe cycle selectons s 3 and 5, It shows that the dstrbuton wth 5 selectons decrease slower than 3 selectons. That mples that the more selectons are, the less steep the curve s. Fgure5 shows the comparson of the real popularty resulted from log fles wth Mult-Selecton Zpf s wth 5 selectons. 3.2 Long-term popularty model Fgure 5. Mult-selecton Zpf dstrbuton & the real dstrbuton of NOD artcle The popularty of artcles changes as tme passes because new artcles are generated at any tme and users 7 prefer new artcles. Long-term popularty model, called as artcle lfe cycle, reflects the changes of the popularty of an artcle. In Fgure6, (a) shows a long-term popularty dstrbuton as a result of statstcal analyss of log fles; t has left skew, whch shows that users prefer new artcles to the old ones. Long-term popularty model can be used for decdng the tme to prefetch an artcle nto the cache to decrease the retreval overhead. In Fgure6 (b), the popularty of artcle A decrease and artcle B ncrease, and the best pont that the

5 artcle A should be replaced wth artcle B n the cache s around the hour, when the curves of two artcles meet. 4. Artcle prefetchng polcy In prevously suggested clustered MNOD server, clents request artcles from CSVR(Control Server), and CSVR fetches artcles from SSVR(Storage Server) and sends them to clents. If popular artcles are prefetched n CSVR, the servce delay can be reduced and network traffcs can be decreased by elmnatng the tme to retreve artcles from SSVR[5]. In prefetchng popular artcles, old artcles must be replaced wth new popular ones when new artcles are generated and the populartes of old artcles decrease. So, we need to consder two popularty models mentoned n the prevous secton for better performance n artcle prefetchng. In ths secton we present the new prefetchng measure combned two popularty models and suggest the artcle prefetchng algorthm employng the measure. 4. The measure of prefetchng We express the measure of short-term artcle popularty n terms of the access frequency durng some tme ntervals and the measure of long-term artcle popularty n terms of the ncreased or decreased amount of frequency durng some tme nterval. For example, the artcle whose access frequency ncrease rapdly may be a new artcle whle the artcle whose access frequency decrease may be less old one and the artcle that access frequency hardly decrease may be old one. Note that we use tme wndow whose nterval s several ten mnutes and calculate f, the access frequency, by cumulatng references to the artcle durng any wndow. By takng a weghted average of fluctuatons n all the wndows, we can calculate a, the expected varaton amount of access frequency n next wndow as follows: a [Eq.] = sf ( f f ) + ( sf ) a In Eq, sf, denotng a smoothng factor, s the weght that ndcate the mportance of frequency varatons of current wndow; The larger sf s, the heaver the weght of the current wndow s frequency s. Consequently we propose a new measure LBF (Lfe-cycle Based Frequency) by combnng f wth a as follows: LBF = f + a [Eq.2] 4.2 Prefetchng scheme CSVR mantans logs of reference to each artcle durng each wndow. From the logs, access frequency f of each artcle n the wndow can be calculated. Compute a usng [Eq.] and LBF of each artcle usng [Eq.2]. Because the artcle that has the largest LBF s expected to be accessed most frequently n next wndow, the artcles that have large LBF have to be prefetched. Once artcles are prefetched n CSVR, the artcles can t be replaced wth others durng current wndow. Fgure7 shows the prefetchng algorthm. whle () { f (a new wndow) { whle (for all artcles n wndow) { a = sf ( f f ) + ( sf ) a sort the artcles n LBF = f + r LBF order; prefetch the artcles that have large untl cache memory s full; Fgure 7. Artcle prefetchng algorthm 4.3 The effects of prefetchng LBF In the proposed clustered MNOD server, the prefetchng n CSVR promotes the effcency of the server system because of decreasng the overhead of retrevals from dsks n SSVR. We employ the mss rate and the concurrent users to show the effects of prefetchng. For smulaton, we used the log fles of Pusanlbo s electronc newspaper n Korea, and the parameters expressed n Table. Table. Smulaton parameters Parameter Block sze The data rate that a request requre The data sze that a request requre The bandwdth of Storage Server Values 256k M/s 3M 6M/s Fgure8 shows the relatons between the amount of prefetchng and the mss rate; as the amount of prefetchng ncreases, the mss rate decreases. Fgure also shows that as the prefetchng amount ncreases, the number of concurrent users by LBF scheme and ones only ncludng prefetchng overheads. Because prefetchng an artcle s equal to processng a request, the overheads of artcle prefetchng are calculated by multplyng the number of prefetched artcles and the overheads that a request requres. Thus the real effect of artcle prefetchng s expressed as the dfference between the number of concurrent users by LBF scheme and 8

