Collaborative Caching in Wireless Video Streaming Through Resource Auctions

Size: px
Start display at page:

Download "Collaborative Caching in Wireless Video Streaming Through Resource Auctions"

Transcription

1 458 IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, VOL. 3, NO. 2, FEBRUARY 22 Collaboratve Cachng n Wreless Vdeo Streamng Through Resource Auctons Je Da, Student Member, IEEE, Fangmng Lu, Member, IEEE, Bo L, Fellow, IEEE, Baochun L, Senor Member, IEEE, and Jangchuan Lu, Senor Member, IEEE Abstract Recent advances n wreless communcatons and moble networng have dramatcally ncreased the popularty of multmeda servces for moble users, wth wreless vdeo streamng at ther fngertps. To facltate effcent acquston of vdeo content, proxy cachng has been wdely used by wreless servce provders (WSPs), whch typcally deploy cache servers at moble swtchng centers (MSCs). However, capacty provsonng of cache servers s challengng, gven the dynamc user demands and the lmted cache server resources. Wth ncreased denstes of wreless servce deployment, t s ncreasngly common that moble users are covered by more than one WSP wthn an area. Ths brngs opportuntes of a collaboratve cachng paradgm among the cache servers deployed at dfferent MSCs. In ths paper, we explore the benefts of collaboratve cachng n wreless streamng servces, addressng both challenges of ncentves and truthfulness of selfsh WSPs. We propose a collaboratve mechansm that maxmzes the socal welfare n the context of Vcrey- Clare-Groves (VCG) auctons, n whch cache servers cooperate n the tradng of ther resources n a self-enforcng manner. Expermental results demonstrate that superor performance can be acheved wth respect to the qualty of vdeo streamng. Index Terms Collaboratve Cachng, VCG Aucton, Incentve Engneerng, Truthfulness. I. INTRODUCTION The popularty of on-demand vdeo streamng has substantally changed the landscape of wreless multmeda applcatons, wth an ncreasng number of content provders, such as YouTube and Netflx, offerng streamng vdeo content to moble users. Meanwhle, a dverse range of wreless access technologes, ncludng HSPA+ cellular access, Femtocells, and recent advances n 4G deployment such as LTE [], have made broadband wreless connectons a near-term realty. Thans to the avalablty of last-mle wreless bandwdth, wreless servce provders (WSPs) have ncreasngly focused on the qualty of wreless vdeo streamng, wth proxy cachng Manuscrpt receved 5 February 2; revsed 2 July 2. The research was support n part by a grant from Huawe Technologes Co. Ltd. under the contract HUAW8-5L8/PN, by a grant from HKUST under the contract RPCEG29, by a grant from The Natonal Natural Scence Foundaton of Chna (NSFC) under grant No J. Da and B. L are wth the Department of Computer Scence and Engneerng, Hong Kong Unversty of Scence and Technology, Clear Water Bay, Kowloon, Hong Kong (e-mal: {jda, bl}@cse.ust.h). F. Lu s wth the Servces Computng Technology and System Lab, Cluster and Grd Computng Lab n the School of Computer Scence and Technology, Huazhong Unversty of Scence and Technology (e-mal: fmlu@mal.hust.edu.cn). B. L s wth the Department of Electrcal and Computer Engneerng, Unversty of Toronto (e-mal: bl@eecg.toronto.edu). J. Lu s wth the School of Computng Scence, Smon Fraser Unversty (e-mal: jclu@cs.sfu.ca). Dgtal Object Identfer.9/JSAC /2/$25. c 22 IEEE beng extensvely utlzed to mprove the streamng qualty [2]. Cache servers have been deployed n Moble Swtchng Centers (MSCs) [3] of WSPs, so as to locate vdeo content closer to end users. For nstance, Verzon has announced ts networ optmzaton deployment [4], n whch vdeo cachng s appled to mprove the experence of wreless users. Nevertheless, the nherent dynamcs of moble users have presented a dauntng challenge to resource provsonng at these cache servers. Moble users frequently jon and leave streamng sessons n the presence of moblty, and consequently the load on deployed cache servers becomes unpredctable. The ntutve soluton of over-provsonng resources would lead to nevtable but unnecessary waste of resources, gven the bursty nature of aggregated demand. Consderng the fact that moble users are often covered by multple WSPs n the same area, all of whch delver vdeo content to ther own clents, a better way to offload ther cache servers s to ncentvze these autonomous and selfsh WSPs to collaborate wth one another. The cache servers belongng to dfferent WSPs may assst one another to stream vdeo content to end users by usng ther own bandwdth and storage resources. Such a collaboraton would mprove the degree of multplexng avalable resources n WSPs, as well as the general avalablty of vdeo content n proxy servers. Exstng wors n the lterature have not demonstrated how these ndependent cache resources owned by WSPs can be organzed n a collaboratve way, so as to acheve more effcent multplexng of resources. The man challenge s the nherently selfsh nature of autonomous WSPs, often belongng to dfferent for-proft corporatons. Though WSPs are ratonal n ther strategc behavor, they wll not cooperate wth others f ther gans do not outwegh ther costs. In ths paper, we focus on engneerng the ncentves to promote and encourage the cache servers owned by dfferent WSPs to truthfully cooperate wth one another. Inspred by the theory of Vcrey-Clare-Groves (VCG) auctons [5], we desgn and analyze a new collaboratve cachng mechansm that can be employed by co-locatng WSPs. VCG auctons are nown as non-cooperatve games, n whch any decson about cooperaton s self-enforced. We treat server bandwdth as commodtes n these auctons, wth dfferent valuatons based on the dynamcs of the streamng systems. Vrtual payments assocated wth the valuaton of bandwdth resources ensure that the contrbuton of cache servers s acnowledged by other partcpants. Truthfulness s also guaranteed whch enables cache servers to fathfully reveal ther true valuaton when

