How To Plan A Network Wide Load Balancing Route For A Network Wde Network (Network)

Size: px
Start display at page:

Download "How To Plan A Network Wide Load Balancing Route For A Network Wde Network (Network)"

Transcription

1 Network-Wde Load Balancng Routng Wth Performance Guarantees Kartk Gopalan Tz-cker Chueh Yow-Jan Ln Florda State Unversty Stony Brook Unversty Telcorda Research Abstract As wde-area network connectvty becomes commodtzed, network servce provders are offerng premum servces that generate hgher revenues by supportng performance senstve traffc (such as voce, multmeda, and onlne tradng). An emergng example s a vrtual prvate network path wth qualty of servce (QoS) guarantees, or QVPN. The man techncal challenge n offerng the QVPN servce s how to allocate a physcal route for each QVPN so as to maxmze the total number of QVPNs that a gven physcal network nfrastructure can support smultaneously. We make the case that the key to addressng ths challenge s to mantan network-wde load balance when selectng QVPN routes. By ensurng that dfferent parts of the network are evenly loaded, no sngle crtcal lnk wll tend to become a bottleneck resource. Ths paper descrbes a Lnk Crtcalty Based Routng (LCBR) algorthm, whch acheves hgh network resource utlzaton effcency whle supportng QVPNs wth end-to-end delay and bandwdth guarantees. Usng a smple yet effectve metrc that accurately quantfes network-wde load balance, LCBR sgnfcantly mproves the total number of supported QVPNs when compared to exstng traffc engneerng approaches. I. INTRODUCTION As Internet connectvty becomes a commodty, large enterprses ncreasngly want hgh levels of performance and relablty guarantees for ther performance senstve traffc (such as voce, multmeda, onlne fnancal tradng or electronc commerce). Carrers and network servce provders are respondng to such demands wth QoS-guaranteed VPN (or QVPN) servce. Technologes such as Mult-Protocol Label Swtchng (MPLS) can meet the QoS requrements of QVPNs by mappng each QVPN nto a Label Swtched Path (LSP). However, a major challenge that carrers face today s how to map each QVPN nto a physcal network path such that as many QVPNs as possble can be supported on ts network nfrastructure, thus maxmzng the revenue base for the carrer. We make the case that key to ths problem s to mantan network-wde load balance durng the route selecton process. Wthout network-wde load balance, t s possble that crtcal lnks become saturated much earler than others, renderng large number of routes unusable. Whle a number of QoS routng [1] [8] and traffc engneerng approaches [9] [14] exst, none of them have attempted to explctly quantfy the noton of network-wde load balance under QoS constrants. Wthout a quanttatve metrc, t becomes dffcult, f not mpossble, to evaluate varous load balancng routng algorthms. In ths paper, we propose the Lnk Crtcalty Based Routng (LCBR) algorthm that explctly mantans network-wde load balance n the route selecton process. LCBR s the frst algorthm to explctly quantfy the noton of network-wde load balance and to use t systematcally n the selecton of routes for QVPNs. LCBR ncorporates a smple and yet effectve measure of lnk crtcalty that helps to capture the mportance of a lnk wth respect to ts current resdual capacty and expected future load. Lnk crtcalty s then used as the bass to derve a network-wde load balancng metrc that characterzes the degree of load balance across the network and encourages the selecton of less crtcal lnks n the routng process. LCBR routes ncomng QVPNs n an onlne fashon. In contrast to offlne schemes, onlne algorthms do not possess apror knowledge of future QVPN arrvals and employ ntellgent heurstcs to adapt to future resource demands. We develop upon an earler verson of LCBR algorthm that was frst brefly ntroduced n [15]. Ths paper s addtonal contrbutons nclude the detaled development and performance evaluaton of a comprehensve set of LCBR algorthms that provde bandwdth guarantees alone, bandwdth-delay guarantees, and relablty. We consder the wde-area physcal network managed by a carrer that has complete admnstratve control of resources n the network. Each customer s QVPN spans from one pont of presence of the carrer s network to another and, once actvated, lasts for several days, weeks, or months. A network management system montors the traffc matrx and per-lnk traffc load n real-tme and uses ths feedback to route