6 the number of concurrent users only ncludng prefetchng M 2M 3M 4M 5M overhead. As a result, prefetchng artcles n CSVR ncreases the performance of MNOD server system by.5 tmes. 5. Artcle Placement Polcy 5. Prevous Data Placement mss rate Fgure 8. Mss rate accordng to prefetchng sze Cuncurrent Users C.U ncludng the overhead M M 2M 3M 4M 5M Fgure. Concurrent users accordng to prefetchng sze We ntended to ncrease user servce throughput by arrangng replca of popular data on dsks. Ths method, however, had a few problems. NOD artcle s not as bg as VOD data n sze. In MNOD system, relatvely large amount of data exsts n the cache, and the probablty that hot data s n the cache s hgher than VOD system. Naturally, dsk access probablty wll be lower relatvely. And, the number of NOD artcles s very much than VOD data, and by reason of storage overhead the number of artcles beng replcated needs to be restrcted to artcles wth hgh popularty. Thus, the data that has lower popularty but more growng probablty of dsk access wll be suffered from long servce delay because of small number of replca due to low popularty. 5.2 New Scenaro of Data Placement The new scenaro we suggest for data placement n the MNOD system s to manage varous numbers of replcas accordng to data popularty. The artcle data wth hgher popularty wll be cached by some cachng method based on data popularty and have lower probablty of dsk access. The artcle data that has lower popularty but more growng probablty of dsk access needs hgher dsk I/O bandwdth. So, more many of replcas need to be assgned for artcle data wth lower popularty. By reason of ths, we emphasze the artcle data that has relatvely lower popularty than some hot data. 5.3 Artcle Placement Algorthm 5.3. Determnng the number of artcle replca Artcle data s arranged n decreasng order accordng to ts popularty and has an dentty number of ts own that s able to be dscerned from other ones. We wll bestow the replcaton level, T, on artcle data a n order to determne the number of replca. The replcaton level T s the nverse of dfference between mean artcle popularty and popularty value of artcle data. Ths algorthm also takes advantage of standard dervaton for assgnng largest value of T to the data that has to be repeated many tmes. And, by that result, the average s shfted. We can express T as [Eq.3] by ntegratng the expressed mentons above E( X) + cs ( X) - p( x ), -5. ˆ c ˆ5. where x : dentty of artcle data px ( ) : popularty of x X : random varable of p( x ) E( X) : expected value (mean) of X s ( X) : standard devaton of X Artcle placement on dsk [Eq. 3] Artcle data conssts of several data blocks that are unts of placement. We place artcle data on the dsk usng round-robn method and the dsk where the frst block of news data s placed s selected randomly. If replcas of artcle data exst, we wll place them usng the same scheme, but we must utlze a random seed for seekng a start dsk that s not the same dsk storng orgnal artcle data, so as to secure hgher dsk I/O bandwdth from dsk access collson. Fgure shows the artcle placement algorthm proposed n ths paper and the

7 parameters used n the algorthm are expressed n Table 2. Table 2. Parameters of artcle placement polcy C The number replcas of each artcle data, x Bnews The number of blocks of each artcle data D Dsk number Tdsk Total amount of dsk S Storage capacty for dsk (the amount of dsk blocks) s The amount of block occuped on the dsk, D s = ; whle ( less than total number of news data) { whle ( less than C ) { D s ntalzed by random functon; whle (less than B ){ news whle () { f ( s <= S ) { dsk store a data block on the dsk D ; /* for usng round-robn method */ ncrease D by one ; D = D % T ; dsk ncrease s by one; break; else { /* dsk D s full */ ncrease D by one ; D = D % T ; dsk Fgure. Artcle placement algorthm Processng flow of user request If the user s request arrves at a CSVR(Control Server) n Fgure, the CSVR wll dvde the request nto several servce blocks and look nto whether the request exsts n cache memory or not wth server control nformaton. If the request exst n cache memory, the user request s served from cache memory. Otherwse, retreval requests of blocks wll be sent to the SSVR(Storage Server) havng requested blocks and be servced form dsks. In order to reduce the servce delay of user request, the SSVR wll process each block as follows: Frst, the SSVR wll look for dsks that have the block and next, compare the length of the queue of each dsk to seek the dsk that has the mnmum length of ts own queue. Fnally, the SSVR assgns the block request to the mnmum loaded dsk. Therefore, by usng ths method load balancng can be acheved among dsks, and the servce delay s decreased. Fgure s a flowchart of user request processng. Fgure. Flowchart of servce procedure for user requests 5.4 Evaluaton of artcle placement polcy 5.4. Smulaton envronment We enforced smulatons n Sun Sparc staton-2. We supposed that user requests requre Mbyte/s capactes of dsk I/O bandwdth, that the sze of data whch s compressed n MPEG II s Mbytes ( seconds), and that the number of data s restrcted to forty. One block sze of dsks s 256K and 24 blocks are processed wthn one second. The volume of a dsk s Gbyte and the number of dsks s 4 ths sze accommodates total data as twce Evaluaton of Smulaton We examned the servce falure rate (deadlne mss request/total servce request) when the popularty level beng replcated fully s changed nto the hgher level %~%, %~2%, and 2%~3%. Fgure2 shows the servce falure rate of a MNOD system wth a cache. The total servce performance s better than that of no cachng system. In ths case, artcles wth the second best popularty have the hghest dsk access probabltes because the cache ht ratos of them are lower. 2

