vskyconf: Cloud-assisted Multi-party Mobile Video Conferencing

Size: px
Start display at page:

Download "vskyconf: Cloud-assisted Multi-party Mobile Video Conferencing"

Transcription

1 vskycof: Cloud-assisted Multi-party Mobile Video Coferecig ABSTRACT Yu Wu Dept. of Computer Sciece The Uiversity of Hog Kog Bo Li Dept. of Computer Sciece ad Egieerig, Hog Kog Uiversity of Sciece ad Techology As a importat applicatio i today s busy world, mobile video coferecig facilitates people s virtual face-to-face commuicatio with frieds, families ad colleagues, via their mobile devices o the move. However, how to provisio high-quality, multi-party video coferecig experieces over mobile devices is still a ope challege. The fudametal reaso behid is the lack of computatio ad commuicatio capacities o the mobile devices, to scale to large coferecig sessios. I this paper, we preset vskycof, a cloud-assisted mobile video coferecig system to fudametally improve the quality ad scale of multi-party mobile video coferecig. By ovelly employig a surrogate virtual machie i the cloud for each mobile user, we allow fully scalable commuicatio amog the coferece participats via their surrogates, rather tha directly. The surrogates exchage coferecig streams amog each other, trascode the streams to the most appropriate bit rates, ad buffer the streams for the most efficiet delivery to the mobile recipiets. A fully decetralized algorithm is desiged to decide the best paths of streams ad the most suitable surrogates for video trascodig alog the paths, such that the limited badwidth is fully utilized to deliver streams of the highest possible quality to the mobile recipiets. We also carefully tailor a bufferig mechaism o each surrogate to cooperate with efficiet stream distributio. We have implemeted vskycof based o Amazo EC2 ad verified the excellet performace of our desig, as compared to the widely adopted uicast solutios. Categories ad Subject Descriptors J. [Computer Applicatios]: Geeral Keywords Cloud computig, Video coferecig, Mobile computig Permissio to make digital or hard copies of all or part of this work for persoal or classroom use is grated without fee provided that copies are ot made or distributed for profit or commercial advatage ad that copies bear this otice ad the full citatio o the first page. Copyrights for compoets of this work owed by others tha ACM must be hoored. Abstractig with credit is permitted. To copy otherwise, or republish, to post o servers or to redistribute to lists, requires prior specific permissio ad/or a fee. Request permissios from Permissios@acm.org. MCC 13, August 12, 213, Hog Kog, Chia. Copyright 213 ACM /13/8...$15.. Chua Wu Dept. of Computer Sciece The Uiversity of Hog Kog cwu@cs.hku.hk Fracis C.M. Lau Dept. of Computer Sciece The Uiversity of Hog Kog fcmlau@cs.hku.hk 1. INTRODUCTION Video coferecig has bee widely deployed for virtual, faceto-face commuicatio amog separate parties, as a greeer solutio to replace may of the eergy-expesive coferece travels. Beefitig from the advaces i mobile ad wireless commuicatio techologies, a umber of mobile video coferecig applicatios [1] have emerged. May of them rely o expesive, dedicated architectures, e.g., multiple cotrol uits (MCU), to process sigalig messages, trascode igress sessio streams ad dissemiate multiple streams to each ed device. Alteratively, distributed peerto-peer (P2P) based mobile video coferecig systems have also sprug up, e.g., Skype mobile leverages itermediate super odes for sessio relays. I order to fid out how well the the existig mobile video coferecig systems support multi-party video coferecig over mobile devices, we have coducted a survey of the seve represetative applicatios, with results give i [1].We observe that applicatios with ifrastructure support ca support more cocurret users i each sessio (but still typically o more tha 4), at the cost of expesive up-frot ivestmet which may prohibit their wide adoptio by small or medium istitutios; P2P-based solutios are reluctat to allow group video calls for a fear of compromisig call qualities, (e.g., Skype oly supports two-way visual commuicatio o mobile phoes), ad most of them stick to a sigle flat streamig rate, or a limited umber of bit rates (e.g., 2).A fullyadaptive, multi-party mobile video coferecig solutio is still pedig to achieve. We summarize the key challeges as follows: (1) The workload o each ode i a video coferecig sessio, i terms of both processig ad trasmissio, scales quadratically to the umber of participats i the sessio, leadig to sigificat challege whe usig mobile devices for multi-party video coferecig. (2) Mobile users are equipped with differet devices ad dowlik speeds; a high-quality solutio should eable differetiated call qualities to differet users, istead of a homogeeous video broadcast quality eforced by the low-ed users, as i a traditioal solutio. I this paper, we preset vskycof, a cloud-assisted mobile video coferecig solutio to fudametally eable high-quality, multiparty video coferecig over heterogeeous mobile devices. The cloud computig paradigm offers ubiquitously accessible computig resources, with o-demad resource provisioig at the modest cost. The paradigm particularly compesates well for the iheret 33

2 resource deficiecies of mobile devices, ad catalyzes the udergoig evolutio i the burgeoig mobile computig idustry. I vskycof, we dyamically provisio a virtual machie i the cloud as the exclusive surrogate for a dialed-i mobile user. Each mobile device dissemiates/receives the video streams to/from its surrogate; the surrogates exchage coferecig streams amog each other, trascode the streams to the most appropriate bit rates, ad buffer the streams for the most efficiet delivery to the mobile recipiets. By leveragig the more powerful processig capabilities ad stable wired etwork badwidths, mobile users shift those otherwise o-device tasks to the cloud, yieldig superior power reductio ad quality ehacemet, as well as achievig fully scalable commuicatio amog the coferece participats. To realize such a solutio, we desig a fully decetralized, efficiet algorithm to decide the best delivery paths of streams amog the surrogates (possibly distributed i differet cloud data ceters), ad the most suitable surrogates for video trascodig alog the paths. We also carefully tailor a bufferig mechaism o each surrogate to cooperate with the efficiet stream distributio. Together they guaratee bouded, small ed-to-ed latecies ad smooth stream playback at the mobile devices at the highest possible qualities, i each video coferecig sessio. We have implemeted a prelimiary versio of vskycof based o Amazo EC2, ad coducted experimets i the real-world settigs. The results reveal the high scalability, full adaptability, ad excellet video coferecig qualities achieved by our desig, as compared to the widely adopted uicast solutios. The remaider of this paper is orgaized as follows. We itroduce related literature i Sec. 2, preset uique challeges ad the system architecture i Sec. 3, ufold desig details i Sec. 4, itroduce our prototype implemetatio ad evalutio i Sec. 5, ad fially coclude the paper i Sec RELATED WORK Despite extesive studies durig the past decades, video coferecig (VC) has recaptured people s iterest i this ew smartphoe era, with a series of works ad systems sprigig up recetly [9][2] [4][7], which ca be categorized ito Server-to-Cliet (S/C) based ad Peer-to-Peer (P2P) based solutios. Compared to the S/C-based structure, P2P has bee deemed as a more promisig ad scalable solutio. Both Poec et al. [9] ad Che et al. [2] formulate utility maximizatio problems ad eable multi-party VC by buildig multi-rate multicast trees. They focus more o the streamig rate allocatio over physical liks, but do ot ivestigate much the trascodig flexibilities. Liag et al. [7] leverage the upload capacities of helpers from differet swarms, i similar ways as adopted by Skype (ot Skype mobile). Though promisig, these solutios are difficult to achieve i practice amog mobile users, who are reluctat to cotribute resources to stragers due to costraied batteries ad expesive cellular data fees. The domiat P2P-based solutio i the real world is still pair-wise uicast (e.g., Frig, Tago, etc.[1]), where a user directly exchages streams with each of the other users i the same coferecig sessio. I this way, the limited uplik badwidths of mobile devices sigificatly restrict the sessio size. Cloud computig, as a agile solutio, compesates well for the deficiecies of mobile devices for media streamig i terms of both processig ad badwidth supports. Traditioal players i the VC market have recetly claimed their support for mobile users via their private clouds. But their solutios are mostly cetralized i a private cloud with abudat ifrastructure resources to support eterprise users. For example, Vidyo [1] provides its cloud-based solutio by provisioig virtual MCUs o top of their Iitiator First/Last-mile Streams Cotrol Sigals Video Streams Participator Participator Figure 1: The architecture of vskycof. Participator VidyoRouters, bearig similar flavors to their traditioal dedicated ifrastructures. I cotrast, our desig ovelly provisios a VM surrogate for each mobile user i a public cloud i a more affordable maer, caterig for the eeds of ordiary mobile users i their daily life. A recet work by Feg et al. [3] optimally leverages iter-dataceter etwork badwidths to maximize the overall throughput of all coferecig sessios, based o itra-sessio etwork codig. vskycof cosiders both dyamic sessio routig ad adaptive sessio trascodig, ad exploits surrogates i a cloud ifrastructure for scalable video coferecig amog mobile devices. Huag et al. [4] leverage clouds to ecode videos ito layered rates usig scalable video codig (SVC), to eable differetiated streamig qualities to users with differet available badwidths. We do ot cosider SVC ecodig i this work, sice the ecodig complexities ievitably icur itolerable delays for a time-sesitive applicatio like video coferecig. 3. ARCHITECTURE vskycof eables efficiet, peer-to-peer fashioed, multi-party mobile video coferecig via a IaaS cloud, with the architecture preseted i Fig. 1. We refer to a video coferece call amog multiple mobile users as a sessio. The user which starts the coferece call is the iitiator of the sessio. A surrogate, i.e., a virtual machie (VM) istace, is created i the IaaS cloud for each mobile user. The IaaS cloud cosists of disparate data ceters i differet geographic locatios, ad the surrogate for each mobile user is assiged i a data ceter proximate to the user. As a proxy for the mobile device, a mobile user s surrogate is resposible for the followig: (i) sessio maiteace, by exchagig cotrol messages with other surrogates i a timely ad efficiet maer; (ii) video stream dissemiatio ad trascodig, by receivig the video stream its correspodig mobile user produces, trascodig it ito appropriate format(s), distributig it to other users surrogates, ad the other way roud as well; (ii) efficiet video stream bufferig for its mobile user, for timely, smooth ad robust streamig to the correspodig mobile device. A mobile user just eeds to sed the stream it geerates ad receive streams other users produce to ad from its surrogate, ad is effectively freed from power-cosumig stream processig ad itra-sessio commuicatio. A gateway server i vskycof loosely keeps track of participatig users ad their surrogates, which ca be implemeted by a stadaloe server or VMs i the IaaS cloud. Fig. 2 depicts the key modules implemeted o a sigle surrogate, which ca be divided ito two parts: Cotrol Plae is the brai of the surrogate, resposible for cotrol sigalig betwee this surrogate ad eighborig surrogates. It measures the latecies ad badwidths o the coectios from/to eighborig surrogates, ad all the collected iformatio is stored i the peer table, which costructs a partial view of the video coferecig topology from this surrogate s poit of view. Utilizig the collected iformatio, the surrogate computes routig paths for 34