2 DAI et al.: COLLABORATIVE CACHING IN WIRELESS VIDEO STREAMING THROUGH RESOURCE AUCTIONS 459 bds are submtted. Our bddng strategy s far smpler than conventonal cooperatve networng mechansms that requre sophstcated strateges to coordnate the behavor across partcpants. Our smulaton results further show that the performance of streamng systems can be notceably mproved by maxmzng the socal welfare n the auctons, n whch the bandwdth unts are used to serve more valuable demands. The remander of the paper s organzed as follows. In Sec. II, we dscuss our contrbuton n the context of related wors. In Sec. III, we formulate the basc model of resource auctons n collaboratve cachng. In Sec. IV, we present the desgn of VCG-based bandwdth resource tradng mechansms. Sec. V dscusses how storage resources are utlzed n our proposed mechansms. Sec. VI proceeds to present an extensve smulaton study to evaluate the effectveness of collaboratve cachng. Fnally, we conclude the paper n Sec. VII. II. RELATED WORK Proxy cachng has long been utlzed n wreless streamng to mprove ts performance. Zhang et al. [2] present a cost-based cache replacement algorthm for a sngle cache server and a server selecton algorthm for multple cache servers n wreless multmeda proxy cachng. They further analyze the Qualty of Servce (QoS) requrement n wreless streamng whle ntroducng the correspondng QoS-adaptve proxy cachng mechansms [6]. Tan et al. [7] ntroduce a smart cachng desgn that duplcates detected vdeo content at access ponts, to effectvely reduce any redundant traffc n the WLAN whle mprovng the response delay of vdeo streamng. Our wor dffers from these studes by mplementng a collaboraton mechansm among cache servers. Our collaboratve cachng encourages the cache servers to actvely delver content across dfferent domans, whle both farness and truthfulness are guaranteed. The potental beneft of collaboratve cachng n vdeo streamng systems has been dscussed n our prevous wor [8] and several other related wors [9] []. Chen et al. [9] consder a collaboratve cachng mechansm n an Internet Protocol Televson (IPTV) system wth a herarchcal archtecture. They suggest that central offces wth lmted storage space should cooperatvely exchange vdeo content that s temporarly unavalable. Borst et al. [] develop dstrbuted cachng algorthms that am to maxmze the traffc volume served from a cache, whle mnmzng bandwdth costs. Ther target applcaton scenaro s a vdeo-on-demand (VoD) system wth tree topologes. These exstng studes generally assume a sngle authorty that owns all of the exstng resources; the cache servers could then spontaneously cooperate wth one another under certan sophstcated and enforced regulatons. In contrast, we consder the more practcal scenaro where cache servers are dstrbuted over multple domans. They may be arbtrarly connected and exhbt selfsh behavor to maxmze ther own benefts, whch presents sgnfcant challenges to the desgn of effcent and truthful cooperaton. Game theoretc models have also been appled n vdeo streamng systems to provde ncentves among cache servers. Fg.. MSC a a WSP Doman Moble User c MSC b MSC Base Staton User Dynamc An llustratve example of collaboratve cachng. Ip et al. [] propose an ncentve scheme to motvate cache servers to share ther caches. The revenue s rewarded based on the amount of cache spaces provded and on a prce functon. In contrast, rather than solely consderng byte ht rates and storage costs [] that are more sutable for tradtonal fle cachng systems, our collaboratve cachng desgn focuses on mprovng the qualty of wreless streamng qualty, wth a partcular emphass on bandwdth provsonng. III. RESOURCE AUCTIONS IN COLLABORATIVE CACHING We start wth an llustratve example to descrbe basc prncples n our collaboratve cachng mechansms for WSPs, n order to acheve better socal effcency. A. An Example and Basc Features In wreless vdeo streamng, cache servers are deployed by WSPs to mprove the overall system performance. Each cache server utlzes ts constraned bandwdth and storage capacty to serve vdeo content requests from moble users, whch consequently results n better streamng qualty for a group of vdeo programs, also referred to as channels. In conventonal cachng deployed by each ndvdual WSP, these resources are always allocated to local users wthn ts own doman. We refer to ths as ndependent cachng n ths paper. Our proposed collaboratve cachng mechansm further explores the cooperaton among cache servers across dfferent admnstratve domans to mprove the streamng qualty. We present a scenaro n Fg. to descrbe how such a collaboraton can be acheved. WSP a and WSP b deploy cache servers at ther own MSCs whch are connected to base statons. A server n doman b only has a lmted storage capacty. Wth ndependent cachng, requests to unavalable content n b have to be relayed to remote servers. However, ths problem can be allevated through collaboraton across dfferent domans. For nstance, after recevng content requests as ndcated by arrow, the cache server wth avalable content n doman a can allocate certan bandwdth to fulfll requests from moble users belongng to doman b. Intutvely, such collaboraton helps allevate the sudden request surge n a doman often encountered when there s a popular channel. As such, the dynamcs of the overall vdeo streamng system can be smoothly dspersed across multple WSPs. The effcent

3 46 IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, VOL. 3, NO. 2, FEBRUARY 22 utlzaton of bandwdth resources across cache servers s crtcal n dealng wth the dynamcs of the system. Despte the potental benefts, there s a lac of a systematc and practcal coordnaton mechansm among cache servers deployed by dfferent WSPs. Gven the autonomous nature of WSPs, one dstnctve feature s the selfshness among deployed cache servers. Ths mples that, () cache servers wthn one WSP are prmarly concerned wth the potental benefts through collaboratons. Thus, there needs to be a proper ncentve mechansm for them to collaborate. (2) selfsh WSPs may not fathfully conform to regulatons of the collaboraton f dshonest behavor can beneft them. Therefore, the truthfulness that forces WSPs to reveal ther true nformaton s also crucal n the desgn. (3) gven the dstrbuted nature of the content dstrbuton, a fully decentralzed coordnaton s requred. Our prmary objectve s to desgn such a collaboratve cachng mechansm that mproves the socal welfare, whle guaranteeng both ncentves and truthfulness. B. Model Formulaton We consder a WSP set I = {, 2,..., N } n collaboratve cachng wth a total of N WSPs that provde streamng servces n the same regon. We ntegrate the servers belongng to the same WSP nto one cache server for the sae of smplcty. The notaton Ican be used nterchangeably to represent WSP or the cache server of WSP. Let the set of vdeo channels be denoted by K = {, 2,..., N },where N denotes the total number of channels that are avalable over multple domans. The bandwdth and storage capacty at server are denoted by W and S. The VCG aucton from game theory fts the need to desgn our collaboraton mechansm such that, () each buyer has to pay after each trade, whch serves as the ncentve to contrbute. (2) the payment method of the VCG aucton ensures truthfulness, n whch buyers are wllng to truthfully reveal ther bddng nformaton n the aucton. (3) auctons can be organzed n a fully decentralzed fashon. We now present the basc model and mechansm desgn nspred by the VCG aucton theory, n order to acheve effcent allocatons that beneft users of dfferent WSPs. When aucton s appled n the context of collaboratve cachng, aucton partcpants become cache servers, wth bandwdth resources beng commodtes. Each cache server acts as both a seller or bdder that trades resources n multple rounds. When actng as a seller, cache server provdes bandwdth W r to remote bdders. We defne the bandwdth unt w as the basc unt to be traded n VCG auctons. Therefore, W r has been dvded nto W r /w homogeneous bandwdth unts. Note that ths forms a mult-unt aucton wth the VCG mechansm [2]. The assgned bandwdth can be mplemented as an external source resdng n a remote MSC, whch adaptvely adjusts ts upln capacty. The followng equaton then holds, n whch W denotes the local bandwdth assgnment to channel K: W r = W K Auctons are conducted when bdders need a certan amount of bandwdth from sellers to fulfll content requests for W () locally unavalable contents or to allevate the pressure on heavly loaded channels. For the buyer server, the number of bandwdth unts requred for channel from server j s denoted as n. Buyer wll submt bds that nclude n and the correspondng valuaton nformaton to seller j. After collectng all bds from the cache servers n other WSPs, the wnner determnaton process at the seller sde can decde the bandwdth allocaton scheme. Consequently, buyers have to pay by vrtual payments that can later be used for bandwdth resource demands, whch serves as an ncentve for contrbutons. The truthfulness wll be guaranteed by the payment method wth any bds devatng from the true valuaton not beneftng n the VCG aucton. Besdes the bandwdth allocaton scheme derved from the above auctons, storage resources on cache servers also requre perodc updates to eep valuable content that beneft ther own payoffs and socal welfare. We show that storage updates n collaboratve cachng help acheve a desrable level of performance wth a comparatvely lower overhead, gven dfferent settngs of cachng mechansms. IV. VCG-BASED BANDWIDTH TRADING In ths secton, we descrbe the VCG-based bandwdth tradng mechansm employed by cache servers. We ntroduce the concept of vrtual bdder b whch submts bds to request n bandwdth unts on behalf of users on channel n doman. b has a prvately nown valuaton functon: v : {,,..., n } R, I, K,nwhchv (n) denotes the valuaton of the beneft of recevng n bandwdth unts. b then submts bds b = {(, ), (,v ()),..., (n,v (n ))} to seller j snce submttng one s true valuaton s the domnant strategy n VCG auctons. We now defne the valuaton functon n VCG auctons. When we consder a vdeo streamng system, the user streamng qualty can be used to reveal the actual beneft of recevng a certan amount of bandwdth unts. We also consder mportant observatons nsde real-world peer-asssted streamng systems through extensve measurements [3] [4]. Accordng to [3], there exsts a postve correlaton between the per-user server bandwdth provsoned and the proporton of users that experence a smooth playbac. In the context of ndependent cachng, the streamng qualty can therefore be defned as: q = γ( W r x ) α (2) In ths defnton, γ s an adjustable scalng parameter. The channel wth streamng rate r has x concurrent users. The value of α satsfes the constrant <α< [4], whch ndcates that the streamng qualty could be mproved when the provsoned bandwdth ncreases, but wth a decreasng margnal gan. Ths formulaton s also consstent wth observatons n [3], n whch the streamng qualty negatvely correlates to channel populatons. Defnton : Valuaton Functon. Based on the formulaton of the streamng qualty under ndependent cachng, we then propose the valuaton functon n collaboratve cachng as: v (n ) =γ(x ) α (( W + e w n r ) α ( W r ) α ) (3)