8 Thus, mantanng the fully replcated popularty level to the hgher level %~2% makes the best performance. W D U Ã H U X O L D I Ã H F L Y U H 6 Fgure 3. Servce falure to every dsk n cachng system Fgure3 present servce falure rates of each dsk when the number of concurrent users s 6 and. The results show that the proposed artcle placement polcy controls the load balance of every dsk as well. 6. Concluson )X O\ÃUHSOLFDWHGÃSRSXODULW\ÃOHYHO Fgure 2. Servce falure to amount of users Servce falure rate $PRXQWÃRIÃXVHUV %~% () %~% (6) % % 2%~3% 2%~3% ~2%() ~2%(6) () (6) Fully replcated popularty level In ths paper we propose popularty-based artcle prefetchng and placement polcy for NOD artcle data. Because NOD data composng news artcles make a dfference to vdeo data of VOD n terms of meda type and sze, lfe cycle of artcles and frequency of clents nteracton, a new NOD artcle popularty model s needed to explan user access patterns. As a result, we show that the popularty dstrbuton of artcles s dfferent from Zpf s, whch s a popularty model of VOD data. In partcular, we propose a new artcle popularty model, whch we call Mult- Selecton Zpf dstrbuton. The proposed model s the shortterm popularty model that denotes popularty dstrbuton among artcles n an nstant tme. If the artcle prefetchng algorthm may be consdered, whle the short-term Total dsk dsk2 dsk3 dsk4 popularty model notfes what artcles are prefetched, longterm popularty model, denotng the lfe cycle of artcles, announces when the artcles are prefetched. Thus, we propose the artcle prefetchng polcy based on the popularty and lfe cycle model of NOD artcles for ncreasng performance n a MNOD system. And, we propose the data placement polcy for NOD data that consders both artcle popularty and dsk access probablty. In ths polcy, we replcate data that have not best popularty, but hgh dsk access probablty because the probablty that the best popular data s n cache memory s hgh and relatvely t's dsk access probablty s low. And, orgnal data and replca are placed to mnmum loaded dsks to acheve load balancng among dsks. The proposed artcle placement polcy decreases the servce falure rate and ncreases the servce throughput by way of load balancng among dsks. Our future work s to develop a mathematcal model of Mult-Selecton Zpf dstrbuton and to nvestgate artcle placement and prefetchng algorthms n clustered archtecture envronment. Reference [] Lous P. Slothouber, "A Model of Web Server Performance", modelpaper.html, 6. [2] Odysseas I. Pentakalos, Danel A. Menascé, "An Approxmate Performance Model of a Untree Mass Storage System", 4 th IEEE Symposum on Mass Storage Systems, pp , Sep.5. [3] J. Alemany, J. S. Thathacher, Random Strpng for News on Demand Servers, Techncal Report UW-CSE , Unv. Washngton, 7. [4] Y. U. Park, G. H. Back, W. I. Seo, Y.J. Km, K. D. Chung, New Buffer polcy for NOD data, Conference on Korea Broadcast Engneerng, Korea, 7. [5] A. Dan, M. Kenzle, D. Staram, A dynamc polcy of segment replcaton for load balancng n vdeo-ondemand servers, Multmeda System 3:3-3, 5. [6] T. D. C. Lttle, D. Venkatesh, Populary - based assgnment of moves to storage devces n vdeo-ondemand system, Multmeda System 2:28-287, 5. [7] S. Berson, R. Muntz, S. Ghandeharzadeh, X. Ju, Staggered Strpng n Multmeda Informaton Systems, SIGMOD Conference, pp. 7-, 4. [8] R. Flynn, W. Tetzlaff, Dsk Strpng and Block Replcaton Algorthm For Vdeo Fle Servers, IEEE 3 rd Internatonal Conference on Multmeda Computng Systems, 6. [] J. E. Ptkow, M. M. Recker, A Smple Yet Robust Cachng Algorthm Based on Dynamc Access Patterns, 2