QVPNs. II. QUANTIFYING NETWORK-WIDE LOAD BALANCE Consder the smple example n Fgure 1. We need to select a route for a QVPN F 1 between nodes S 1 and D 1. There are two canddate routes: (S 1, E, B, C, D, D 1 ) and (S 1, E, F, G, D, D 1 ). Whch of these two routes s better from perspectve of long-term network resource usage effcency? Suppose t s lkely that future QVPN requests may arrve between (S 2, D 2 ) and (S 3, D 3 ) as well, but we do not know the exact QoS requrements of these QVPNs. Then the better route to select for F 1 would be (S 1, E, F, G, D, D 1 ) because t leaves the resources along the lnks (B, C) and (C, D) free for future QVPN requests between (S 2, D 2 ) and (S 3, D 3 ). Hence, the routng algorthm should, as far as possble, avod overloadng the lnks whch are crtcal to a large number of source-destnaton pars, so that no sngle lnk becomes a bottleneck for the whole network. The fundamental challenge here s the followng: Wthout precse knowledge of future QVPN request pattern, how exactly can we determne the mportance of a network lnk and ts mpact on network-wde load balance? In ths secton, we answer ths queston by quantfyng the notons of lnk crtcalty,

2 S3 S2 S1 Fg. 1. A E Route 1 B F C D3 G Route 2 An example of load balancng route selecton problem. expected load and network-wde load balance. A. Lnk Crtcalty and Expected Load The mportance of a lnk n LCBR s measured by the future expected load on each lnk,.e. the expected amount of traffc between dfferent source-destnaton pars to be carred over each lnk. A lnk that carres hgher amount of traffc between dfferent source-destnaton nodes s consdered more crtcal than one that carres less. More formally, assume that a total of x network routes are possble between a source-destnaton par (s, d) and y of these routes pass through a lnk l. Then the crtcalty φ l (s, d) of the lnk l wth respect to sourcedestnaton par (s, d) s defned as the fracton y/x. The total expected load φ l on lnk l s defned as the fractonal sum of expected bandwdth demands on the lnk from all possble source-destnaton pars n the network. Snce φ l represents a future expected load on the lnk, we need to begn wth an ntal estmate of φ l and then refne t ncrementally at run-tme as successve QVPNs are admtted. The ntal estmate of φ l s computed usng a matrx of expected bandwdth demands B(s, d) between each source-destnaton par (s, d) n the network. B(s, d) can be obtaned from measured daly traffc profles and/or servce-level agreements. If B(s, d) were to dstrbute equally over each possble route between s and d, then the ntal estmate of φ l can be calculated as φ l = (s,d) φ l(s, d)b(s, d). Of course, the ntal estmate of φ l may not be accurate because (1) actual bandwdth demands may devate from the B(s, d) values and (2) equal-load dstrbuton assumpton may not hold at run-tme. Secton III descrbes how the ntal φ l values are ncrementally corrected at run-tme wth each new QVPN. Another mportant consderaton n computng φ l s that the total number of routes between any source-destnaton par grows quckly wth network sze and connectvty. In practce, we can tackle ths ssue as follows. (1) Snce only a small subset of all the n nodes n the network are typcally possble sources or destnatons for QVPN traffc, computng φ l does not nvolve an exhaustve computaton of all the n 2 possble values of φ l (s, d). (2) Between a gven source-destnaton par, we can restrct the choce of routes to k-shortest canddate routes (or dsjont route pars), where k s typcally small (around 5 n our evaluatons). (3) Fnally, the lnk crtcalty φ l (s, d) tself s largely a statc metrc that changes only when the topology tself changes. Thus the values of φ l can be perodcally precomputed (such as on a daly bass) and kept ready for use. D D1 D2 Lnk 2 Cost cost(2) Ideal Trajectory Current Operatng Pont (cost(1), cost(2) ) X ^ φ 2/C 2 ( φ 1/C φ 1, 2/C 2 ) cost = ^ X = (cost(1) φ 1/C1 ) ( cost(2) φ 2/C 2 ) φ 1/C1 cost(1) Lnk 1 Cost Fg. 2. Cost of a smple 2-lnk network. An n-dmensonal plot would represent a network wth n lnks. B. Metrc for Network-wde Load Balancng Let C l be the total bandwdth capacty of a lnk and R l be ts resdual capacty at any tme. We begn by defnng the dynamc cost of each lnk, cost(l) = φ l /R l, as the expected load per unt of avalable lnk capacty. Thus, a lnk wth smaller resdual capacty R l or larger expected load φ l wll be consdered more expensve n routng QVPNs. Networkwde