4 DAI et al.: COLLABORATIVE CACHING IN WIRELESS VIDEO STREAMING THROUGH RESOURCE AUCTIONS 46 The valuaton functon reflects the streamng qualty mprovement over channel by recevng n unts of bandwdth from server j. The notaton e ( <e ) represents the degradaton of benefts due to the dfferentaton of coverage areas among dfferent domans of server and j. Defnton 2: Wnner Determnaton. Each cache server j needs to determne the allocaton of Wj r after recevng bds from other servers. The wnner determnaton s consdered to be an effcent allocaton f t maxmzes the socal welfare as the followng: Maxmze Subject to: v ( n ) K I n W j r w K I n n I, K n s the optmzaton varable n the formulaton. Snce the valuaton functon s prvately nown, the recevng bds only nclude dscrete values for each possble number of acnowledged unts. Therefore, ths problem becomes an nteger programmng problem, whch s n general NP-hard. It has been further shown that truthfulness s no longer guaranteed n a VCG aucton f approxmaton algorthms are appled [5]. Fortunately, we show n Sec. IV-A that an optmal soluton can be acheved here by explorng the unque structure of the valuaton functon. Defnton 3: VCG Payments. The VCG-based aucton mechansm results n a payment for bdder b as: p = v ( n )+ max v b ( n ) v ( n ) = {(,)} K I K I (5) The notaton n denotes the optmal soluton to the wnner determnaton problem (4). The mplcatons of vrtual payment are two-fold. Frst, t acts as an ncentve to resource contrbutors whch ncur costs n delverng content. Second, the truthfulness of collaboratve cachng s also guaranteed whch s proven n Sec. IV-A. Defnton 4: The utlty ganed by vrtual bdders n a VCG aucton s u = v ( n ) p. A. Optmal Bandwdth Allocaton Strateges The conventonal wnner determnaton n the form of nteger programmng results n a hgh computatonal complexty [5]. However, the valuaton functon n our proposed mechansms satsfes a downward slopng property [2], whch leads to a polynomal-tme optmal allocaton. Theorem : The valuaton functon gven n (3) satsfes the downward slopng property. Proof: The frst order dervatve of (3) s gven as: v n (4) γ α e w(x = ) α r α (W + e w n > (6) ) α Ths ensures that the valuaton monotoncally ncreases wth a growng number of provsoned bandwdth unts. The second order dervatve of the valuaton functon s gven as: Algorthm A Collaboratve Cachng Framewor through VCG Resource Auctons : n r r j W j w 2: f n r j K I n then 3: n n, I, K 4: else 5: n, I, K 6: for n r j > do 7: {, } =argmaxv ( n I, K 8: n n + 9: n r j nr j : end for : end f +) v ( n ) 2 v 2 = γ(α2 α)(e w) 2 (x ) α r α (W + e w n < (7) )2 α n The second dervatve s constantly smaller than, whch ndcates that the valuaton functon gven n (3) satsfes the downward slopng property. Thus, the seller can then acheve the optmal soluton to problem (4) n polynomal tme wthout obtanng any detals of the valuaton functon. The procedure to determne the wnner s gven n Algorthm wth a computatonal complexty of O( I K n r j ) for each aucton. Theorem 2: Utlty u ganed by vrtual bdder b s a nonnegatve value. Proof: Suppose n s determned n the wnner determnaton process. The resultng payment s p gven by Eq. (5). In the optmal soluton to the problem of max K I v ( n ) n case of b = {(, )}, n = n. Therefore, ths soluton also satsfes the constrants n Problem (4) f remanng bds are not changed. Therefore, ts socal welfare s smaller than or equal to I v ( ).Thenp v ( n ), u. K n Theorem 3: Bddng on one s true valuaton v ( n ) maxmzes the utlty obtaned by vrtual bdders. Proof: The utlty ganed from the true valuaton s u = K I v ( n ) maxb K I ={(,)} v ( n ). Consder that b submts a false valuaton such that v ( n ) v ( n ). In ths case, the utlty ganed from the false valuaton s u = max K I v ( n ) v ( n ) max b ={(,)} K I v ( n )+v ( n ),where n s the optmal allocaton to max K I v ( n ).The utlty dfferentaton between two scenaros s denoted as u u = ( K I v ( n ) v ( n )) (max K I v ( n ) v ( n )). Snce n s an effcent allocaton to problem (4) n the context of true valuatons, the authentcty of remanng bds ensures that u u. Therefore, we have u u.