9 In the Proceedng of the Second Internatonal WWW Conference, 7. [] S. A. Barnett, G. J. Ando, H. W. Beadle, Cachng Polces n a Dstrbuted Vdeo-on-Demand System, In the Proceedng of Australan Telecommuncaton Networks and Applcatons Conference, 5. [] C. Grwodz, M. Bar, L. C. Wolf, Long-term Move Popularty Models n Vdeo-on-Demand Systems or The Lfe of an on-demand Move, The ffth ACM Internatonal Multmeda Conference, 7. [2] I. Tatarnov, V. Solovev, A. Rousskov, Statc Cachng n Web Servers, submtted to the 6 th nternatonal Conference on Computer Communcatons and Networks. [3] Lee D., Cho J., J. Km, S. H. Noh, S. L. Mn, Y. Cho, C. S. Km, LRFU Replacement Polcy: A spectrum of block replacement polces, In Seoul Natonal Unversty Techncal Report SNU-CE-AN-6-4, Korea, 6. [4] Renu Tewar, Harrck M. Vn, Ast Dan & Dnkar Staram, Resource Based Cachng for Web Servers, Proceedngs of SPIE/ACM Conference on Multmeda Computng and Network, 8. [5] Renu Tewar, Rrajat Mukherjee, Danel M. Das, Harrck M. Vn, "Real-Tme Issues for Clustered Multmeda Servers", IBM Research Center, Techncal Report, 5. 22

Clustered Multimedia NOD : Popularity-Based Article Prefetching and Placement

Clustered Multimedia NOD : Popularity-Based Article Prefetching and Placement Clustered Multimedia NOD : Popularity-Based Article Prefetching and Placement Y.J.Kim, T.U.Choi, K.O.Jung, Y.K.Kang, S.H.Park, Ki-Dong Chung Department of Computer Science, Pusan National University, Korea

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

Efficient Distributed File System (EDFS)

Efficient Distributed File System (EDFS) Effcent Dstrbuted Fle System (EDFS) (Sem-Centralzed) Debessay(Debsh) Fesehaye, Rahul Malk & Klara Naherstedt Unversty of Illnos-Urbana Champagn Contents Problem Statement, Related Work, EDFS Desgn Rate

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

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

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

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 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

Cluster Analysis of Electrical Behavior

Cluster Analysis of Electrical Behavior Journal of Computer and Communcatons, 205, 3, 88-93 Publshed Onlne May 205 n ScRes. http://www.scrp.org/ournal/cc http://dx.do.org/0.4236/cc.205.350 Cluster Analyss of Electrcal Behavor Ln Lu Ln Lu, School

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

Analysis of Collaborative Distributed Admission Control in x Networks

Analysis of Collaborative Distributed Admission Control in x Networks 1 Analyss of Collaboratve Dstrbuted Admsson Control n 82.11x Networks Thnh Nguyen, Member, IEEE, Ken Nguyen, Member, IEEE, Lnha He, Member, IEEE, Abstract Wth the recent surge of wreless home networks,

More information

Virtual Machine Migration based on Trust Measurement of Computer Node

Virtual Machine Migration based on Trust Measurement of Computer Node Appled Mechancs and Materals Onlne: 2014-04-04 ISSN: 1662-7482, Vols. 536-537, pp 678-682 do:10.4028/www.scentfc.net/amm.536-537.678 2014 Trans Tech Publcatons, Swtzerland Vrtual Machne Mgraton based on

More information

Real-Time Guarantees. Traffic Characteristics. Flow Control

Real-Time Guarantees. Traffic Characteristics. Flow Control Real-Tme Guarantees Requrements on RT communcaton protocols: delay (response s) small jtter small throughput hgh error detecton at recever (and sender) small error detecton latency no thrashng under peak

More information

SRB: Shared Running Buffers in Proxy to Exploit Memory Locality of Multiple Streaming Media Sessions

SRB: Shared Running Buffers in Proxy to Exploit Memory Locality of Multiple Streaming Media Sessions SRB: Shared Runnng Buffers n Proxy to Explot Memory Localty of Multple Streamng Meda Sessons Songqng Chen,BoShen, Yong Yan, Sujoy Basu, and Xaodong Zhang Department of Computer Scence Moble and Meda System

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

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

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

Efficient Broadcast Disks Program Construction in Asymmetric Communication Environments

Efficient Broadcast Disks Program Construction in Asymmetric Communication Environments Effcent Broadcast Dsks Program Constructon n Asymmetrc Communcaton Envronments Eleftheros Takas, Stefanos Ougaroglou, Petros copoltds Department of Informatcs, Arstotle Unversty of Thessalonk Box 888,

More information

Priority-Based Scheduling Algorithm for Downlink Traffics in IEEE Networks

Priority-Based Scheduling Algorithm for Downlink Traffics in IEEE Networks Prorty-Based Schedulng Algorthm for Downlnk Traffcs n IEEE 80.6 Networks Ja-Mng Lang, Jen-Jee Chen, You-Chun Wang, Yu-Chee Tseng, and Bao-Shuh P. Ln Department of Computer Scence Natonal Chao-Tung Unversty,