load balancng depends upon the cumulatve mpact of not only the magntude of ndvdual lnk costs but also the varatons among them. Fgure 2 geometrcally llustrates ths relatonshp for a smple network wth two lnks that have expected loads φ 1 and φ 2. The term φ l /C l represents the mnmum value of the lnk cost when the resdual capacty s maxmum at R l = C l. For deal load balance, resdual lnk capactes R 1 and R 2 should deally evolve towards zero such that (cost(1), cost(2)) ndeed stays along the deal load-balance trajectory. At the same tme, n order to mnmze the amount of resources consumed, t may not always be possble to follow the deal trajectory. The next best alternatve s to select routes that mnmze the dstance between the current operatng pont of the network (cost(1), cost(2)) and dle-state operatng pont (φ 1 /C 1, φ 2 /C 2 ). We defne the extent of load balance n a network G as the squared magntude of the dstance vector between the actual and the dle-state operatng pont. cost(g) = ( cost(l) φ ) 2 l (1) C l l G III. LOAD BALANCING ROUTE SELECTION We now present the Prmary LCBR () algorthm that selects a sngle prmary route for a QVPN F N between source s and destnaton d that mnmzes the load balance metrc cost(g). F N requres two forms of QoS guarantees: (a) longterm bandwdth requrement ρ N of F N must be satsfed at each lnk along the route, and (b) end-to-end delay encountered by packets of F N should be smaller than D N. A. Wth Bandwdth Guarantees In ths secton, we consder the QVPNs that requre bandwdth guarantees alone. When we select any route for a QVPN F N, that requres a long-term bandwdth guarantee of ρ N, the resdual capacty R l of each lnk l n the selected route would decrease by ρ N. The contrbuton of each lnk to the network-wde metrc cost(g) s gven by (φ l /R l φ l /C l ) 2.

3 Correspondngly, cost(g) ncreases due to smaller resdual capacty R l along the selected lnks. In order to fnd the route whch produces the smallest ncrement n cost(g), we can frst elmnate lnks wth avalable bandwdth smaller than ρ N and then apply Djkstra s shortest path algorthm on the reduced network graph, where the weght of each lnk s defned as follows: w l = ( φl R l ρ N φ l C l ) 2 ( φl φ ) 2 l (2) R l C l The term w l represents ncrease n lnk l s contrbuton to cost(g) when F N s routed through l. It s straghtforward to see that the route X N wth the mnmum value of l X N w l s the one whch produces the least ncrease n cost(g). Every tme bandwdth ρ N s reserved along a route for a new QVPN F N, the expected load φ l of each lnk n the network needs to be dynamcally updated because (1) the actual traffc profle mght devate from orgnally expected traffc profle B(s, d) and (2) the orgnal φ l values were computed under the assumpton of equal traffc dstrbuton among all canddate routes. Ths update can be performed ncrementally and wth low overhead, wthout recomputng each φ l from scratch. For each lnk l, ts φ l value s ncrementally updated as follows: φ l = φ l + (1 φ l (s, d))ρ N l X N φ l = φ l φ l (s, d)ρ N l / X N (3) B. Wth Bandwdth and Delay Guarantees Consder a QVPN F N that requres an end-to-end delay guarantee of D N, n addton to a bandwdth guarantee of ρ N. Each lnk l s capable of supportng a range of queung delay bounds for F N dependng upon the amount of bandwdth reserved for F N at the lnk l. Ths rases a number of possbltes for parttonng the end-to-end delay budget of a QVPN among the ndvdual lnks of a route [8], [16], [17]. However, n order to perform the delay parttonng, we need to know the complete set of lnks along the selected route. Snce the classcal shortest path algorthm ncrementally bulds the route one lnk at a tme, t cannot handle the end-to-end delay parttonng problem mentoned above. At the same tme, t s also mpractcal to examne the delay parttonng along every possble canddate route between a gven source and destnaton because the number of routes between any source-destnaton par grows quckly wth network sze. In practce, shorter routes typcally tend to utlze fewer network resources n comparson to longer routes and hence t s more lkely that the route whch best mnmzes the network-wde cost metrc s one among the k-shortest canddate routes. The algorthm for bandwdth-delay guaranteed route selecton apples ths nsght to narrow down the set of canddate routes that can mnmze cost(g) to those havng fewer lnks. performs route selecton n two phases - offlne and