5 462 IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, VOL. 3, NO. 2, FEBRUARY 22 All bds submtted by vrtual bdders are therefore truthful accordng to Theorem 2 and 3. Meanwhle, the demands wth hgher valuatons are satsfed wth hgher prortes, whch optmze the socal welfare. The bddng strategy s smplfed, whch benefts the practcal deployment of the mechansms. B. Implementaton Issues Frst, local resource allocaton. In each tradng round, a fracton of β ( β ) of the total bandwdth capacty s used for the local content dssemnaton such that K W = βw. Snce the status of local channels s always nown to the respectve WSP, the decson of W s also based on the valuaton functon (3) to optmze the performance. Second, the determnaton of n. The decson of n s determned by the remanng budget n auctons. The cache server formulates an optmzaton problem that mnmzes the number of requested unts K n, where the constrant s gven as the total valuaton of requested unts K v(n ) beng greater than the remanng budget. Demands wth hgher valuatons can be found through a local search n polynomal tme, and requests are therefore ssued. Thrd, server selecton for ssung requests. Ths decson s made based on the external bandwdth avalablty and the content avalablty, whch can be acqured through nformaton exchange. Those cache servers wth more avalable bandwdth Wj r are preferred to be selected. Fourth, the frequency of tradng. If the trade s performed n each tme slot, the resource allocaton s acheved wth fner granularty but excessve computatonal overhead. To mtgate such overhead, the tradng result can be potentally used for a longer perod of tme. The effect of changng the tradng frequency s evaluated n Sec. VI. V. COLLABORATIVE CACHING STORAGE One of the man benefts of applyng collaboratve cachng n wreless vdeo streamng s the ablty of requestng locally unavalable contents whch mproves the storage utlzaton through resource multplexng. Gven bandwdth resource auctons ntroduced n Sec. IV, the lmted cache storage should also be carefully utlzed. Wthout loss of generalty, cache servers are requred to eep full copes of vdeo channels. The storage decson of cache server can then be denoted as a = {, }, I, K. In ths secton, we analyze the effect of dfferent cache storage strateges under both ndependent and collaboratve cachng mechansms. A. Performance Metrc The bandwdth resource of cache servers cannot be assumed to be abundant. Even f a vdeo s cached, there s no guarantee that the currently avalable bandwdth s suffcent to satsfy all the content requests. Therefore, the vdeo request can be counted as a cache ht only when the vdeo s cached and served wth allocated bandwdth. Accordngly, we defne the performance metrc of acheved cache ht rato as: K R = φ x I (8) K x n whch φ represents a bnary value such that: φ = {, If ( W w + j I n ) >,, Otherwse. The acheved cache ht rato defned n Eq. (8) s the proporton of onlne users whose content requests are served ether by a local WSP through local resource allocaton, or external WSPs through resource auctons. B. Storage Update Strateges Under both ndependent and collaboratve cachng mechansms, the perodcal storage update ams to mprove the overall streamng qualty n the system. It can be observed from Eq. (3) that t s more benefcal to assgn bandwdth resource to popular channels. Therefore, conventonal ndependent cachng prefers to cache hghly popular vdeos as: Maxmze Subject to: a x K a f S K a = {, } N (9) () Here f denotes the fle sze of vdeo and S represents the storage capacty of the cache server. In practce, the statstcal result of vdeo popularty over the last storage update nterval s used for the update decson. The storage update decson needs to further coordnate wth resource auctons wth the proposed collaboratve cachng mechansm. Cache servers update ther contents based on the local requests and the demands n the current maret, as both the locally and socally valuable content should be stored. The local demands are reflected by the local resource and the content requests n. The predcton of maret demands s to tae hstorcal tradng nformaton nto consderaton. Thus, the socally valuable content can be represented by the number of bandwdth unts satsfed to other allocaton W WSPs n j. The basc utlty maxmzaton formulaton s then gven as: Maxmze Subject to: (λa (W w + K j I a f S K a = {, } N n )+( λ)a n j ) j I () Here λ ( λ ) specfes the desgn choce for the storage update strategy n the collaboratve cachng mechansm. If we set λ =, the storage decson s made solely based on the local demands. If we set λ =, the storage update strategy s fully domnated by demands n the current maret. When we set λ as a certan value between and, trepresents a combned strategy that benefts both demands of the local WSP and content requests from other WSPs.

6 DAI et al.: COLLABORATIVE CACHING IN WIRELESS VIDEO STREAMING THROUGH RESOURCE AUCTIONS 463 Fg. 2. cachng C Cachng I Cachng Tme Slots The streamng qualty of collaboratve cachng versus ndependent Acheved Cache Ht Rato.9.7. C Cachng I Cachng Tme Slots Fg. 3. The acheved cache ht rato of collaboratve cachng versus ndependent cachng. C. Impact of Storage Update Interval The vdeo popularty throughout the entre system dynamcally changes over tme. In ths case, the cache server wth an ndependent cachng mechansm needs to frequently update ts storage to cache the most popular content, n order to acheve an effcent use of bandwdth resources. However, frequent updates would cause consderable cost n dstrbutng content from the centralzed content server to cache servers. Such costs can be sgnfcantly reduced n our proposed collaboratve cachng by enablng content delvery among dfferent WSPs. The dversfed contents among multple WSPs can help satsfy requests for vdeos that are not locally avalable even wth a long nterval of storage update. Ths cost reducton s llustrated n our performance evaluaton secton. VI. PERFORMANCE EVALUATION In ths secton, we examne the performance of the proposed collaboratve cachng scheme, n comparson wth the conventonal ndependent cachng mechansm. Our evaluaton s based on a tme-slotted smulator mplemented usng Python. The performance of dfferent cachng mechansms s analyzed n terms of the overall streamng qualty,.e., the percentage of onlne users that experence a smooth playbac, and the acheved cache ht rato, defned n Sec. V. We also tae a close loo at the potental cost and overhead of collaboratve cachng, whch s crtcal n practcal system desgn. We smulate four WSPs that deploy cache servers to assst wreless vdeo streamng. There are 5 vdeo channels and the pea popularty s dstrbuted over to 5. The bandwdth resource auctons are carred out perodcally accordng to a predefned allocaton update nterval. The storage space of cache servers s randomly flled at the ntal phase, and then perodcally updated wth dfferent strateges n Sec. V. A. Performance Improvement through Collaboratve Cachng We frst plot the overall streamng qualty mprovement brought forth by the proposed collaboratve cachng (denoted as C-Cachng) n Fg. 2. The allocaton update nterval s one tme slot, and the storage update nterval s 5 tme slots. The sze of the bandwdth unt n the aucton s the same as the vdeo streamng rate of 5 bps. Each cache server can store 25% of all the exstng vdeo channels, and the bandwdth capacty s set to 6, bandwdth unts whch account for 2% of the total bandwdth requrement n the system. The value of parameter β and λ s set to and, respectvely. We frst observe that the overall streamng qualty of C- Cachng consstently outperforms that of the ndependent cachng (denoted as I-Cachng). Due to the random ntalzaton of each cache storage, the streamng qualty of C- Cachng and I-Cachng stay stable at around 5% and 3%, respectvely. After the frst storage update, the streamng qualty of I-Cachng ncreases to 4%, as more bandwdth unts are utlzed for supportng channels wth hgh valuaton. However, the streamng qualty of C-Cachng does not explctly ncrease snce the resource auctons already fulfll the demands for those unavalable popular vdeos before the storage update. In subsequent tme slots, the streamng qualty of I-Cachng fluctuates between 2% and 4%, whle C- Cachng always mantans a satsfactory level of performance. Ths demonstrates the stablty of the proposed collaboratve cachng mechansm. Smlar performance patterns can be observed n the acheved cache ht rato as plotted by Fg. 3. On average, the acheved cache ht rato of C-Cachng s almost 3 tmes hgher than that of I-Cachng. The ratonale s that I-Cachng prefers to assgn bandwdth resources to the most popular vdeos to acheve a better overall performance. As such, only a smaller porton of vdeos can be satsfed by deployed cache servers. In comparson, the dversfed content exchanges under C- Cachng and the dedcated bandwdth resource n resource auctons could help mantan a hgher cache ht rato and relatvely stable streamng qualty. Fg. 4 compares dfferent scenaros of the bandwdth and storage capacty settngs. Along wth an ncrease n storage capacty, the streamng qualty under I-Cachng ncreases almost lnearly snce the negatve mpact of the content unavalablty brought forth by system dynamcs can be allevated. In comparson, the streamng qualty mprovement under C-Cachng s ntally sgnfcant and becomes less sgnfcant f the storage capacty exceeds 4%. Ths clearly demonstrates that collaboratve cachng through multplexng can substantally reduce the overall storage requrement to acheve a desrable streamng qualty. Fg. 5 depcts the mpact of the porton of bandwdth resources that are allowed to be utlzed by other WSPs.