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

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers IOSR Journal of Electroncs and Communcaton Engneerng (IOSR-JECE) e-issn: 78-834,p- ISSN: 78-8735.Volume 9, Issue, Ver. IV (Mar - Apr. 04), PP 0-07 Content Based Image Retreval Usng -D Dscrete Wavelet wth

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

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

An Efficient Garbage Collection for Flash Memory-Based Virtual Memory Systems

An Efficient Garbage Collection for Flash Memory-Based Virtual Memory Systems S. J and D. Shn: An Effcent Garbage Collecton for Flash Memory-Based Vrtual Memory Systems 2355 An Effcent Garbage Collecton for Flash Memory-Based Vrtual Memory Systems Seunggu J and Dongkun Shn, Member,

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

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

Cache Performance 3/28/17. Agenda. Cache Abstraction and Metrics. Direct-Mapped Cache: Placement and Access

Cache Performance 3/28/17. Agenda. Cache Abstraction and Metrics. Direct-Mapped Cache: Placement and Access Agenda Cache Performance Samra Khan March 28, 217 Revew from last lecture Cache access Assocatvty Replacement Cache Performance Cache Abstracton and Metrcs Address Tag Store (s the address n the cache?

More information

FRES-CAR: An Adaptive Cache Replacement Policy

FRES-CAR: An Adaptive Cache Replacement Policy FRES-CAR: An Adaptve Cache Replacement Polcy George Palls, Athena Vaal, Eythms Sdropoulos Department of Informatcs Arstotle Unversty of Thessalon, 54124, Thessalon, Greece gpalls@ccf.auth.gr, {avaal, eythms}@csd.auth.gr

More information

FAHP and Modified GRA Based Network Selection in Heterogeneous Wireless Networks

FAHP and Modified GRA Based Network Selection in Heterogeneous Wireless Networks 2017 2nd Internatonal Semnar on Appled Physcs, Optoelectroncs and Photoncs (APOP 2017) ISBN: 978-1-60595-522-3 FAHP and Modfed GRA Based Network Selecton n Heterogeneous Wreless Networks Xaohan DU, Zhqng

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

FEATURE EXTRACTION. Dr. K.Vijayarekha. Associate Dean School of Electrical and Electronics Engineering SASTRA University, Thanjavur

FEATURE EXTRACTION. Dr. K.Vijayarekha. Associate Dean School of Electrical and Electronics Engineering SASTRA University, Thanjavur FEATURE EXTRACTION Dr. K.Vjayarekha Assocate Dean School of Electrcal and Electroncs Engneerng SASTRA Unversty, Thanjavur613 41 Jont Intatve of IITs and IISc Funded by MHRD Page 1 of 8 Table of Contents

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

A Traffic Aware Routing Protocol for Congestion Avoidance in Content-Centric Network

A Traffic Aware Routing Protocol for Congestion Avoidance in Content-Centric Network , pp.69-80 http://dx.do.org/10.14257/jmue.2014.9.9.08 A Traffc Aware Routng Protocol for Congeston Avodance n Content-Centrc Network Jung-Jae Km 1, Mn-Woo Ryu 2*, S-Ho Cha 3 and Kuk-Hyun Cho 1 1 Dept.

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 Topology-aware Random Walk

A Topology-aware Random Walk A Topology-aware Random Walk Inkwan Yu, Rchard Newman Dept. of CISE, Unversty of Florda, Ganesvlle, Florda, USA Abstract When a graph can be decomposed nto clusters of well connected subgraphs, t s possble

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

Concurrent Apriori Data Mining Algorithms

Concurrent Apriori Data Mining Algorithms Concurrent Apror Data Mnng Algorthms Vassl Halatchev Department of Electrcal Engneerng and Computer Scence York Unversty, Toronto October 8, 2015 Outlne Why t s mportant Introducton to Assocaton Rule Mnng

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

Research Article Efficient Load Balancing Using Active Replica Management in a Storage System

Research Article Efficient Load Balancing Using Active Replica Management in a Storage System Mathematcal Problems n Engneerng Volume 2016, Artcle ID 4751829, 9 pages http://dx.do.org/10.1155/2016/4751829 Research Artcle Effcent Load Balancng Usng Actve Replca Management n a Storage System Ju-Pn

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

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision SLAM Summer School 2006 Practcal 2: SLAM usng Monocular Vson Javer Cvera, Unversty of Zaragoza Andrew J. Davson, Imperal College London J.M.M Montel, Unversty of Zaragoza. josemar@unzar.es, jcvera@unzar.es,

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

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

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 KIND OF ROUTING MODEL IN PEER-TO-PEER NETWORK BASED ON SUCCESSFUL ACCESSING RATE

A KIND OF ROUTING MODEL IN PEER-TO-PEER NETWORK BASED ON SUCCESSFUL ACCESSING RATE A KIND OF ROUTING MODEL IN PEER-TO-PEER NETWORK BASED ON SUCCESSFUL ACCESSING RATE 1 TAO LIU, 2 JI-JUN XU 1 College of Informaton Scence and Technology, Zhengzhou Normal Unversty, Chna 2 School of Mathematcs