onlne. In the offlne phase, performed once for the entre network, pre-computes the set of k- shortest canddate routes between each source and destnaton and computes the expected load φ l for each lnk based on the computed canddate routes. A set of k-shortest canddate routes can be pre-computed usng well known algorthms such as [18]. Fortunately, as results n Secton V-C wll demonstrate, a small value of k s suffcent n practce to acheve good network-wde load balance, whch also lowers the computaton overhead. The onlne phase of executes upon the arrval of each new route set up request. The algorthm frst computes cost(l) = φ l /R l for each lnk l n network usng the precomputed value φ l and current resdual capacty R l. For each pre-computed canddate route X between s and d, performs the followng sequence of three operatons. (1) It checks f the QoS requrements (D N, ρ N ) of F N can be satsfed by the avalable resources along route X. (2) If there are suffcent resources, then parttons the end-toend delay D N among the lnks of route X. Specfc delay parttonng algorthms are descrbed n detal n [8], [16], [17]. The result of delay parttonng s a bandwdth reservaton value ρ Nl ρ N for each lnk l along route X that guarantees a perlnk delay budget D Nl such that l X D Nl D N. (3) Next, recomputes the per-lnk remanng capacty R l and the projected value of cost(g) that would result f the route X s assgned to F N. The route set up request for F N s rejected f ether (a) no route X has suffcent resources to satsfy F N s QoS requrements or (b) the mnmum projected value of cost(g) for any route X s greater than a pre-defned cost threshold α. If these two checks do not reject the QVPN F N, then assgns F N to that route X N whch yelds the mnmum ncrement n value of cost(g). The φ l values are correspondngly updated as follows. φ l = φ l φ l (s, d)ρ N + ρ Nl l X N φ l = φ l φ l (s, d)ρ N l / X N (4) IV. PRIMARY-BACKUP ROUTE SELECTION Gven a new QVPN request F N, the goal of Prmary-Backup LCBR (PB-LCBR) algorthm s to select a dsjont prmarybackup route par (X N, Y N ) that mnmzes cost(g). The dsjont backup route Y N guarantees that, f at most one network element (a lnk or a node) fals and the faled element les on the prmary route X N, then F N s traffc would be dverted to Y N wth the same QoS guarantees on bandwdth ρ N and endto-end delay bound D N. The PB-LCBR algorthm smultaneously examnes both prmary and backup components of canddate route pars. The algorthm s smlar n structure to algorthm, although wth mportant varatons that deserve menton. When a QVPN F N requres only a bandwdth guarantee of ρ N, then PB-LCBR fnds the shortest path-par, rather than the shortest path, that ncreases cost(g) by the smallest margn. Specfcally, PB- LCBR selects the path-par (X N, Y N ) wth mnmum value of l (X N Y N ) w l, where w l s gven by Equaton 2. Correspondng to Djkstra s shortest path algorthm, [19] descrbes a shortest path-par algorthm. When F N requres an end-toend delay bound of D N n addton to bandwdth guarantee of ρ N, usng the shortest path-par algorthm s not feasble due to reasons mentoned earler n Secton III-B. In ths case,

4 PB-LCBR follows the framework of n Secton III-B wth two major varatons. The frst varaton s that the offlne phase of PB-LCBR pre-computes a set of k = k 1 k 2 canddate prmary-backup route pars for every source-destnaton par (as opposed to just k 1 canddate prmary routes n the case of P- LCBR). The second varaton s that the nput to the onlne phase conssts of route pars (X, Y ) that were pre-computed durng the offlne phase. The dfference from the onlne phase of s that for each canddate route par (X, Y ) (1) both prmary route X and backup route Y are checked to ensure that suffcent resources are avalable to satsfy new QVPN F N s QoS requrements and (2) the end-to-end delay requrement D N s parttoned along both X and Y. Fnally, the canddate route par (X N, Y N ) that yelds mnmum cost(g) s chosen for F N. In order to maxmze resource usage effcency along backup routes, we now ntroduce the noton of backup resource aggregaton whch attempts to share QVPN reservatons along common backup lnks. Backup resource aggregaton was frst proposed n the RAFT approach [20] n the context of QVPNs that requre bandwdth guarantees alone. Our addtonal contrbuton here s the applcaton of the resource aggregaton concept to both bandwdth guaranteed as well as delay-bandwdth guaranteed QVPNs wthn