3 Data Plae Data I/O paths, trascodig, jitter maagemet Jitter Mask I-gress Queue Trascoder E-gress Queue Cotrol Plae Resource allocatio, sessio maiteace ad call routig Route Computig Flow Request Lik Measuremet Heartbeat Other surrogates Peer Table Path Broadcast Figure 2: The key modules of a surrogate. streams from its correspodig mobile user to other mobile users, ad participates i the costructio of optimal video dissemiatio trees. It also moitors the call qualities ad determies the best video ecodig parameters (codecs, bitrates, etc.) for streams from/to its mobile device. Data Plae is resposible for processig i/out video streams, i terms of both trascodig ad forwardig, as directed by the cotrol plae. The video stream from its mobile user is captured cotiuously ad dissemiated to other surrogates after ecessary trascodig. I the reverse directio, all video streams from other mobile users, via their respective surrogates, are trascoded ito appropriate rates (if ecessary) ad delivered to the mobile user by a key module jitter mask, which deals with radom jitters caused by fluctuatios of processig ad etwork latecies, as well as ay aomalies alog the dissemiatio paths. Our desig of vskycof observes the followig priciples. Decetralized Cotrol. Except ecessary bootstrappig from the gateway server, each sessio is to be maitaied by the surrogate of the iitiator of a coferece sessio, i order to provide good scalability ad flexibility. The video routig ad trascodig decisios are to be made i a fully distributed fashio by collaboratios amog surrogates. Self-Evolvig Routig Topology with Full Adaptivity. We seek to build a best routig topology amog the surrogates for dissemiatig the stream from each participat, which achieves a small ed-to-ed latecy ad fully exploits the available badwidths amog the surrogates. Trascodig decisios to covert the origial stream to acceptable formats/bit rates of the recipiets are dyamically made at the best poits alog the dissemiatio paths, accordig to differet computatio capacities of the surrogates ad eeds of dowstream mobile devices. Robust, Smooth Video Streamig. To guaratee smooth stream playback at each mobile user eve i cases of iaccurate route computatio (e.g., due to iaccurate estimates of lik badwidths ad latecies), we seek to desig a advaced error correctio mechaism to search for better routig paths before the call quality drops, by moitorig a carefully desiged jitter buffer at each surrogate. 4. DETAILED DESIGN 4.1 Sessio Maiteace Establishmet: Whe a mobile user logs i to the vskycof system via the gateway server, it is assiged a surrogate VM. The gateway ca maitai iformatio o a pool of available, pre-iitiated VMs i the IaaS cloud, ad assig oe from the pool to a mobile user based o geographic proximity of the two, to expedite the service. The surrogate of the sessio iitiator fids out IP addresses of surrogates of the other olie users from the gateway server, which it wishes to ivite to joi the video coferecig sessio. The iitiator the cotacts ad ivites the iterested participats through their surrogates directly, ad maitais a list of IP addresses of all active surrogates i the sessio. Tear-dow: Whe a mobile user leaves the system, its surrogate VM is released ad retured to the pool of available VMs i the IaaS cloud. If the iitiator of a sessio departs, its hostig role is haded over to aother substitute surrogate i the participat list. 4.2 Routig Computatio I a video coferecig sessio with S users, there are S streams, each produced by oe of the mobile users, to be delivered to all the other users. We model a mathematical optimizatio problem for costructig efficiet dissemiatio topologies of all streams i a sessio ad decidig the optimal trascodig locatios. We the desig efficiet, fully distributed heuristic to approach the optimal solutio i a dyamic system. For trascodig, we practically oly cosider dow-samplig of a stream, i.e., the reductio of streamig bit rate, but ot the reverse, sice up-samplig provides o quality improvemet but cosumes uecessary badwidth. We also focus o trascodig due to mismatched bit rates of streams of the same format, while the case of trascodig from oe format to aother ca be readily addressed with similar efforts Optimizatio Formulatio Let graph G =(S, E) represet the etwork of surrogates i a sessio, where S is the set of surrogates ad E is the set of directed coectios amog the surrogates. For each surrogate m 2 S, let ˆm represet the correspodig mobile user. Let S = S. Suppose C ij is the maximum available badwidth o lik (i, j) 2E, ad d ij deotes the lik latecy. We refer to the stream from a surrogate m 2Sas flow m, with source rate R (m) ˆm, which is the rate of icomig stream from mobile user ˆm to surrogate m, determied by the source capturig rate by the user s mobile camera ad the uplik rate from the mobile user. Let R (m) ˆ be the maximum acceptable bit rate of flow m at mobile user ˆ, as decided by the last-mile dow-lik badwidth from surrogate to ˆ, ad the allocatio of this dow-lik badwidth amog streams from differet users, e.g., if user ˆ sizes playback widows of streams from S 1 1 S other coferece participats equally o its device scree, of 1 the dow-lik badwidth should be allocated to each stream. The multicast flow m from surrogate m to all other surrogates ca be viewed as cosistig of S 1 coceptual uicast flows [6], from m to each of the other surrogates, respectively. These coceptual uicast flows co-exist i the etwork without cotedig for lik badwidths, ad the multicast flow rate o a lik is the maximum of the rates of all the uicast flows goig alog this lik. For ease of trascodig implemetatio, we restrict each uicast flow from m to to be a itegral flow alog oe path with the ed-to-ed rate r (m), ad the multicast topology is the overlap of all the S 1 uicast flow paths. Let biary variable Iij m idicate whether the coceptual uicast flow from m to traverses lik (i, j) 2E, ad c (m) ij deote the actual rate of the multicast flow m o lik (i, j). Let fuctio ' (r 1,r 2) give the trascodig latecy at surrogate, if the rate r 1 of a igress flow received by is higher tha the rate r 2 of the egress flow from. ' (r 1,r 2) = if r 1 apple r 2. Typical trascodig steps are to decode the source stream of rate r 1 to a itermediate format, ad the re-ecode the stream from the itermedia format to the destiatio rate r 2 [8]. Hece, trascodig delay ' (r 1,r 2) is mootoously icreasig o both r 1 ad r 2, ad depeds o computatio capacity of the surrogate VM : the more powerful the VM is, the faster the trascodig ca be accomplished. The quality of service i the coferecig sessio relies o two aspects: (i) the ed-to-ed latecy ad (ii) the flow rate received by each participat for each flow. We boud the ed-to-ed latecy, 35