More information

Supporting Consistency Management in Dynamic Content Distribution Overlays

Supporting Consistency Management in Dynamic Content Distribution Overlays Supportng Consstency Management n Dynamc Content Dstrbuton Overlays Zhou Su, Jro Katto, Yasuhko Yasuda Faculty of Scence and engneerng, Waseda Unversty Contact E-mal: suzhou@waseda.p Abstract Content Dstrbuton

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

Tsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance

Tsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance Tsnghua Unversty at TAC 2009: Summarzng Mult-documents by Informaton Dstance Chong Long, Mnle Huang, Xaoyan Zhu State Key Laboratory of Intellgent Technology and Systems, Tsnghua Natonal Laboratory for

More information

Support Strong Consistency for Mobile Dynamic Contents Delivery Network

Support Strong Consistency for Mobile Dynamic Contents Delivery Network Nnth IEEE Internatonal Symposum on Multmeda 27 Support Strong Consstency for Moble Dynamc Contents Delvery Networ Zhou Su, Jro Katto and Yasuho Yasuda Graduate School of Scence and Engneerng, Waseda Unversty,

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

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

A Semi-Distributed Load Balancing Architecture and Algorithm for Heterogeneous Wireless Networks

A Semi-Distributed Load Balancing Architecture and Algorithm for Heterogeneous Wireless Networks A Sem-Dstrbuted oad Balancng Archtecture and Algorthm for Heterogeneous reless Networks Md. Golam Rabul Ala Choong Seon Hong * Kyung Hee Unversty, Korea rob@networkng.khu.ac.kr, cshong@khu.ac.kr Abstract

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

A Load-balancing and Energy-aware Clustering Algorithm in Wireless Ad-hoc Networks

A Load-balancing and Energy-aware Clustering Algorithm in Wireless Ad-hoc Networks A Load-balancng and Energy-aware Clusterng Algorthm n Wreless Ad-hoc Networks Wang Jn, Shu Le, Jnsung Cho, Young-Koo Lee, Sungyoung Lee, Yonl Zhong Department of Computer Engneerng Kyung Hee Unversty,

More information

Adaptive Power-Aware Prefetch in Wireless Networks

Adaptive Power-Aware Prefetch in Wireless Networks Adaptve Power-Aware Prefetch n Wreless Networks Langzhong Yn, Student Member, IEEE, and Guohong Cao, Member, IEEE Abstract Most of the prefetch technques used n the current cache management schemes do

More information

Determining the Optimal Bandwidth Based on Multi-criterion Fusion

Determining the Optimal Bandwidth Based on Multi-criterion Fusion Proceedngs of 01 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 5 (01) (01) IACSIT Press, Sngapore Determnng the Optmal Bandwdth Based on Mult-crteron Fuson Ha-L Lang 1+, Xan-Mn

More information

Quantifying Performance Models

Quantifying Performance Models Quantfyng Performance Models Prof. Danel A. Menascé Department of Computer Scence George Mason Unversty www.cs.gmu.edu/faculty/menasce.html 1 Copyrght Notce Most of the fgures n ths set of sldes come from

More information

Strategic Prefetching of VoD Programs Based on ART2 driven Request Clustering

Strategic Prefetching of VoD Programs Based on ART2 driven Request Clustering Strategc Prefetchng of VoD Programs Based on ART2 drven Request Clusterng Dr. T. R. Gopalakrshnan Nar 1 and P. Jayarekha 2 Drector, Research and Industry Incubaton Centre, DSI, Bangalore trgnar@yahoo.com

More information

BioTechnology. An Indian Journal FULL PAPER. Trade Science Inc.

BioTechnology. An Indian Journal FULL PAPER. Trade Science Inc. [Type text] [Type text] [Type text] ISSN : 0974-74 Volume 0 Issue BoTechnology 04 An Indan Journal FULL PAPER BTAIJ 0() 04 [684-689] Revew on Chna s sports ndustry fnancng market based on market -orented

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

A New Approach For the Ranking of Fuzzy Sets With Different Heights

A New Approach For the Ranking of Fuzzy Sets With Different Heights New pproach For the ankng of Fuzzy Sets Wth Dfferent Heghts Pushpnder Sngh School of Mathematcs Computer pplcatons Thapar Unversty, Patala-7 00 Inda pushpndersnl@gmalcom STCT ankng of fuzzy sets plays

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

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

Evaluation of an Enhanced Scheme for High-level Nested Network Mobility

Evaluation of an Enhanced Scheme for High-level Nested Network Mobility IJCSNS Internatonal Journal of Computer Scence and Network Securty, VOL.15 No.10, October 2015 1 Evaluaton of an Enhanced Scheme for Hgh-level Nested Network Moblty Mohammed Babker Al Mohammed, Asha Hassan.