7 464 IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, VOL. 3, NO. 2, FEBRUARY I Cachng, BW= C Cachng, BW= I Cachng, BW=2 C Cachng, BW=2 Storage Capacty Fg. 4. The streamng qualty of collaboratve cachng versus ndependent cachng, under dfferent settngs of bandwdth and storage capacty..9.7 C Cachng, ST=2% C Cachng, ST=4% C Cachng, ST=6%. I Cachng, ST=6%..7.9 Parameter λ n Storage Updates Fg. 6. The streamng qualty versus the parameter λ n the cache storage update, under dfferent settngs of storage capactes and cachng mechansms..9 C Cachng, BW= C Cachng, BW=6 C Cachng, BW=2.7 I Cachng, BW= β Fg. 5. The streamng qualty versus the percentage of bandwdth provsoned to other WSPs, under dfferent settngs of bandwdth capactes and cachng mechansms..7 C Cachng I Cachng Storage Update Intervals Fg. 7. The streamng qualty versus the cache storage update nterval, under dfferent settngs of cachng mechansms. It can be observed that the streamng qualty s maxmzed when an adequate porton of bandwdth s allocated for other WSPs (.e., for auctons), and both excessve local and external bandwdth provson n C-Cachng cannot yeld the maxmal streamng qualty. The reason s that when β s small, most of content requests from other WSPs can be satsfed even wth a low valuaton of the bandwdth unts. On the contrary, f the value of β s as large as, cache servers can no longer beneft from resource multplexng across dfferent WSPs, n cases of popularty varaton or content unavalablty. Fg. 5 suggests that we can properly desgn the β to acheve better streamng qualty n practcal engneerng. Fg. 6 denotes the effect of parameter λ n the storage update under dfferent storage capacty settngs. We observe that a local demands frst strategy s preferred when the storage capacty s lmted (ST= 2%) n C-Cachng. However, the overall content dversty s satsfactory by cachng respectve popular contents, whch can be concluded from Fg. 3 that exhbts a satsfactory level of the acheved cache ht rato. The senstvty of λ decreases when the storage capacty s relatvely abundant, snce both locally preferred and socally preferred contents could be ept wth a larger storage space, regardless of specfc settngs of λ. B. Overhead wth Collaboratve Cachng We next examne the overhead ncurred by the proposed C-Cachng by analyzng the mpact of the update nterval of cache storage and that of bandwdth allocaton. Fg. 7 llustrates the tradeoff between the streamng qualty and the frequency of cache storage updates. It can be seen that the performance gap between the two mechansms becomes more sgnfcant when the update nterval gradually gets larger. The ncreasng gap s caused by fluctuatons n channel popularty, whch can easly be resolved wth nter-wsp resource auctons. Ths shows that I-Cachng has to frequently update ts storage whch brngs a hgher cost assocated wth the cache replacement. In sharp contrast, the performance of C-Cachng remans stable despte specfc cache update strateges. We then analyze the tradeoff between the streamng qualty and the update nterval of bandwdth resource allocaton n Fg. 8. We evaluate several scenaros n whch decsons of bandwdth allocaton from resource auctons are utlzed over multple rounds. Fg. 8 clearly shows that the overall streamng qualty only slghtly decreases along wth an ncrease of the allocaton update nterval. Ths mples that n practce, t s feasble to reuse exstng bandwdth allocaton results, whch further reduces the overhead of aucton mechansms. C. Farness among WSPs When the aucton s conducted n practce among multple WSPs, t s ndspensable to further evaluate the farness of the system. Fg. 9 plots the respectve streamng qualty of dfferent WSPs n our smulaton. It shows that the varaton of streamng qualty at WSPs exhbts a smlar pattern over tme. Ths demonstrates that the proposed resource aucton can mprove the socal welfare by mutual assstance. Fg.

8 DAI et al.: COLLABORATIVE CACHING IN WIRELESS VIDEO STREAMING THROUGH RESOURCE AUCTIONS 465 C Cachng I Cachng Allocaton Update Intervals Fg. 8. The streamng qualty versus the update nterval of bandwdth allocaton, under dfferent settngs of cachng mechansms. Receved Bandwdth Unts 2 x 6.8 WSP WSP2.6 WSP3.4 WSP Tme Slots Fg.. The number of receved pacets from other WSPs n collaboratve cachng over tme. 5 5 WSP WSP2 WSP3 WSP Tme Slots Fg. 9. The streamng qualty of dfferent WSPs n collaboratve cachng over tme. further plots the cumulatve amount of receved pacets from other WSPs. It can be observed that the trend of all curves remans smlar overall, whch also reflects the farness of the collaboratve cachng mechansm. [5] N. Nsan and A. Ronen, Computatonally Feasble VCG Mechansms, Journal of Artfcal Intellgence Research, vol. 29, no., 27. [6] Q. Zhang, F. Yang, and W. Zhu, Cross-layer QoS Support for Multmeda Delvery over Wreless Internet, EURASIP J. Appl. Sgnal Process., vol. 25, no. 2, 25. [7] E. Tan, L. Guo, S. Chen, and X. Zhang, SCAP: Smart Cachng n Wreless Access Ponts to Improve P2P Streamng, n Proc. IEEE ICDCS, Jun. 27. [8] J. Da, B. L, F. Lu, B. L, and H. Jn, On the Effcency of Collaboratve Cachng n ISP-aware P2P Networs, n Proc. IEEE INFOCOM, Apr. 2. [9] L. Chen, M. Meo, and A. Sccchtano, Cachng Vdeo Contents n IPTV Systems wth Herarchcal Archtecture, n Proc. IEEE ICC, Jun. 29. [] S. Borst, V. Gupta, and A. Wald, Dstrbuted Cachng Algorthms for Content Dstrbuton Networs, n Proc. IEEE INFOCOM, Mar. 2. [] A. T. S. Ip, J. C. S. Lu, and J. Lu, A Revenue-rewardng Scheme of Provdng Incentve for Cooperatve Proxy Cachng for Meda Streamng Systems, ACM Trans. Multmeda Comput. Commun. Appl., vol. 4, no., 28. [2] S. Dobzns and N. Nsan, Mechansms for Mult-Unt Auctons, Journal of Artfcal Intellgence Research, vol. 37, no., 2. [3] C. Wu, B. L, and S. Zhao, Dagnosng Networ-wde P2P Lve Streamng Ineffcences, n Proc. IEEE INFOCOM, Apr. 29. [4] C. Wu and B. L, Mult-channel Lve P2P Streamng: Refocusng on Servers, n Proc. IEEE INFOCOM, Apr. 28. VII. CONCLUSION In ths paper, we have proposed a collaboratve cachng mechansm for wreless vdeo streamng that coordnates cache resource provsonng among selfsh WSPs. In partcular, we focus on engneerng the ncentves to promote and encourage cache servers from dfferent WSPs to truthfully cooperate wth one another n the context of VCG auctons. We have derved solutons to allocate server bandwdth and analyzed the utlzaton of server storage space. Smulaton results demonstrate that the performance of streamng systems can be sgnfcantly mproved, wth maxmzed socal welfare n auctons conducted by the collaboratve cachng mechansm. Je Da s currently a Ph.D. student n the Department of Computer Scence and Engneerng, Hong Kong Unversty of Scence and Technology. Pror to that, he receved hs B.Eng. degree n computer scence from Huazhong Unversty of Scence and Technology, Wuhan, Chna, n 27. From Aprl to August 2, he was a vstng student at the School of Computng Scence, Smon Fraser Unversty, Canada. Hs research nterests nclude collaboratve cachng systems, ISP networs, cloud computng, peer-to-peer networs, and multmeda systems. He s a student member of IEEE. REFERENCES [] F. Khan, LTE for 4G Moble Broadband: Ar Interface Technologes and Performance. Cambrdge Unversty Press, 29. [2] Q. Zhang, Z. Xang, W. Zhu, and L. Gao, Cost-Based Cache Replacement and Server Selecton for Multmeda Proxy across Wreless Internet, IEEE Transactons on Multmeda, vol. 6, no. 4, 24. [3] H. Chen and Y. Xao, Cache Access and Replacement for Future Wreless Internet, IEEE Communcatons Magazne, vol. 44, no. 5, 26. [4] Explanaton of Optmzaton Deployment, com/vzwoptmzaton/.