4 from the time a source surrogate m emits flow m to the time a receiver surrogate is ready to push the stream to its correspodig mobile user, by L (m), whose value is dyamically set as discussed i Sec Let U( r(m) R (m) ˆ ) be a icreasig, cocave utility fuctio o the rate of flow m received by surrogate, r (m). We maximize the aggregate utility of all receivers i all flows as our objective. The optimizatio problem is formulated i (1). subject to: Iij m i:(i,j)2e where (i,j)2e max m2s 2S,6=m I m k:(j,k)2e U( r(m) R (m) ˆ ) (1) = b m j, 8j, m, 2S,m6=, (2) Iij m r (m) apple c (m) ij, 8(i, j) 2E,m,2S,m6=, (3) c (m) ij apple C ij, 8(i, j) 2E, (4) I m ij d ij + m2s (i,j)2e k:(j,k)2e +' ( I m ij I m Ij m c (m) j,r(m) ˆ j:(j,)2e ' j(c (m) ij,c (m) ) ) apple L(m), 8m, 2S,m6=, (5) Iij m 2{, 1}, 8m, 2S,m6=, (i, j) 2E, (6) 8 >< b m j = >: apple r (m) apple r (m) 1, j = m 1, j =, otherwise apple R (m) ˆm, 8m, 2S, (7) apple R (m) ˆ, 8m, 2S, (8) Costraits (2) ad (6) eforce a sigle path for the uicast flow from surrogate m to, ad esures flow coservatio alog the path. Costrait (3) implies that the uicast flow from m to with is coceptual, hidde i the actual multicast flow m with rate c (m) ij, o each lik (i, j). Costrait (4) requires that the overall rate of actual flows from differet sources should ot exceed the capacity of each lik. Costrait (5) bouds the ed-to-ed delay alog the path from source surrogate m to receiver surrogate, which cosists of three parts: (i) the overall lik delay alog the rate r (m) d ij; (ii) the sum of potetial trascodig delay path, P (i,j)2e Im ij at itermediate surrogates j s alog the path, i.e., P (i,j)2e Pk:(j,k)2E Im ij I m ' j(c (m) ij,c (m) ), where a surrogate j is o the path if there exist eighborig liks (i, j) ad (j, k),. such that Iij m = 1 ad I m = 1, ad a trascodig delay occurs if the flow rate o (i, j), c (m) ij, is larger tha the flow rate o (j, k), c (m) ; (iii) the potetial trascodig delay at surrogate, ' P ( j:(j,)2e Im ), to trascode the received stream j c (m) j,r(m) ˆ to the maximum receivig rate allowed at mobile user ˆ, if eeded. Costraits (7) ad (8) restrict the ed-to-ed rate of virtual uicast flow from surrogate m to to be o larger tha the maximum sedig rate from mobile user ˆm ad the maximum receivig rate at mobile user ˆ. The solutios to the optimizatio problem, r (m), c (m) ij, I m ij, 8m, 2S, 6= m, (i, j) 2E, give us (i) the rate at which each mobile user ˆm should sed its stream to its surrogate m, which is the maximum of all coceptual uicast flow rates from m to the other surrogates, max 2S,6=m r (m) ; (ii) the delivery rate of flow m alog each lik (i, j) ad hece the flow routig topology amog the surrogates (c (m) ij =idicates flow m is ot to be routed over lik (i, j)); ad (iii) where the trascodig of each flow m should happe, i.e., a surrogate j where a egress flow rate c (m) is smaller tha the igress rate c (m) ij alog the same coceptual uicast path, should trascode flow m to the lower rate Distributed Heuristic The optimizatio problem (1) is o-covex with iteger variables.we desig a efficiet heuristic algorithm, as give i Alg. 1 ad Alg. 2, to decide flow routig, rate assigmet ad trascodig locatios i a fully distributed fashio. Algorithm 1 Flow Routig ad Rate Allocatio 1: Costruct shortest-path trees from each surrogate m, T (m) ; 2: if 9m, 2S,! (m) >L (m) the 3: No feasible solutio exists; retur ; 4: ed if 5: N ij := Number of dissemiatio trees o (i, j); 6: 8(a, b) 2 T (m),c (m) a,b := mi k2s,(i,j)2t (m) {R(m), C ij }; ˆk N ij 7: Search for better routig paths, followig Alg. 2; represets the overall latecy (icludig both lik ad ecessary trascodig latecies) for flow m from surrogate m to surrogate. A shortest-path tree is costructed from surrogate m to all the other surrogates, usig a distributed Bellmaford algorithm (Lie 1 i Alg. 1). If the overall lik latecy o the We first decide a basic, feasible dissemiatio topology for each flow m, o which the ed-to-ed delay costrait for each receiver, costrait (5), is satisfied. Though the optimizatio problem (1) does ot restrict the topologies ito trees, we seek to costruct a dissemiatio tree for each flow for ease of practical implemetatio. For cociseess,! (m), we kow that this pre-set ed-to-ed latecy boud is by o meas satisfiable, ad should be adjusted to a more reasoable value (Lies 2-4). We the decide a basic, ed-to-ed rate of flow m o this shortest path tree, from surrogate m to all the other surrogates: the capacity C ij of each lik (i, j) is evely divided by the (actual) flows geerated by differet surrogates, that pass through this lik; the ed-to-ed rate of each flow m is set to the rate allocated to this flow o the bottleeck lik its shortest-path tree spas (Lies 5-6). path from surrogates m to is larger tha L (m) Algorithm 2 Self-Evolvig Route/Rate Adjustmet at i Flow m 1: while 9(j, ) 2 T (m), c (m) j <R(m) do 2: if 9(i, k) 2 T (m), mi{c m ik, C k } >c (m) j the 3: :={} S {q :(, q) 2 T m }; 4: if 8p 2,! (m) p apple L (m) p the 5: T (m) := T (m) (j, )+(k, ); 6: ed if 7: ed if 8: ed while Based o the basic dissemiatio topology, each surrogate the carries out dyamic edge ad rate adjustmets by followig Alg. 2, to maximally utilize the available capacity to stream high-quality streams, without violatig the latecy costraits. For each flow 36

5 Mobile Device m m D = + + m Network Mobile Device B (m) Time Figure 3: A illustratio of the ed-to-ed delay for flow m. m, suppose surrogate j is the paret to surrogate o the curret dissemiatio tree of flow m. cotacts other eighborig surrogates i the flow, to discover if there is a better path from source surrogate m with higher capacity via aother paret k. It compares from j with the potetial receivig rate from k, mi(c (m) ik, C k ), where we suppose surrogate i is the paret of k i the curret tree, ad C k is the remaiig available badwidth o lik (k, ) (Lie 2 i Alg. 2). If the potetial receivig rate via k is larger, eeds to further evaluate the icreased latecy alog the ew path, due to chages of lik latecies ad potetial trascodig latecies at k ad. Oly if the latecy of the curret receivig rate c (m) j the ew path from m to, i.e.,! (m), is still withi L (m), ad the updated latecy to each of the descet surrogates from o the tree is still withi the respective delay boud, ca safely chage its paret from j to k (Lies 3-6). 4.3 Jitter Maskig I multi-party video coferecig, sychroizatio amog streams received at all users is crucial to users perceived quality of experiece. It is much desired that the video frames captured at all users at the same time, are played at all the recipiet user devices at the same time. We desig a effective bufferig mechaism, which collaborates with the routig algorithms, for this purpose. maitais a buffer B (m) for each stream m 2S/{} from each of the other surrogates. The buffer holds video packets of flow m, ready to be delivered to mobile device. vskypecof eforces a ed-to-ed delay of D, from whe a video frame is captured at oe mobile device, to the time it is sychroously played at all the other mobile devices. The value of D ca be set based o reasoable estimatio of the maximum delay betwee two mobile users i the system, ad should fall i the acceptable delay rage for real-time commuicatio. Let m idicate delay betwee mobile device ˆm ad its surrogate m, 8m 2S. For a frame i buffer B (m), which is produced at t at the source ˆm, it will be pushed out from the buffer o earlier tha t + L (m), where L (m) = D m, i order to guaratee playback of the frame at the mobile device ˆ at t + D (Fig. 3). If there were o jitter i the cloud, we could set the delay boud i optimizatio (1), used to fid the routig path from surrogate m to surrogate, to L (m) = L (m), ad rest assured that the buffer will ever starve. However, i a practical system, jitter may occur due to various reasos, e.g., variatio of trascodig delay at surrogates, iaccurate estimate of lik delay ad badwidth whe L (m) ruig our routig algorithm, etc. Hece, L (m) i the optimiza-, i order to absorb the iaccuracy ad jitter. A series of measuremet work [5] have show that jitter o a etwork path approximately follows a ormal distributio [11]. Let tio for route selectio should be set smaller tha L (m) be a radom variable, represetig the path delay from surrogate m to surrogate, such that J (m) N(µ, 2 ), where µ J (m) is the mea ad is the stadard deviatio. For a ormal distributio, we ca derive that 99.97% of the samples fall withi the rage of ( 1,µ +3.4 ). If we set L (m) to the mea µ i the path delay distributio while allowig L (m) = µ +3.4, we derive L (m) = L (m) 3.4. Usig this L (m) i solvig optimizatio (1), we ca make sure that 99.97% of the video packets, followig the path selected, ca be set out from surrogate by L (m), ad catch their playback deadlies at the mobile device ˆm. I vskycof, each surrogate dyamically estimates the delay variace alog the path from m to, based o iter-packet latecies of flow m it receives. It also observes the curret queueig delay i buffer B (m), ad adjusts L (m) used i path selectio accordig to L (m) = L (m) 3.4. That is, if there are less packets i the buffer caused by larger delay variace, it tues L (m) dow to be more striget o the latecy requiremet i the path selectio; otherwise, it tues L (m) up to explore paths with better badwidths. I this way, this bufferig mechaism at the surrogates collaborates with the routig algorithm, to deal with radomess i the system ad iaccuracy i the computatio, while maximally guarateeig sychroized playback of all streams at all the mobile devices. 5. PERFORMANCE EVALUATION We implemet a prototype of vskycof ad deploy it i Amazo Elastic Compute Cloud (EC2). s are provisioed from ap-southeast-1a (Sigapore), eu-west-1a (Irelad), us-west- 1b (Califoria) ad us-east-1a (Virgiia), for users located ear the respective regio. Each mobile user is emulated by a machie ear its assiged EC2 istace (withi 5 ms), where video frames are geerated at a costat rate aroud 768 kbps. We implemet a applicatio-layer packet cotroller to limit the uplik ad dowlik badwidths of each user withi the rage of [1.5, 2] Mbps the same as those o regular 3G cellular coectios. We apply the cocave fuctio log(x) as the utility fuctio i our routig computatio. Each surrogate dyamically measures the lik delays to its eighborig surrogates. The trascodig latecies are pre-evaluated o the VM istaces ad used i our routig computatio, for trascodig from 768kbps to 256kbps, from 768kbps to 128kbps, from 256kbps to 128kbps, respectively. At each user, the stream from oe of the other coferece participats is displayed i a large scree (correspodig to a maximal acceptable streamig rate of 768 kbps), ad streams from other participats are displayed usig smaller screes (correspodig to maximal acceptable streamig rates of 128 kbps or 256 kbps). A fixed 4 ms ed-to-ed delay (D i Sec. 4.3) is cofigured, ad the buffer for each flow at each surrogate is set to a size correspodig to 4ms stream playback. 5.1 Adaptive Flow Rates We test a video coferecig sessio amog 1 participats: 5 from Hog Kog, 1 from Europe, 2 from US West ad 2 from US East. As a potetial bottleeck for scalability, the surrogate of the sessio iitiator is resposible for sessio maiteace by exchagig messages with the other surrogates i the sessio. We therefore ivestigate the coferecig performace at the iitiator s surrogate: if its performace is satisfactory, the the performace at the other surrogates should be eve better. Fig. 4 plots the flow rates of streams from 3 out of the other 9 participats. Flow-b is the flow with a maximum streamig rate of 768kbps; flow-a ad flow-c correspod to a maximum streamig rate of 128kbps ad 256kbps respectively, with the user of the latter joiig the sessio at a later time. We ca see that each flow goes through a fast start stage, whe the basic stream dissem- 37