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

Goals and Approach Type of Resources Allocation Models Shared Non-shared Not in this Lecture In this Lecture

Goals and Approach Type of Resources Allocation Models Shared Non-shared Not in this Lecture In this Lecture Goals and Approach CS 194: Dstrbuted Systems Resource Allocaton Goal: acheve predcable performances Three steps: 1) Estmate applcaton s resource needs (not n ths lecture) 2) Admsson control 3) Resource

More information

Journal of Chemical and Pharmaceutical Research, 2014, 6(10): Research Article. Study on the original page oriented load balancing strategy

Journal of Chemical and Pharmaceutical Research, 2014, 6(10): Research Article. Study on the original page oriented load balancing strategy Avalable onlne www.jocpr.com Journal of hemcal and Pharmaceutcal Research, 2014, 6(10):274-280 Research Artcle IN : 0975-7384 ODEN(UA) : JPR5 tudy on the orgnal page orented load balancng strategy Kunpeng

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

Adaptive Energy and Location Aware Routing in Wireless Sensor Network

Adaptive Energy and Location Aware Routing in Wireless Sensor Network Adaptve Energy and Locaton Aware Routng n Wreless Sensor Network Hong Fu 1,1, Xaomng Wang 1, Yngshu L 1 Department of Computer Scence, Shaanx Normal Unversty, X an, Chna, 71006 fuhong433@gmal.com {wangxmsnnu@hotmal.cn}

More information

Comparison of Function Based Web Proxy Cache Replacement Strategies

Comparison of Function Based Web Proxy Cache Replacement Strategies Comparson of Functon Based Web Proxy Cache Replacement trateges Hala ElAarag and am Romano Department of Mathematcs and Computer cence tetson Unversty 421 N Woodland Blvd, Deland, FL 32723 {sromano,helaarag}@stetson.edu

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

Use of Genetic Algorithms in Efficient Scheduling for Multi Service Classes

Use of Genetic Algorithms in Efficient Scheduling for Multi Service Classes Use of Genetc Algorthms n Effcent Schedulng for Mult Servce Classes Shyamale Thlakawardana and Rahm Tafazoll Centre for Communcatons Systems Research (CCSR), Unversty of Surrey, Guldford, GU27XH, UK Abstract

More information

Efficient Load-Balanced IP Routing Scheme Based on Shortest Paths in Hose Model. Eiji Oki May 28, 2009 The University of Electro-Communications

Efficient Load-Balanced IP Routing Scheme Based on Shortest Paths in Hose Model. Eiji Oki May 28, 2009 The University of Electro-Communications Effcent Loa-Balance IP Routng Scheme Base on Shortest Paths n Hose Moel E Ok May 28, 2009 The Unversty of Electro-Communcatons Ok Lab. Semnar, May 28, 2009 1 Outlne Backgroun on IP routng IP routng strategy

More information

Cell Count Method on a Network with SANET

Cell Count Method on a Network with SANET CSIS Dscusson Paper No.59 Cell Count Method on a Network wth SANET Atsuyuk Okabe* and Shno Shode** Center for Spatal Informaton Scence, Unversty of Tokyo 7-3-1, Hongo, Bunkyo-ku, Tokyo 113-8656, Japan

More information

AN INDEXING METHOD FOR SUPPORTING SPATIAL QUERIES IN STRUCTURED PEER-TO-PEER SYSTEMS

AN INDEXING METHOD FOR SUPPORTING SPATIAL QUERIES IN STRUCTURED PEER-TO-PEER SYSTEMS AN INDEXING METHOD FOR SUPPORTING SPATIAL QUERIES IN STRUCTURED PEER-TO-PEER SYSTEMS Lngku Meng a, Wenun Xe a, *, Dan Lu a, b a School of Remote Sensng and Informaton Engneerng, Wuhan Unversty, Wuhan 4379,

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

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

Scalable Download Protocols

Scalable Download Protocols Scalable Download Protocols A Thess Submtted to the College of Graduate Studes and Research In Partal Fulfllment of the Requrements For the Degree of Doctor of Phlosophy In the Department of Computer Scence

More information

CS1100 Introduction to Programming

CS1100 Introduction to Programming Factoral (n) Recursve Program fact(n) = n*fact(n-) CS00 Introducton to Programmng Recurson and Sortng Madhu Mutyam Department of Computer Scence and Engneerng Indan Insttute of Technology Madras nt fact

More information

An Improved Image Segmentation Algorithm Based on the Otsu Method

An Improved Image Segmentation Algorithm Based on the Otsu Method 3th ACIS Internatonal Conference on Software Engneerng, Artfcal Intellgence, Networkng arallel/dstrbuted Computng An Improved Image Segmentaton Algorthm Based on the Otsu Method Mengxng Huang, enjao Yu,