9 466 IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, VOL. 3, NO. 2, FEBRUARY 22 Fangmng Lu receved hs B.Engr. degree n 25 from Department of Computer Scence and Technology, Tsnghua Unversty, Beng, Chna; and hs Ph.D. degree n Computer Scence and Engneerng from the Hong Kong Unversty of Scence and Technology n 2. He s currently an Assocate Professor n the Servces Computng Technology and System Lab, Cluster and Grd Computng Lab, School of Computer Scence and Technology, Huazhong Unversty of Scence and Technology, Wuhan, Chna. From Aug 29 to Feb 2, he was a vstng student at the Computer Engneerng Group, Department of Electrcal and Computer Engneerng, Unversty of Toronto, Canada. Hs research nterests are n the area of peer-to-peer networs, rch-meda dstrbuton, cloud computng and large-scale datacenter networng. He s a member of IEEE and IEEE Communcatons Socety. Bo L (F ) s a professor n the Department of Computer Scence and Engneerng, Hong Kong Unversty of Scence and Technology. He s a Cheung Kong Char Professor n Shangha Jao Tong Unversty, Chna, and the Chef Techncal Advsor for Chna Cache Corp. (NASDAQ:CCIH). He was prevously wth IBM Networng System Dvson, Research Trangle Par, and an adjunct researcher at Mcrosoft Research Asa. Hs recent nterests nclude: large-scale content dstrbuton n the Internet, Peer-to-Peer meda streamng, the Internet topology, cloud computng, green computng and communcatons. He receved hs B. Eng. Degree n the Computer Scence from Tsnghua Unversty, Beng, and hs Ph.D. degree n the Electrcal and Computer Engneerng from Unversty of Massachusetts at Amherst. Baochun L receved the B.Engr. degree from the Department of Computer Scence and Technology, Tsnghua Unversty, Chna, n 995 and the M.S. and Ph.D. degrees from the Department of Computer Scence, Unversty of Illnos at UrbanaChampagn, Urbana, n 997 and 2. Snce 2, he has been wth the Department of Electrcal and Computer Engneerng at the Unversty of Toronto, where he s currently a Professor and the Bell Canada Endowed Char n Computer Engneerng. Hs research nterests nclude cloud computng, peer-to-peer networs, applcatons of networ codng, and wreless networs. He was the recpent of the IEEE Communcatons Socety Leonard G. Abraham Award n the Feld of Communcatons Systems n 2. In 29, he was a recpent of the Multmeda Communcatons Best Paper Award from the IEEE Communcatons Socety, and a recpent of the Unversty of Toronto McLean Award. He s a member of ACM and a senor member of IEEE. Jangchuan Lu (S -M 3-SM 8) receved the BEng degree (cum laude) from Tsnghua Unversty, Chna, n 999, and the PhD degree from The Hong Kong Unversty of Scence and Technology n 23. He s a co-recpent of the Best Student Paper Award of IWQoS 28 and the Multmeda Communcatons Best Paper Award from the IEEE Communcatons Socety. He s currently an Assocate Professor at Smon Fraser Unversty, Brtsh Columba, Canada, and was an Assstant Professor at The Chnese Unversty of Hong Kong from 23 to 24. Hs research nterests nclude cloud computng, peer-to-peer systems, multmeda communcatons, and wreless networng. He s an Assocate Edtor of IEEE Transactons on Multmeda, and an edtor of IEEE Communcatons Surveys and Tutorals.

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

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

A Proactive Non-Cooperative Game-theoretic Framework for Data Replication in Data Grids

A Proactive Non-Cooperative Game-theoretic Framework for Data Replication in Data Grids Eghth IEEE Internatonal Symposum on Cluster Computng and the Grd A Proactve Non-Cooperatve Game-theoretc Framewor for Data Replcaton n Data Grds Al H. Elghran, Student Member, IEEE, Ry Subrata, Member,

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

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

Regional Load Balancing Circuitous Bandwidth Allocation Method Based on Dynamic Auction Mechanism

Regional Load Balancing Circuitous Bandwidth Allocation Method Based on Dynamic Auction Mechanism ATEC Web of Conferences 76, (8) IFID 8 https://do.org/./matecconf/876 Regonal Load Balancng Crcutous Bandwdth Allocaton ethod Based on Dynamc Aucton echansm Wang Chao, Zhang Dalong, Ran Xaomn atonal Dgtal

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

Advanced radio access solutions for the new 5G requirements

Advanced radio access solutions for the new 5G requirements Advanced rado access solutons for the new 5G requrements Soumaya Hamouda Assocate Professor, Unversty of Carthage Tuns, Tunsa Soumaya.hamouda@supcom.tn IEEE Summt 5G n Future Afrca. May 3 th, 2017 Pretora,

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

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

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

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

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

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

Network Coding as a Dynamical System

Network Coding as a Dynamical System Network Codng as a Dynamcal System Narayan B. Mandayam IEEE Dstngushed Lecture (jont work wth Dan Zhang and a Su) Department of Electrcal and Computer Engneerng Rutgers Unversty Outlne. Introducton 2.

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

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

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

Private Information Retrieval (PIR)

Private Information Retrieval (PIR) 2 Levente Buttyán Problem formulaton Alce wants to obtan nformaton from a database, but she does not want the database to learn whch nformaton she wanted e.g., Alce s an nvestor queryng a stock-market

More information

HIERARCHICAL SCHEDULING WITH ADAPTIVE WEIGHTS FOR W-ATM *

HIERARCHICAL SCHEDULING WITH ADAPTIVE WEIGHTS FOR W-ATM * Copyrght Notce c 1999 IEEE. Personal use of ths materal s permtted. However, permsson to reprnt/republsh ths materal for advertsng or promotonal purposes or for creatng new collectve wors for resale or

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

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

Optimized caching in systems with heterogeneous client populations

Optimized caching in systems with heterogeneous client populations Performance Evaluaton 42 (2000) 163 185 Optmzed cachng n systems wth heterogeneous clent populatons Derek L. Eager a,, Mchael C. Ferrs b, Mary K. Vernon b a Department of Computer Scence, Unversty of Saskatchewan,

More information

Game Based Virtual Bandwidth Allocation for Virtual Networks in Data Centers

Game Based Virtual Bandwidth Allocation for Virtual Networks in Data Centers Avaable onlne at www.scencedrect.com Proceda Engneerng 23 (20) 780 785 Power Electroncs and Engneerng Applcaton, 20 Game Based Vrtual Bandwdth Allocaton for Vrtual Networks n Data Centers Cu-rong Wang,

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

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

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

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

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

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

ARTICLE IN PRESS. Signal Processing: Image Communication

ARTICLE IN PRESS. Signal Processing: Image Communication Sgnal Processng: Image Communcaton 23 (2008) 754 768 Contents lsts avalable at ScenceDrect Sgnal Processng: Image Communcaton journal homepage: www.elsever.com/locate/mage Dstrbuted meda rate allocaton

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

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

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

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

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

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

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

Pricing Network Resources for Adaptive Applications in a Differentiated Services Network

Pricing Network Resources for Adaptive Applications in a Differentiated Services Network IEEE INFOCOM Prcng Network Resources for Adaptve Applcatons n a Dfferentated Servces Network Xn Wang and Hennng Schulzrnne Columba Unversty Emal: {xnwang, schulzrnne}@cs.columba.edu Abstract The Dfferentated