6 iatio topology is beig costructed (as itroduced i Sec. 4.2), ad the evolve towards their maximal acceptable rates. Fig. 5 presets the load i the jitter buffer for flow-b at the iitiator s surrogate, where we see that the bufferig level varies sigificatly whe flow-b takes a path with large delay jitters. Whe our routig algorithm redirects flow-b through a better path, a more stable bufferig level is achieved later o. Fig. 6 shows the latecy experieced by each flow, from the correspodig source surrogate to the iitiator s surrogate. We observe that latecies oly vary slightly wheever the routig paths are adjusted, ad ca well meet the ed-to-ed latecy required (4ms). Bitrates (kbps) Buffer Load (%) 1 5 Flow a Flow b Flow c Time (secods) Figure 4: Flow rates at the iitiator s surrogate. 1% 5% Time (secods) Figure 5: Load of flow-b s buffer at the iitiator s surrogate. Latecy (ms) Flow a Flow b Flow c Time (secods) Figure 6: Flow latecies at the iitiator s surrogate. 5.2 Compariso with a Uicast Solutio We ext evaluate the performace of vskycof agaist a uicast scheme typically applied i P2P video coferecig, where each flow is directly trasmitted from the source to the destiatio via the etwork. We establish a 3-user video coferecig sessio, ad emulate a 5-miute log coferecig sessio with oe user comig from each of the regios, Hog Kog, Europe ad west US. Fig. 7 shows the perceived ed-to-ed latecies of the two flows received at the Hog Kog user, where eu stads for Europe ad usw stads for west US. We ca see that the ed-to-ed latecy achieved with vskycof is geerally smaller, ad much more stable tha that achieved by the uicast solutio, verifyig the smooth stream playback experieced by vskycof users. This validates that our cloud-assisted desig is suitable to achieve high-quality video coferecig amog multiple mobile participats. 6. CONCLUSION AND FUTURE WORK This paper presets vskycof, a cloud-assisted mobile video coferecig solutio, desiged to fudametally improve the quality Latecy (ms) 1 5 uicast eu uicast usw vskycof eu vskycof usw Time (secods) Figure 7: Ed-to-ed delay at the Hog Kog user. ad scale of multi-party mobile video coferecig. We employ a virtual machie i a cloud ifrastructure as the proxy for each mobile user, to sed ad to receive coferecig streams, ad to trascode the streams ito proper formats/rates. We desig a fully decetralized, efficiet algorithm to decide the best paths of stream dissemiatio ad the most suitable surrogates for video trascodig alog the paths, ad tailor a bufferig mechaism o each surrogate to cooperate with efficiet stream distributio. We have implemeted the vskycof prototype o Amazo EC2 ad verified its excellet performace. I our ogoig work, we are implemetig vskycof o real mobile devices ad evaluatig its performace uder more dyamic settigs. 7. ACKNOWLEDGMENTS The research was supported i part by a grat from Hog Kog RGC uder the cotract HKU E. 8. REFERENCES [1] Survey o mobile video coferecig apps, [2]. Che, M. Che, B. Li, Y. Zhao, Y. Wu, ad J. Li. Celerity: a low-delay multi-party coferecig solutio. I Proceedigs of the 19th ACM iteratioal coferece o Multimedia, New York, NY, USA, 211. [3] Y. Feg, B. Li, ad B. Li. Airlift: Video coferecig as a cloud service usig iter-dataceter etworks bibtex. I Proceedigs of IEEE ICNP, 212. [4] Z. Huag, C. Mei, L. E. Li, ad T. Woo. Cloudstream: Deliverig high-quality streamig videos through a cloud-based svc proxy. I Proceedigs of INFOCOM, 211. [5] M. J. Karam ad F. A. Tobagi. Aalysis of delay ad delay jitter of voice traffic i the iteret. Computer Networks, 4(6): , Dec. 22. [6] Z. Li, B. Li, D. Jiag, ad L. C. Lau. O achievig optimal throughput with etwork codig. I Proceedigs of IEEE INFOCOM, 25. [7] C. Liag, M. Zhao, ad Y. Liu. Optimal badwidth sharig i multiswarm multiparty p2p video-coferecig systems. IEEE/ACM Tras. Netw., 19(6): , 211. [8] J. L. Ozer. Video Compressio for Flash, Apple Devices ad HTML5. Doceo Publishig, USA, 211. [9] M. Poec, S. Segupta, M. Che, J. Li, ad P. A. Chou. Optimizig multi-rate peer-to-peer video coferecig applicatios. IEEE Trasactios o Multimedia, 211. [1] Y. Wu, C. Wu, B. Li, ad F. C. Lau. vskycof: Cloud-assisted Multi-party Mobile Video Coferecig. Techical report, [11] E. R. Ziegel. Probability ad Statistics for Egieerig ad the Scieces (8th Ed.), by Jay L. Devore. America Statistical Associatio, eight editio,

Domain 1 - Describe Cisco VoIP Implementations

Domain 1 - Describe Cisco VoIP Implementations Maual ONT (642-8) 1-800-418-6789 Domai 1 - Describe Cisco VoIP Implemetatios Advatages of VoIP Over Traditioal Switches Voice over IP etworks have may advatages over traditioal circuit switched voice etworks.

More information

Recovery time guaranteed heuristic routing for improving computation complexity in survivable WDM networks

Recovery time guaranteed heuristic routing for improving computation complexity in survivable WDM networks Computer Commuicatios 30 (2007) 1331 1336 wwwelseviercom/locate/comcom Recovery time guarateed heuristic routig for improvig computatio complexity i survivable WDM etworks Lei Guo * College of Iformatio

More information

Domain 1 Components of the Cisco Unified Communications Architecture

Domain 1 Components of the Cisco Unified Communications Architecture Maual CCNA Domai 1 Compoets of the Cisco Uified Commuicatios Architecture Uified Commuicatios (UC) Eviromet Cisco has itroduced what they call the Uified Commuicatios Eviromet which is used to separate

More information

Modified Line Search Method for Global Optimization

Modified Line Search Method for Global Optimization Modified Lie Search Method for Global Optimizatio Cria Grosa ad Ajith Abraham Ceter of Excellece for Quatifiable Quality of Service Norwegia Uiversity of Sciece ad Techology Trodheim, Norway {cria, ajith}@q2s.tu.o

More information

June 3, 1999. Voice over IP

June 3, 1999. Voice over IP Jue 3, 1999 Voice over IP This applicatio ote discusses the Hypercom solutio for providig ed-to-ed Iteret protocol (IP) coectivity i a ew or existig Hypercom Hybrid Trasport Mechaism (HTM) etwork, reducig

More information

Evaluating Model for B2C E- commerce Enterprise Development Based on DEA

Evaluating Model for B2C E- commerce Enterprise Development Based on DEA , pp.180-184 http://dx.doi.org/10.14257/astl.2014.53.39 Evaluatig Model for B2C E- commerce Eterprise Developmet Based o DEA Weli Geg, Jig Ta Computer ad iformatio egieerig Istitute, Harbi Uiversity of

More information

(VCP-310) 1-800-418-6789

(VCP-310) 1-800-418-6789 Maual VMware Lesso 1: Uderstadig the VMware Product Lie I this lesso, you will first lear what virtualizatio is. Next, you ll explore the products offered by VMware that provide virtualizatio services.