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

A Deflected Grid-based Algorithm for Clustering Analysis

A Deflected Grid-based Algorithm for Clustering Analysis A Deflected Grd-based Algorthm for Clusterng Analyss NANCY P. LIN, CHUNG-I CHANG, HAO-EN CHUEH, HUNG-JEN CHEN, WEI-HUA HAO Department of Computer Scence and Informaton Engneerng Tamkang Unversty 5 Yng-chuan

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

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

RAP. Speed/RAP/CODA. Real-time Systems. Modeling the sensor networks. Real-time Systems. Modeling the sensor networks. Real-time systems:

RAP. Speed/RAP/CODA. Real-time Systems. Modeling the sensor networks. Real-time Systems. Modeling the sensor networks. Real-time systems: Speed/RAP/CODA Presented by Octav Chpara Real-tme Systems Many wreless sensor network applcatons requre real-tme support Survellance and trackng Border patrol Fre fghtng Real-tme systems: Hard real-tme:

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

Analysis of Continuous Beams in General

Analysis of Continuous Beams in General Analyss of Contnuous Beams n General Contnuous beams consdered here are prsmatc, rgdly connected to each beam segment and supported at varous ponts along the beam. onts are selected at ponts of support,

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

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

USING GRAPHING SKILLS

USING GRAPHING SKILLS Name: BOLOGY: Date: _ Class: USNG GRAPHNG SKLLS NTRODUCTON: Recorded data can be plotted on a graph. A graph s a pctoral representaton of nformaton recorded n a data table. t s used to show a relatonshp

More information

Design of the Application-Level Protocol for Synchronized Multimedia Sessions

Design of the Application-Level Protocol for Synchronized Multimedia Sessions Desgn of the Applcaton-Level Protocol for Synchronzed Multmeda Sessons Chun-Chuan Yang Multmeda and Communcatons Laboratory Department of Computer Scence and Informaton Engneerng Natonal Ch Nan Unversty,

More information

CS 268: Lecture 8 Router Support for Congestion Control

CS 268: Lecture 8 Router Support for Congestion Control CS 268: Lecture 8 Router Support for Congeston Control Ion Stoca Computer Scence Dvson Department of Electrcal Engneerng and Computer Scences Unversty of Calforna, Berkeley Berkeley, CA 9472-1776 Router

More information

Overview. Basic Setup [9] Motivation and Tasks. Modularization 2008/2/20 IMPROVED COVERAGE CONTROL USING ONLY LOCAL INFORMATION

Overview. Basic Setup [9] Motivation and Tasks. Modularization 2008/2/20 IMPROVED COVERAGE CONTROL USING ONLY LOCAL INFORMATION Overvew 2 IMPROVED COVERAGE CONTROL USING ONLY LOCAL INFORMATION Introducton Mult- Smulator MASIM Theoretcal Work and Smulaton Results Concluson Jay Wagenpfel, Adran Trachte Motvaton and Tasks Basc Setup

More information

Dynamic Bandwidth Provisioning with Fairness and Revenue Considerations for Broadband Wireless Communication

Dynamic Bandwidth Provisioning with Fairness and Revenue Considerations for Broadband Wireless Communication Ths full text paper was peer revewed at the drecton of IEEE Communcatons Socety subject matter experts for publcaton n the ICC 008 proceedngs. Dynamc Bandwdth Provsonng wth Farness and Revenue Consderatons

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

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

Application of VCG in Replica Placement Strategy of Cloud Storage

Application of VCG in Replica Placement Strategy of Cloud Storage Internatonal Journal of Grd and Dstrbuted Computng, pp.27-40 http://dx.do.org/10.14257/jgdc.2016.9.4.03 Applcaton of VCG n Replca Placement Strategy of Cloud Storage Wang Hongxa Computer Department, Bejng

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

On the Exact Analysis of Bluetooth Scheduling Algorithms

On the Exact Analysis of Bluetooth Scheduling Algorithms On the Exact Analyss of Bluetooth Schedulng Algorth Gl Zussman Dept. of Electrcal Engneerng Technon IIT Hafa 3000, Israel glz@tx.technon.ac.l Ur Yechal Dept. of Statstcs and Operatons Research School of

More information

Intelligent Traffic Conditioners for Assured Forwarding Based Differentiated Services Networks 1

Intelligent Traffic Conditioners for Assured Forwarding Based Differentiated Services Networks 1 Intellgent Traffc Condtoners for Assured Forwardng Based Dfferentated Servces Networks B. Nandy, N. Seddgh, P. Peda, J. Ethrdge Nortel Networks, Ottawa, Canada Emal:{bnandy, nseddgh, ppeda, jethrdg}@nortelnetworks.com

More information