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

Cognitive Radio Resource Management Using Multi-Agent Systems

Cognitive Radio Resource Management Using Multi-Agent Systems Cogntve Rado Resource Management Usng Mult- Systems Jang Xe, Ivan Howtt, and Anta Raja Department of Electrcal and Computer Engneerng Department of Software and Informaton Systems The Unversty of North

More information

Internet Traffic Managers

Internet Traffic Managers Internet Traffc Managers Ibrahm Matta matta@cs.bu.edu www.cs.bu.edu/faculty/matta Computer Scence Department Boston Unversty Boston, MA 225 Jont work wth members of the WING group: Azer Bestavros, John

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

An Optimal Bandwidth Allocation and Data Droppage Scheme for Differentiated Services in a Wireless Network

An Optimal Bandwidth Allocation and Data Droppage Scheme for Differentiated Services in a Wireless Network Purdue Unversty Purdue e-pubs ECE Techncal Reports Electrcal and Computer Engneerng 3--7 An Optmal Bandwdth Allocaton and Data Droppage Scheme for Dfferentated Servces n a Wreless Network Waseem Shekh

More information

2892 IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, VOL. 33, NO. 12, DECEMBER 2015

2892 IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, VOL. 33, NO. 12, DECEMBER 2015 2892 IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, VOL. 33, NO. 12, DECEMBER 2015 Incentve Mechansms for Economc and Emergency Demand Responses of Colocaton Datacenters Nguyen H. Tran, Member, IEEE,

More information

DUE to the recent popularization of hand-held mobile

DUE to the recent popularization of hand-held mobile IEEE TRANSACTIONS ON MOBILE COMPUTING, VOL. 15, NO. 4, APRIL 2016 803 Contact Duraton Aware Data Replcaton n DTNs wth Lcensed and Unlcensed Spectrum Jng Zhao, Student Member, IEEE, Xuejun Zhuo, Qnghua

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

Distributed Middlebox Placement Based on Potential Game

Distributed Middlebox Placement Based on Potential Game Int. J. Communcatons, Network and System Scences, 2017, 10, 264-273 http://www.scrp.org/ournal/cns ISSN Onlne: 1913-3723 ISSN Prnt: 1913-3715 Dstrbuted Mddlebox Placement Based on Potental Game Yongwen

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

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

Delay Variation Optimized Traffic Allocation Based on Network Calculus for Multi-path Routing in Wireless Mesh Networks

Delay Variation Optimized Traffic Allocation Based on Network Calculus for Multi-path Routing in Wireless Mesh Networks Appl. Math. Inf. Sc. 7, No. 2L, 467-474 2013) 467 Appled Mathematcs & Informaton Scences An Internatonal Journal http://dx.do.org/10.12785/ams/072l13 Delay Varaton Optmzed Traffc Allocaton Based on Network

More information

IP mobility support is becoming very important as the

IP mobility support is becoming very important as the 706 IEEE TRANSACTIONS ON COMPUTERS, VOL. 52, NO. 6, JUNE 2003 A New Scheme for Reducng Lnk and Sgnalng Costs n Moble IP Young J. Lee and Ian F. Akyldz, Fellow, IEEE Abstract IP moblty support s provded

More information

Fast Retransmission of Real-Time Traffic in HIPERLAN/2 Systems

Fast Retransmission of Real-Time Traffic in HIPERLAN/2 Systems Fast Retransmsson of Real-Tme Traffc n HIPERLAN/ Systems José A Afonso and Joaqum E Neves Department of Industral Electroncs Unversty of Mnho, Campus de Azurém 4800-058 Gumarães, Portugal {joseafonso,

More information

Research of Dynamic Access to Cloud Database Based on Improved Pheromone Algorithm

Research of Dynamic Access to Cloud Database Based on Improved Pheromone Algorithm , pp.197-202 http://dx.do.org/10.14257/dta.2016.9.5.20 Research of Dynamc Access to Cloud Database Based on Improved Pheromone Algorthm Yongqang L 1 and Jn Pan 2 1 (Software Technology Vocatonal College,

More information

Quality Improvement Algorithm for Tetrahedral Mesh Based on Optimal Delaunay Triangulation

Quality Improvement Algorithm for Tetrahedral Mesh Based on Optimal Delaunay Triangulation Intellgent Informaton Management, 013, 5, 191-195 Publshed Onlne November 013 (http://www.scrp.org/journal/m) http://dx.do.org/10.36/m.013.5601 Qualty Improvement Algorthm for Tetrahedral Mesh Based on

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

Hermite Splines in Lie Groups as Products of Geodesics

Hermite Splines in Lie Groups as Products of Geodesics Hermte Splnes n Le Groups as Products of Geodescs Ethan Eade Updated May 28, 2017 1 Introducton 1.1 Goal Ths document defnes a curve n the Le group G parametrzed by tme and by structural parameters n the

More information

THere are increasing interests and use of mobile ad hoc

THere are increasing interests and use of mobile ad hoc 1 Adaptve Schedulng n MIMO-based Heterogeneous Ad hoc Networks Shan Chu, Xn Wang Member, IEEE, and Yuanyuan Yang Fellow, IEEE. Abstract The demands for data rate and transmsson relablty constantly ncrease

More information

MOBILE Cloud Computing (MCC) extends the capabilities

MOBILE Cloud Computing (MCC) extends the capabilities 1 Resource Sharng of a Computng Access Pont for Mult-user Moble Cloud Offloadng wth Delay Constrants Meng-Hs Chen, Student Member, IEEE, Mn Dong, Senor Member, IEEE, Ben Lang, Fellow, IEEE arxv:1712.00030v2

More information

an assocated logc allows the proof of safety and lveness propertes. The Unty model nvolves on the one hand a programmng language and, on the other han

an assocated logc allows the proof of safety and lveness propertes. The Unty model nvolves on the one hand a programmng language and, on the other han UNITY as a Tool for Desgn and Valdaton of a Data Replcaton System Phlppe Quennec Gerard Padou CENA IRIT-ENSEEIHT y Nnth Internatonal Conference on Systems Engneerng Unversty of Nevada, Las Vegas { 14-16

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

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

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

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

Hybrid-clustering game Algorithm for Resource Allocation in Macro-Femto HetNet

Hybrid-clustering game Algorithm for Resource Allocation in Macro-Femto HetNet KSII TRANSACTIONS ON INTERNET AND INFORMATION SYSTEMS VOL. 12, NO. 4, Apr. 2018 1638 Copyrght c 2018 KSII Hybrd-clusterng game Algorthm for Resource Allocaton n Macro-Femto HetNet Fang Ye 1, Jng Da 1 and

More information

Avoiding congestion through dynamic load control

Avoiding congestion through dynamic load control Avodng congeston through dynamc load control Vasl Hnatyshn, Adarshpal S. Seth Department of Computer and Informaton Scences, Unversty of Delaware, Newark, DE 976 ABSTRACT The current best effort approach

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

Control strategies for network efficiency and resilience with route choice

Control strategies for network efficiency and resilience with route choice Control strateges for networ effcency and reslence wth route choce Andy Chow Ru Sha Centre for Transport Studes Unversty College London, UK Centralsed strateges UK 1 Centralsed strateges Some effectve

More information

Post-Streaming Rate Analysis - A New Approach to Mobile Video Streaming with Predictable Performance

Post-Streaming Rate Analysis - A New Approach to Mobile Video Streaming with Predictable Performance The fnal verson of record s avalable at http://dx.do.org/10.1109/tmc.2017.2694418 IEEE TRANSACTIONS ON MOBILE COMPUTING, MANUSCRIPT ID TMC-2016-05-0324 1 Post-Streamng Rate Analyss - A New Approach to