More information

Domain 1: Designing a SQL Server Instance and a Database Solution

Domain 1: Designing a SQL Server Instance and a Database Solution Maual SQL Server 2008 Desig, Optimize ad Maitai (70-450) 1-800-418-6789 Domai 1: Desigig a SQL Server Istace ad a Database Solutio Desigig for CPU, Memory ad Storage Capacity Requiremets Whe desigig a

More information

On the Capacity of Hybrid Wireless Networks

On the Capacity of Hybrid Wireless Networks O the Capacity of Hybrid ireless Networks Beyua Liu,ZheLiu +,DoTowsley Departmet of Computer Sciece Uiversity of Massachusetts Amherst, MA 0002 + IBM T.J. atso Research Ceter P.O. Box 704 Yorktow Heights,

More information

Capacity of Wireless Networks with Heterogeneous Traffic

Capacity of Wireless Networks with Heterogeneous Traffic Capacity of Wireless Networks with Heterogeeous Traffic Migyue Ji, Zheg Wag, Hamid R. Sadjadpour, J.J. Garcia-Lua-Aceves Departmet of Electrical Egieerig ad Computer Egieerig Uiversity of Califoria, Sata

More information

Engineering Data Management

Engineering Data Management BaaERP 5.0c Maufacturig Egieerig Data Maagemet Module Procedure UP128A US Documetiformatio Documet Documet code : UP128A US Documet group : User Documetatio Documet title : Egieerig Data Maagemet Applicatio/Package

More information

A Combined Continuous/Binary Genetic Algorithm for Microstrip Antenna Design

A Combined Continuous/Binary Genetic Algorithm for Microstrip Antenna Design A Combied Cotiuous/Biary Geetic Algorithm for Microstrip Atea Desig Rady L. Haupt The Pesylvaia State Uiversity Applied Research Laboratory P. O. Box 30 State College, PA 16804-0030 haupt@ieee.org Abstract:

More information

IT Support. 020 8269 6878 n www.premierchoiceinternet.com n support@premierchoiceinternet.com. 30 Day FREE Trial. IT Support from 8p/user

IT Support. 020 8269 6878 n www.premierchoiceinternet.com n support@premierchoiceinternet.com. 30 Day FREE Trial. IT Support from 8p/user IT Support IT Support Premier Choice Iteret has bee providig reliable, proactive & affordable IT Support solutios to compaies based i Lodo ad the South East of Eglad sice 2002. Our goal is to provide our

More information

SYSTEM INFO. MDK - Multifunctional Digital Communications System. Efficient Solutions for Information and Safety

SYSTEM INFO. MDK - Multifunctional Digital Communications System. Efficient Solutions for Information and Safety Commuicatios Systems for Itercom, PA, Emergecy Call ad Telecommuicatios MDK - Multifuctioal Digital Commuicatios System SYSTEM INFO ms NEUMANN ELEKTRONIK GmbH Efficiet Solutios for Iformatio ad Safety

More information

Vladimir N. Burkov, Dmitri A. Novikov MODELS AND METHODS OF MULTIPROJECTS MANAGEMENT

Vladimir N. Burkov, Dmitri A. Novikov MODELS AND METHODS OF MULTIPROJECTS MANAGEMENT Keywords: project maagemet, resource allocatio, etwork plaig Vladimir N Burkov, Dmitri A Novikov MODELS AND METHODS OF MULTIPROJECTS MANAGEMENT The paper deals with the problems of resource allocatio betwee

More information

A model of Virtual Resource Scheduling in Cloud Computing and Its

A model of Virtual Resource Scheduling in Cloud Computing and Its A model of Virtual Resource Schedulig i Cloud Computig ad Its Solutio usig EDAs 1 Jiafeg Zhao, 2 Wehua Zeg, 3 Miu Liu, 4 Guagmig Li 1, First Author, 3 Cogitive Sciece Departmet, Xiame Uiversity, Xiame,

More information

The Fundamental Capacity-Delay Tradeoff in Large Mobile Ad Hoc Networks

The Fundamental Capacity-Delay Tradeoff in Large Mobile Ad Hoc Networks The Fudametal Capacity-Delay Tradeoff i Large Mobile Ad Hoc Networks Xiaoju Li ad Ness B. Shroff School of Electrical ad Computer Egieerig, Purdue Uiversity West Lafayette, IN 47907, U.S.A. {lix, shroff}@ec.purdue.edu

More information

Chatpun Khamyat Department of Industrial Engineering, Kasetsart University, Bangkok, Thailand ocpky@hotmail.com

Chatpun Khamyat Department of Industrial Engineering, Kasetsart University, Bangkok, Thailand ocpky@hotmail.com SOLVING THE OIL DELIVERY TRUCKS ROUTING PROBLEM WITH MODIFY MULTI-TRAVELING SALESMAN PROBLEM APPROACH CASE STUDY: THE SME'S OIL LOGISTIC COMPANY IN BANGKOK THAILAND Chatpu Khamyat Departmet of Idustrial

More information

Optimize your Network. In the Courier, Express and Parcel market ADDING CREDIBILITY

Optimize your Network. In the Courier, Express and Parcel market ADDING CREDIBILITY Optimize your Network I the Courier, Express ad Parcel market ADDING CREDIBILITY Meetig today s challeges ad tomorrow s demads Aswers to your key etwork challeges ORTEC kows the highly competitive Courier,

More information

Characterizing End-to-End Packet Delay and Loss in the Internet

Characterizing End-to-End Packet Delay and Loss in the Internet Characterizig Ed-to-Ed Packet Delay ad Loss i the Iteret Jea-Chrysostome Bolot Xiyu Sog Preseted by Swaroop Sigh Layout Itroductio Data Collectio Data Aalysis Strategy Aalysis of packet delay Aalysis of

More information

Multi-server Optimal Bandwidth Monitoring for QoS based Multimedia Delivery Anup Basu, Irene Cheng and Yinzhe Yu

Multi-server Optimal Bandwidth Monitoring for QoS based Multimedia Delivery Anup Basu, Irene Cheng and Yinzhe Yu Multi-server Optimal Badwidth Moitorig for QoS based Multimedia Delivery Aup Basu, Iree Cheg ad Yizhe Yu Departmet of Computig Sciece U. of Alberta Architecture Applicatio Layer Request receptio -coectio

More information

1 Computing the Standard Deviation of Sample Means

1 Computing the Standard Deviation of Sample Means Computig the Stadard Deviatio of Sample Meas Quality cotrol charts are based o sample meas ot o idividual values withi a sample. A sample is a group of items, which are cosidered all together for our aalysis.

More information

IntelliSOURCE Comverge s enterprise software platform provides the foundation for deploying integrated demand management programs.

IntelliSOURCE Comverge s enterprise software platform provides the foundation for deploying integrated demand management programs. ItelliSOURCE Comverge s eterprise software platform provides the foudatio for deployig itegrated demad maagemet programs. ItelliSOURCE Demad maagemet programs such as demad respose, eergy efficiecy, ad

More information

Domain 1: Configuring Domain Name System (DNS) for Active Directory

Domain 1: Configuring Domain Name System (DNS) for Active Directory Maual Widows Domai 1: Cofigurig Domai Name System (DNS) for Active Directory Cofigure zoes I Domai Name System (DNS), a DNS amespace ca be divided ito zoes. The zoes store ame iformatio about oe or more

More information

5 Boolean Decision Trees (February 11)

5 Boolean Decision Trees (February 11) 5 Boolea Decisio Trees (February 11) 5.1 Graph Coectivity Suppose we are give a udirected graph G, represeted as a boolea adjacecy matrix = (a ij ), where a ij = 1 if ad oly if vertices i ad j are coected

More information

Research Article Sign Data Derivative Recovery

Research Article Sign Data Derivative Recovery Iteratioal Scholarly Research Network ISRN Applied Mathematics Volume 0, Article ID 63070, 7 pages doi:0.540/0/63070 Research Article Sig Data Derivative Recovery L. M. Housto, G. A. Glass, ad A. D. Dymikov

More information

Taking DCOP to the Real World: Efficient Complete Solutions for Distributed Multi-Event Scheduling

Taking DCOP to the Real World: Efficient Complete Solutions for Distributed Multi-Event Scheduling Taig DCOP to the Real World: Efficiet Complete Solutios for Distributed Multi-Evet Schedulig Rajiv T. Maheswara, Milid Tambe, Emma Bowrig, Joatha P. Pearce, ad Pradeep araatham Uiversity of Souther Califoria

More information

Systems Design Project: Indoor Location of Wireless Devices

Systems Design Project: Indoor Location of Wireless Devices Systems Desig Project: Idoor Locatio of Wireless Devices Prepared By: Bria Murphy Seior Systems Sciece ad Egieerig Washigto Uiversity i St. Louis Phoe: (805) 698-5295 Email: bcm1@cec.wustl.edu Supervised

More information

Security Functions and Purposes of Network Devices and Technologies (SY0-301) 1-800-418-6789. Firewalls. Audiobooks

Security Functions and Purposes of Network Devices and Technologies (SY0-301) 1-800-418-6789. Firewalls. Audiobooks Maual Security+ Domai 1 Network Security Every etwork is uique, ad architecturally defied physically by its equipmet ad coectios, ad logically through the applicatios, services, ad idustries it serves.

More information

Baan Service Master Data Management

Baan Service Master Data Management Baa Service Master Data Maagemet Module Procedure UP069A US Documetiformatio Documet Documet code : UP069A US Documet group : User Documetatio Documet title : Master Data Maagemet Applicatio/Package :

More information

CS100: Introduction to Computer Science

CS100: Introduction to Computer Science Review: History of Computers CS100: Itroductio to Computer Sciece Maiframes Miicomputers Lecture 2: Data Storage -- Bits, their storage ad mai memory Persoal Computers & Workstatios Review: The Role of

More information

INVESTMENT PERFORMANCE COUNCIL (IPC)

INVESTMENT PERFORMANCE COUNCIL (IPC) INVESTMENT PEFOMANCE COUNCIL (IPC) INVITATION TO COMMENT: Global Ivestmet Performace Stadards (GIPS ) Guidace Statemet o Calculatio Methodology The Associatio for Ivestmet Maagemet ad esearch (AIM) seeks

More information

CHAPTER 3 DIGITAL CODING OF SIGNALS

CHAPTER 3 DIGITAL CODING OF SIGNALS CHAPTER 3 DIGITAL CODING OF SIGNALS Computers are ofte used to automate the recordig of measuremets. The trasducers ad sigal coditioig circuits produce a voltage sigal that is proportioal to a quatity

More information

Output Analysis (2, Chapters 10 &11 Law)

Output Analysis (2, Chapters 10 &11 Law) B. Maddah ENMG 6 Simulatio 05/0/07 Output Aalysis (, Chapters 10 &11 Law) Comparig alterative system cofiguratio Sice the output of a simulatio is radom, the comparig differet systems via simulatio should

More information

Automatic Tuning for FOREX Trading System Using Fuzzy Time Series

Automatic Tuning for FOREX Trading System Using Fuzzy Time Series utomatic Tuig for FOREX Tradig System Usig Fuzzy Time Series Kraimo Maeesilp ad Pitihate Soorasa bstract Efficiecy of the automatic currecy tradig system is time depedet due to usig fixed parameters which

More information

COMPARISON OF THE EFFICIENCY OF S-CONTROL CHART AND EWMA-S 2 CONTROL CHART FOR THE CHANGES IN A PROCESS

COMPARISON OF THE EFFICIENCY OF S-CONTROL CHART AND EWMA-S 2 CONTROL CHART FOR THE CHANGES IN A PROCESS COMPARISON OF THE EFFICIENCY OF S-CONTROL CHART AND EWMA-S CONTROL CHART FOR THE CHANGES IN A PROCESS Supraee Lisawadi Departmet of Mathematics ad Statistics, Faculty of Sciece ad Techoology, Thammasat

More information

In nite Sequences. Dr. Philippe B. Laval Kennesaw State University. October 9, 2008

In nite Sequences. Dr. Philippe B. Laval Kennesaw State University. October 9, 2008 I ite Sequeces Dr. Philippe B. Laval Keesaw State Uiversity October 9, 2008 Abstract This had out is a itroductio to i ite sequeces. mai de itios ad presets some elemetary results. It gives the I ite Sequeces

More information

C.Yaashuwanth Department of Electrical and Electronics Engineering, Anna University Chennai, Chennai 600 025, India..

C.Yaashuwanth Department of Electrical and Electronics Engineering, Anna University Chennai, Chennai 600 025, India.. (IJCSIS) Iteratioal Joural of Computer Sciece ad Iformatio Security, A New Schedulig Algorithms for Real Time Tasks C.Yaashuwath Departmet of Electrical ad Electroics Egieerig, Aa Uiversity Cheai, Cheai

More information

Data Center Ethernet Facilitation of Enterprise Clustering. David Flynn, Linux Networx Orlando, Florida March 16, 2004

Data Center Ethernet Facilitation of Enterprise Clustering. David Flynn, Linux Networx Orlando, Florida March 16, 2004 Data Ceter Etheret Facilitatio of Eterprise Clusterig David Fly, Liux Networx Orlado, Florida March 16, 2004 1 2 Liux Networx builds COTS based clusters 3 Clusters Offer Improved Performace Scalability

More information

Chapter 6: Variance, the law of large numbers and the Monte-Carlo method

Chapter 6: Variance, the law of large numbers and the Monte-Carlo method Chapter 6: Variace, the law of large umbers ad the Mote-Carlo method Expected value, variace, ad Chebyshev iequality. If X is a radom variable recall that the expected value of X, E[X] is the average value

More information

Mobility. Manage Agility with Our Technology

Mobility. Manage Agility with Our Technology Mobility Maage Agility with Our Techology Smart mobile strategies are redefiig busiess performace today. Mobility as a eabler is oe of our key busiess objectives. As a radical orgaizatio, we defie comprehesive

More information

Accurate and Efficient Traffic Monitoring Using Adaptive Non-linear Sampling Method

Accurate and Efficient Traffic Monitoring Using Adaptive Non-linear Sampling Method Accurate ad Efficiet Traffic Moitorig Usig Adaptive No-liear Samplig Method Chegche Hu, Sheg Wag, Jia Tia, Bi Liu Tsighua Uiversity Beijig, Chia, {hucc,wags,tiaj}@mails.tsighua.edu.c liub@tsighua.edu.c

More information

Optimal Adaptive Bandwidth Monitoring for QoS Based Retrieval

Optimal Adaptive Bandwidth Monitoring for QoS Based Retrieval 1 Optimal Adaptive Badwidth Moitorig for QoS Based Retrieval Yizhe Yu, Iree Cheg ad Aup Basu (Seior Member) Departmet of Computig Sciece Uiversity of Alberta Edmoto, AB, T6G E8, CAADA {yizhe, aup, li}@cs.ualberta.ca

More information

Department of Computer Science, University of Otago

Department of Computer Science, University of Otago Departmet of Computer Sciece, Uiversity of Otago Techical Report OUCS-2006-09 Permutatios Cotaiig May Patters Authors: M.H. Albert Departmet of Computer Sciece, Uiversity of Otago Micah Colema, Rya Fly

More information

Document Control Solutions

Document Control Solutions Documet Cotrol Solutios State of the art software The beefits of Assai Assai Software Services provides leadig edge Documet Cotrol ad Maagemet System software for oil ad gas, egieerig ad costructio. AssaiDCMS

More information

A Churn-prevented Bandwidth Allocation Algorithm for Dynamic Demands In IaaS Cloud

A Churn-prevented Bandwidth Allocation Algorithm for Dynamic Demands In IaaS Cloud A Chur-preveted Badwidth Allocatio Algorithm for Dyamic Demads I IaaS Cloud Jilei Yag, Hui Xie ad Jiayu Li Departmet of Computer Sciece ad Techology, Tsighua Uiversity, Beijig, P.R. Chia Tsighua Natioal

More information

How to read A Mutual Fund shareholder report

How to read A Mutual Fund shareholder report Ivestor BulletI How to read A Mutual Fud shareholder report The SEC s Office of Ivestor Educatio ad Advocacy is issuig this Ivestor Bulleti to educate idividual ivestors about mutual fud shareholder reports.