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

Proper Choice of Data Used for the Estimation of Datum Transformation Parameters

Proper Choice of Data Used for the Estimation of Datum Transformation Parameters Proper Choce of Data Used for the Estmaton of Datum Transformaton Parameters Hakan S. KUTOGLU, Turkey Key words: Coordnate systems; transformaton; estmaton, relablty. SUMMARY Advances n technologes and

More information

Optimal Caching Placement for D2D Assisted Wireless Caching Networks

Optimal Caching Placement for D2D Assisted Wireless Caching Networks Optmal Cachng Placement for DD Asssted Wreless Cachng Networks Jun Rao, ao Feng, Chenchen Yang, Zhyong Chen, and Bn Xa Department of Electronc Engneerng, Shangha Jao Tong Unversty, Shangha, P. R. Chna

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

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

The Research of Support Vector Machine in Agricultural Data Classification

The Research of Support Vector Machine in Agricultural Data Classification The Research of Support Vector Machne n Agrcultural Data Classfcaton Le Sh, Qguo Duan, Xnmng Ma, Me Weng College of Informaton and Management Scence, HeNan Agrcultural Unversty, Zhengzhou 45000 Chna Zhengzhou

More information

QoS-aware composite scheduling using fuzzy proactive and reactive controllers

QoS-aware composite scheduling using fuzzy proactive and reactive controllers Khan et al. EURASIP Journal on Wreless Communcatons and Networkng 2014, 2014:138 http://jwcn.euraspjournals.com/content/2014/1/138 RESEARCH Open Access QoS-aware composte schedulng usng fuzzy proactve

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

Efficient QoS Provisioning at the MAC Layer in Heterogeneous Wireless Sensor Networks

Efficient QoS Provisioning at the MAC Layer in Heterogeneous Wireless Sensor Networks Effcent QoS Provsonng at the MAC Layer n Heterogeneous Wreless Sensor Networks M.Soul a,, A.Bouabdallah a, A.E.Kamal b a UMR CNRS 7253 HeuDaSyC, Unversté de Technologe de Compègne, Compègne Cedex F-625,

More information

Distortion-Memory Tradeoffs in Cache-Aided Wireless Video Delivery

Distortion-Memory Tradeoffs in Cache-Aided Wireless Video Delivery Dstorton-Memory Tradeoffs n Cache-Aded Wreless Vdeo Delvery P. Hassanzadeh, E. Erkp, J. Llorca, A. Tulno arxv:1511.03932v1 [cs.it] 12 Nov 2015 Abstract Moble network operators are consderng cachng as one

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

LECTURE NOTES Duality Theory, Sensitivity Analysis, and Parametric Programming

LECTURE NOTES Duality Theory, Sensitivity Analysis, and Parametric Programming CEE 60 Davd Rosenberg p. LECTURE NOTES Dualty Theory, Senstvty Analyss, and Parametrc Programmng Learnng Objectves. Revew the prmal LP model formulaton 2. Formulate the Dual Problem of an LP problem (TUES)

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

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

Resource Control for Loss-Sensitive Traffic in CDMA Networks

Resource Control for Loss-Sensitive Traffic in CDMA Networks Resource Control for Loss-Senstve Traffc n CDMA Networks Vaslos A. Srs Insttute of Computer Scence, FORTH and Department of Computer Scence, Unversty of Crete P. O. Box 1385, GR 711 1, Heraklon, Crete,

More information

Resource and Virtual Function Status Monitoring in Network Function Virtualization Environment

Resource and Virtual Function Status Monitoring in Network Function Virtualization Environment Journal of Physcs: Conference Seres PAPER OPEN ACCESS Resource and Vrtual Functon Status Montorng n Network Functon Vrtualzaton Envronment To cte ths artcle: MS Ha et al 2018 J. Phys.: Conf. Ser. 1087

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

MobileGrid: Capacity-aware Topology Control in Mobile Ad Hoc Networks

MobileGrid: Capacity-aware Topology Control in Mobile Ad Hoc Networks MobleGrd: Capacty-aware Topology Control n Moble Ad Hoc Networks Jle Lu, Baochun L Department of Electrcal and Computer Engneerng Unversty of Toronto {jenne,bl}@eecg.toronto.edu Abstract Snce wreless moble

More information

Channel-Quality Dependent Earliest Deadline Due Fair Scheduling Schemes for Wireless Multimedia Networks

Channel-Quality Dependent Earliest Deadline Due Fair Scheduling Schemes for Wireless Multimedia Networks Channel-Qualty Dependent Earlest Deadlne Due Far Schedulng Schemes for Wreless Multmeda Networks Ahmed K. F. Khattab Khaled M. F. Elsayed ahmedkhattab@eng.cu.edu.eg khaled@eee.org Department of Electroncs

More information

Intelligent Information Acquisition for Improved Clustering

Intelligent Information Acquisition for Improved Clustering Intellgent Informaton Acquston for Improved Clusterng Duy Vu Unversty of Texas at Austn duyvu@cs.utexas.edu Mkhal Blenko Mcrosoft Research mblenko@mcrosoft.com Prem Melvlle IBM T.J. Watson Research Center

More information

An Application of the Dulmage-Mendelsohn Decomposition to Sparse Null Space Bases of Full Row Rank Matrices

An Application of the Dulmage-Mendelsohn Decomposition to Sparse Null Space Bases of Full Row Rank Matrices Internatonal Mathematcal Forum, Vol 7, 2012, no 52, 2549-2554 An Applcaton of the Dulmage-Mendelsohn Decomposton to Sparse Null Space Bases of Full Row Rank Matrces Mostafa Khorramzadeh Department of Mathematcal

More information

End-to-end Distortion Estimation for RD-based Robust Delivery of Pre-compressed Video

End-to-end Distortion Estimation for RD-based Robust Delivery of Pre-compressed Video End-to-end Dstorton Estmaton for RD-based Robust Delvery of Pre-compressed Vdeo Ru Zhang, Shankar L. Regunathan and Kenneth Rose Department of Electrcal and Computer Engneerng Unversty of Calforna, Santa

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

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

WITH rapid improvements of wireless technologies,

WITH rapid improvements of wireless technologies, JOURNAL OF SYSTEMS ARCHITECTURE, SPECIAL ISSUE: HIGHLY-RELIABLE CPS, VOL. 00, NO. 0, MONTH 013 1 Adaptve GTS Allocaton n IEEE 80.15.4 for Real-Tme Wreless Sensor Networks Feng Xa, Ruonan Hao, Je L, Naxue

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

Session 5.3. Switching/Routing and Transmission planning

Session 5.3. Switching/Routing and Transmission planning ITU Regonal Semnar Belgrade Serba and Montenegro 20-24 24 June 2005 Sesson 5.3 Swtchng/Routng and Transmsson plannng volvng nfrastructures to NGN and related Plannng Strateges and Tools I.S. Sesson 5.3-1

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

Needed Information to do Allocation

Needed Information to do Allocation Complexty n the Database Allocaton Desgn Must tae relatonshp between fragments nto account Cost of ntegrty enforcements Constrants on response-tme, storage, and processng capablty Needed Informaton to

More information

TN348: Openlab Module - Colocalization

TN348: Openlab Module - Colocalization TN348: Openlab Module - Colocalzaton Topc The Colocalzaton module provdes the faclty to vsualze and quantfy colocalzaton between pars of mages. The Colocalzaton wndow contans a prevew of the two mages

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