More information

E-Plex Enterprise Access Control System

E-Plex Enterprise Access Control System Eterprise Access Cotrol System Egieered for Flexibility Modular Solutio The Eterprise Access Cotrol System is a modular solutio for maagig access poits. Employig a variety of hardware optios, system maagemet

More information

Digital Enterprise Unit. White Paper. Web Analytics Measurement for Responsive Websites

Digital Enterprise Unit. White Paper. Web Analytics Measurement for Responsive Websites Digital Eterprise Uit White Paper Web Aalytics Measuremet for Resposive Websites About the Authors Vishal Machewad Vishal Machewad has over 13 years of experiece i sales ad marketig, havig worked as a

More information

Verifying the Availability of Cloud Applications

Verifying the Availability of Cloud Applications Melaie Siebehaar, Olga Wege, Roy Has, Hasa Terca, Ralf Steimetz: Verifyig the Availability of Cloud Applicatios. I: Proceedigs of the 3rd Iteratioal Coferece o Cloud Computig ad Services Sciece (CLOSER

More information

Full Lifecycle Project Cost Controls

Full Lifecycle Project Cost Controls Full Lifecycle Project Cost Cotrols EcoSys EPC is a ext geeratio plaig ad cost cotrols software solutio deliverig best practices for full lifecycle project cost maagemet i a itegrated, easy-to-use web

More information

Location, Location, Location! Modeling Data Proximity in the Cloud

Location, Location, Location! Modeling Data Proximity in the Cloud Locatio, Locatio, Locatio! Modelig Data Proximity i the Cloud Birjodh Tiwaa tiwaa@eecs.umich.edu Uiversity of Michiga rbor, MI Hitesh Ballai hiballa@microsoft.com Microsoft Research Cambridge, UK Mahesh

More information

Case Study. Normal and t Distributions. Density Plot. Normal Distributions

Case Study. Normal and t Distributions. Density Plot. Normal Distributions Case Study Normal ad t Distributios Bret Halo ad Bret Larget Departmet of Statistics Uiversity of Wiscosi Madiso October 11 13, 2011 Case Study Body temperature varies withi idividuals over time (it ca

More information

Online Banking. Internet of Things

Online Banking. Internet of Things Olie Bakig & The Iteret of Thigs Our icreasigly iteretcoected future will mea better bakig ad added security resposibilities for all of us. FROM DESKTOPS TO SMARTWATCHS Just a few years ago, Americas coducted

More information

Study on the application of the software phase-locked loop in tracking and filtering of pulse signal

Study on the application of the software phase-locked loop in tracking and filtering of pulse signal Advaced Sciece ad Techology Letters, pp.31-35 http://dx.doi.org/10.14257/astl.2014.78.06 Study o the applicatio of the software phase-locked loop i trackig ad filterig of pulse sigal Sog Wei Xia 1 (College

More information

INVESTMENT PERFORMANCE COUNCIL (IPC) Guidance Statement on Calculation Methodology

INVESTMENT PERFORMANCE COUNCIL (IPC) Guidance Statement on Calculation Methodology Adoptio Date: 4 March 2004 Effective Date: 1 Jue 2004 Retroactive Applicatio: No Public Commet Period: Aug Nov 2002 INVESTMENT PERFORMANCE COUNCIL (IPC) Preface Guidace Statemet o Calculatio Methodology

More information

client communication

client communication CCH Portal cliet commuicatio facig today s challeges Like most accoutacy practices, we ow use email for most cliet commuicatio. It s quick ad easy, but we do worry about the security of sesitive data.

More information

Skytron Asset Manager

Skytron Asset Manager Skytro Asset Maager Meet Asset Maager Skytro Asset Maager is a wireless, pateted RFID asset trackig techology specifically desiged for hospital facilities to deliver istat ROI withi a easy to istall, fully

More information

Domain 1: Identifying Cause of and Resolving Desktop Application Issues Identifying and Resolving New Software Installation Issues

Domain 1: Identifying Cause of and Resolving Desktop Application Issues Identifying and Resolving New Software Installation Issues Maual Widows 7 Eterprise Desktop Support Techicia (70-685) 1-800-418-6789 Domai 1: Idetifyig Cause of ad Resolvig Desktop Applicatio Issues Idetifyig ad Resolvig New Software Istallatio Issues This sectio

More information

Authentication - Access Control Default Security Active Directory Trusted Authentication Guest User or Anonymous (un-authenticated) Logging Out

Authentication - Access Control Default Security Active Directory Trusted Authentication Guest User or Anonymous (un-authenticated) Logging Out FME Server Security Table of Cotets FME Server Autheticatio - Access Cotrol Default Security Active Directory Trusted Autheticatio Guest User or Aoymous (u-autheticated) Loggig Out Authorizatio - Roles

More information

ContactPro Desktop for Multi-Media Contact Center

ContactPro Desktop for Multi-Media Contact Center CotactPro Desktop for Multi-Media Cotact Ceter CCT CotactPro (CP) is the perfect solutio for the aget desktop i a Avaya multimedia call ceter eviromet. CotactPro empowers agets to efficietly serve customers

More information

Professional Networking

Professional Networking Professioal Networkig 1. Lear from people who ve bee where you are. Oe of your best resources for etworkig is alumi from your school. They ve take the classes you have take, they have bee o the job market

More information

Tradigms of Astundithi and Toyota

Tradigms of Astundithi and Toyota Tradig the radomess - Desigig a optimal tradig strategy uder a drifted radom walk price model Yuao Wu Math 20 Project Paper Professor Zachary Hamaker Abstract: I this paper the author iteds to explore

More information

MTO-MTS Production Systems in Supply Chains

MTO-MTS Production Systems in Supply Chains NSF GRANT #0092854 NSF PROGRAM NAME: MES/OR MTO-MTS Productio Systems i Supply Chais Philip M. Kamisky Uiversity of Califoria, Berkeley Our Kaya Uiversity of Califoria, Berkeley Abstract: Icreasig cost

More information

Your organization has a Class B IP address of 166.144.0.0 Before you implement subnetting, the Network ID and Host ID are divided as follows:

Your organization has a Class B IP address of 166.144.0.0 Before you implement subnetting, the Network ID and Host ID are divided as follows: Subettig Subettig is used to subdivide a sigle class of etwork i to multiple smaller etworks. Example: Your orgaizatio has a Class B IP address of 166.144.0.0 Before you implemet subettig, the Network

More information

An Architecture for a Next Generation VoIP Transmission Systems

An Architecture for a Next Generation VoIP Transmission Systems A Architecture for a Next Geeratio VoIP Trasmissio Systems Christia Hoee 1, Kai Clüver 2, Ja Weil 2 1 Wilhelm-Schickard-Istitute, Uiversity of Tübige, Germay 2 Fachgebiet Nachrichteübertragug, Techical

More information

Convention Paper 6764

Convention Paper 6764 Audio Egieerig Society Covetio Paper 6764 Preseted at the 10th Covetio 006 May 0 3 Paris, Frace This covetio paper has bee reproduced from the author's advace mauscript, without editig, correctios, or

More information

LECTURE 13: Cross-validation

LECTURE 13: Cross-validation LECTURE 3: Cross-validatio Resampli methods Cross Validatio Bootstrap Bias ad variace estimatio with the Bootstrap Three-way data partitioi Itroductio to Patter Aalysis Ricardo Gutierrez-Osua Texas A&M

More information

OpenText Cloud Fax Sevices

OpenText Cloud Fax Sevices OpeText Cloud Fax Sevices The Market Leader i Cloud Fax Techology For over 25 years, OpeText Cloud Fax Services has helped may compaies go paperless with solutios that itegrate with both email ad back-ed

More information

Radio Dispatch Systems

Radio Dispatch Systems Radio Dispatch Systems ZETRON DISPATCH SOLUTIONS: AT THE CENTER OF YOUR CRITICAL OPERATIONS Your dispatch system is the ceterpoit through which your key operatios are coordiated ad cotrolled. That s why

More information

Multiplexers and Demultiplexers

Multiplexers and Demultiplexers I this lesso, you will lear about: Multiplexers ad Demultiplexers 1. Multiplexers 2. Combiatioal circuit implemetatio with multiplexers 3. Demultiplexers 4. Some examples Multiplexer A Multiplexer (see

More information

COMPUTING EFFICIENCY METRICS FOR SYNERGIC INTELLIGENT TRANSPORTATION SYSTEMS

COMPUTING EFFICIENCY METRICS FOR SYNERGIC INTELLIGENT TRANSPORTATION SYSTEMS Trasport ad Teleuicatio Vol, No 4, 200 Trasport ad Teleuicatio, 200, Volume, No 4, 66 74 Trasport ad Teleuicatio Istitute, Lomoosova, Riga, LV-09, Latvia COMPUTING EFFICIENCY METRICS FOR SYNERGIC INTELLIGENT

More information

5: Introduction to Estimation

5: Introduction to Estimation 5: Itroductio to Estimatio Cotets Acroyms ad symbols... 1 Statistical iferece... Estimatig µ with cofidece... 3 Samplig distributio of the mea... 3 Cofidece Iterval for μ whe σ is kow before had... 4 Sample

More information

Unicenter TCPaccess FTP Server

Unicenter TCPaccess FTP Server Uiceter TCPaccess FTP Server Release Summary r6.1 SP2 K02213-2E This documetatio ad related computer software program (hereiafter referred to as the Documetatio ) is for the ed user s iformatioal purposes

More information

Research Article Allocating Freight Empty Cars in Railway Networks with Dynamic Demands

Research Article Allocating Freight Empty Cars in Railway Networks with Dynamic Demands Discrete Dyamics i Nature ad Society, Article ID 349341, 12 pages http://dx.doi.org/10.1155/2014/349341 Research Article Allocatig Freight Empty Cars i Railway Networks with Dyamic Demads Ce Zhao, Lixig

More information

Major Coefficients Recovery: a Compressed Data Gathering Scheme for Wireless Sensor Network

Major Coefficients Recovery: a Compressed Data Gathering Scheme for Wireless Sensor Network This full text paper was peer reviewed at the directio of IEEE Commuicatios Society subject matter experts for publicatio i the IEEE Globecom proceedigs. Major Coefficiets Recovery: a Compressed Data Gatherig

More information

The Power of Free Branching in a General Model of Backtracking and Dynamic Programming Algorithms

The Power of Free Branching in a General Model of Backtracking and Dynamic Programming Algorithms The Power of Free Brachig i a Geeral Model of Backtrackig ad Dyamic Programmig Algorithms SASHKA DAVIS IDA/Ceter for Computig Scieces Bowie, MD sashka.davis@gmail.com RUSSELL IMPAGLIAZZO Dept. of Computer

More information

CCH CRM Books Online Software Fee Protection Consultancy Advice Lines CPD Books Online Software Fee Protection Consultancy Advice Lines CPD

CCH CRM Books Online Software Fee Protection Consultancy Advice Lines CPD Books Online Software Fee Protection Consultancy Advice Lines CPD Books Olie Software Fee Fee Protectio Cosultacy Advice Advice Lies Lies CPD CPD facig today s challeges As a accoutacy practice, maagig relatioships with our cliets has to be at the heart of everythig

More information

The analysis of the Cournot oligopoly model considering the subjective motive in the strategy selection

The analysis of the Cournot oligopoly model considering the subjective motive in the strategy selection The aalysis of the Courot oligopoly model cosiderig the subjective motive i the strategy selectio Shigehito Furuyama Teruhisa Nakai Departmet of Systems Maagemet Egieerig Faculty of Egieerig Kasai Uiversity

More information

Center, Spread, and Shape in Inference: Claims, Caveats, and Insights

Center, Spread, and Shape in Inference: Claims, Caveats, and Insights Ceter, Spread, ad Shape i Iferece: Claims, Caveats, ad Isights Dr. Nacy Pfeig (Uiversity of Pittsburgh) AMATYC November 2008 Prelimiary Activities 1. I would like to produce a iterval estimate for the

More information

A Network Monitoring System with a Peer-to- Peer Architecture

A Network Monitoring System with a Peer-to- Peer Architecture A Network Moitorig System with a Peer-to- Peer Architecture Paulo Salvador ad Rui Valadas Uiversity of Aveiro Istitute of Telecommuicatios Summary Motivatio Peer-to-peer architecture System elemets Data

More information

iprox sensors iprox inductive sensors iprox programming tools ProxView programming software iprox the world s most versatile proximity sensor

iprox sensors iprox inductive sensors iprox programming tools ProxView programming software iprox the world s most versatile proximity sensor iprox sesors iprox iductive sesors iprox programmig tools ProxView programmig software iprox the world s most versatile proximity sesor The world s most versatile proximity sesor Eato s iproxe is syoymous

More information

ODBC. Getting Started With Sage Timberline Office ODBC

ODBC. Getting Started With Sage Timberline Office ODBC ODBC Gettig Started With Sage Timberlie Office ODBC NOTICE This documet ad the Sage Timberlie Office software may be used oly i accordace with the accompayig Sage Timberlie Office Ed User Licese Agreemet.

More information

Subject CT5 Contingencies Core Technical Syllabus

Subject CT5 Contingencies Core Technical Syllabus Subject CT5 Cotigecies Core Techical Syllabus for the 2015 exams 1 Jue 2014 Aim The aim of the Cotigecies subject is to provide a groudig i the mathematical techiques which ca be used to model ad value

More information

Bio-Plex Manager Software

Bio-Plex Manager Software Multiplex Suspesio Array Bio-Plex Maager Software Extract Kowledge Faster Move Your Research Forward Bio-Rad cotiues to iovate where it matters most. With Bio-Plex Maager 5.0 software, we offer valuable

More information

Telecom. White Paper. Actionable Intelligence in the SDN Ecosystem: Optimizing Network Traffic through FRSA

Telecom. White Paper. Actionable Intelligence in the SDN Ecosystem: Optimizing Network Traffic through FRSA Telecom White Paper Actioable Itelligece i the SDN Ecosystem: Optimizig Network Traffic through FRSA About the Authors Sumit Kapoor Sumit is a solutio architect i the telecom busiess uit at Tata Cosultacy

More information

ANALYTICS. Insights that drive your business

ANALYTICS. Insights that drive your business ANALYTICS Isights that drive your busiess Eterprises are trasformig their busiesses by supplemetig their databases with real ad up-to-date customer data. Aalytics, as a catalyst, refies raw data ad aligs

More information

Chapter 7: Confidence Interval and Sample Size

Chapter 7: Confidence Interval and Sample Size Chapter 7: Cofidece Iterval ad Sample Size Learig Objectives Upo successful completio of Chapter 7, you will be able to: Fid the cofidece iterval for the mea, proportio, ad variace. Determie the miimum

More information

facing today s challenges As an accountancy practice, managing relationships with our clients has to be at the heart of everything we do.

facing today s challenges As an accountancy practice, managing relationships with our clients has to be at the heart of everything we do. CCH CRM cliet relatios facig today s challeges As a accoutacy practice, maagig relatioships with our cliets has to be at the heart of everythig we do. That s why our CRM system ca t be a bolt-o extra it

More information

Agenda. Outsourcing and Globalization in Software Development. Outsourcing. Outsourcing here to stay. Outsourcing Alternatives

Agenda. Outsourcing and Globalization in Software Development. Outsourcing. Outsourcing here to stay. Outsourcing Alternatives Outsourcig ad Globalizatio i Software Developmet Jacques Crocker UW CSE Alumi 2003 jc@cs.washigto.edu Ageda Itroductio The Outsourcig Pheomeo Leadig Offshore Projects Maagig Customers Offshore Developmet

More information

On the Use of Adaptive OFDM to Preserve Energy in Ad Hoc Wireless Networks

On the Use of Adaptive OFDM to Preserve Energy in Ad Hoc Wireless Networks O the Use of Adaptive OFDM to Preserve Eergy i Ad Hoc Wireless etworks Kamol Kaemarugsi ad Prashat Krishamurthy Telecommuicatios Program, School of Iformatio Sciece, Uiversity of Pittsburgh 135 orth Bellefield

More information

*The most important feature of MRP as compared with ordinary inventory control analysis is its time phasing feature.

*The most important feature of MRP as compared with ordinary inventory control analysis is its time phasing feature. Itegrated Productio ad Ivetory Cotrol System MRP ad MRP II Framework of Maufacturig System Ivetory cotrol, productio schedulig, capacity plaig ad fiacial ad busiess decisios i a productio system are iterrelated.

More information

Business Rules-Driven SOA. A Framework for Multi-Tenant Cloud Computing

Business Rules-Driven SOA. A Framework for Multi-Tenant Cloud Computing Lect. Phd. Liviu Gabriel CRETU / SPRERS evet Traiig o software services, Timisoara, Romaia, 6-10 dec 2010 www.feaa.uaic.ro Busiess Rules-Drive SOA. A Framework for Multi-Teat Cloud Computig Lect. Ph.D.

More information

Incremental calculation of weighted mean and variance

Incremental calculation of weighted mean and variance Icremetal calculatio of weighted mea ad variace Toy Fich faf@cam.ac.uk dot@dotat.at Uiversity of Cambridge Computig Service February 009 Abstract I these otes I eplai how to derive formulae for umerically

More information

International Journal on Emerging Technologies 1(2): 48-56(2010) ISSN : 0975-8364

International Journal on Emerging Technologies 1(2): 48-56(2010) ISSN : 0975-8364 e t Iteratioal Joural o Emergig Techologies (): 48-56(00) ISSN : 0975-864 Dyamic load balacig i distributed ad high performace parallel eterprise computig by embeddig MPI ad ope MP Sadip S. Chauha, Sadip

More information

AP Calculus AB 2006 Scoring Guidelines Form B

AP Calculus AB 2006 Scoring Guidelines Form B AP Calculus AB 6 Scorig Guidelies Form B The College Board: Coectig Studets to College Success The College Board is a ot-for-profit membership associatio whose missio is to coect studets to college success

More information

Optimization of Large Data in Cloud computing using Replication Methods

Optimization of Large Data in Cloud computing using Replication Methods Optimizatio of Large Data i Cloud computig usig Replicatio Methods Vijaya -Kumar-C, Dr. G.A. Ramachadhra Computer Sciece ad Techology, Sri Krishadevaraya Uiversity Aatapuramu, AdhraPradesh, Idia Abstract-Cloud

More information

Desktop Management. Desktop Management Tools

Desktop Management. Desktop Management Tools Desktop Maagemet 9 Desktop Maagemet Tools Mac OS X icludes three desktop maagemet tools that you might fid helpful to work more efficietly ad productively: u Stacks puts expadable folders i the Dock. Clickig

More information

Configuring Additional Active Directory Server Roles

Configuring Additional Active Directory Server Roles Maual Upgradig your MCSE o Server 2003 to Server 2008 (70-649) 1-800-418-6789 Cofigurig Additioal Active Directory Server Roles Active Directory Lightweight Directory Services Backgroud ad Cofiguratio

More information

Determining the sample size

Determining the sample size Determiig the sample size Oe of the most commo questios ay statisticia gets asked is How large a sample size do I eed? Researchers are ofte surprised to fid out that the aswer depeds o a umber of factors

More information