the comprehensve admsson control and resource allocaton framework of the LCBR algorthm. Two QVPNs are sad to ntersect wth each other at a network element e f both ther prmary routes pass through element e. Every lnk l has one prmary set, P rmary(l), that contans the IDs of all QVPNs whose prmary routes pass through the lnk l. In addton, each lnk has a total of (m + n) backup sets of QVPN reservaton, where each set corresponds to one network element; m s the number of lnks and n s the number of nodes n the entre network. The backup sets at any lnk l are represented by Backup(l, e), 1 e (m + n), where each backup set corresponds to one network element e. Recovery from falure of a sngle network element occurs as follows. Durng normal operatons, each lnk scheduler operates wth the reservatons for QVPNs n ts prmary set P rmary(l). Whenever a network element e fals, ts correspondng backup sets, Backup(l, e), are actvated at all the lnks l and the respectve lnk schedulers start operatng wth reservatons n prmary set P rmary(l) plus those n backup set Backup(l, e). In the meantme, the route selecton mechansm would attempt to recover completely by redscoverng new prmary and/or backup routes of those QVPNs that are affected by the falure of network element e. V. PERFORMANCE EVALUATION We now compare the performance of LCBR aganst two earler traffc engneerng based approaches the Wdest Shortest Path (WSP) [2] based algorthm and the Mnmum Interference Routng Algorthm (MIRA) [11] that were orgnally proposed for bandwdth guaranteed prmary route selecton. The prmary route selecton verson of WSP (P- WSP) works as follows. For QVPNs that requre bandwdth guarantees alone, selects that route whch has maxmum Prmary-Backup Route Selecton PB-WSP PB-MIRA PB-LCBR 3000 Fg. 3. Number of QVPNs admtted wth bandwdth guarantees alone. ρ avg = 10Mbps, k = Prmary-Backup Route Selecton PB-WSP PB-MIRA PB-LCBR 1000 Fg. 4. Number of QVPNs admtted wth bandwdth-delay guarantees. ρ avg = 10Mbps, D = 10ms, k = 5 resdual bottleneck lnk capacty from among all the feasble routes havng mnmum length. The prmary-backup WSP (PB- WSP), selects that feasble route par wth mnmum-length prmary path whch has maxmum bottleneck lnk capacty. For bandwdth-delay guaranteed route selecton, (PB- WSP) examnes a set of k canddate routes (route pars) and select the mnmum length canddate wth maxmum bottleneck capacty. MIRA defnes the weght of a lnk l as the number of source-destnaton pars whose mncuts nclude l. For bandwdth guaranteed route selecton, (PB-MIRA) selects the route (route par) wth mnmum sum of lnk weghts, as defned above. For bandwdth-delay guaranteed route selecton, (PB-MIRA) examnes a set of k canddate routes (route pars) and selects the one wth mnmum sum of lnk weghts. We developed a smulaton tool to compare the resource usage effcency of the three algorthms. In ths paper, we present smulaton results over the AT&T natonwde backbone topology wth 41 nodes and 64 lnks. Due to space constrants, the results for two other topologes the North Amercan IP backbone topology of Sprnt and a 5 5 Grd topology are presented n [21]. Lnk capactes are chosen as a mx from 5 Gbps to 20 Gbps and sources-destnaton pars are selected unformly. Bandwdth demand profle B(s, d), s unformly dstrbuted between 5 Gbps and 20 Gbps. New QVPNs request an average bandwdth guarantee of 10Mbps. Excludng sgnal propagaton delays, whch can typcally range anywhere around ms, the end-to-end queung delay budget of up to 20ms can be requested by each QVPN. QVPNs are constantly admtted tll the network saturates. Intra-path delay parttonng s performed usng LSS algorthm [16] and backup resource aggregaton s appled to all the three PB-* algorthms.

5 Std. Devaton n Lnk Loads (%) Fg. 5. Comparson of standard devaton n lnk loads for WSP, MIRA and LCBR. ρ avg = 10Mbps, D = 10ms, k = 5. Average number of flows admtted A. Effectveness of LCBR Algorthm Canddate Set Sze (k) Fg. 6. Number of QVPNs admtted vs. canddate set sze k. ρ avg = 10Mbps, D = 10ms. We frst provde a snapshot comparson of the performance of LCBR aganst WSP and MIRA. Fgure 3 and 4 plot the number of QVPNs admtted wth bandwdth guarantees and bandwdthdelay guarantees respectvely by dfferent algorthms for 20 smulaton runs, where each run uses a dfferent random seed to controls lnk capactes. The fgures demonstrate that n all scenaros, LCBR consstently admts more QVPNs than WSP and MIRA algorthms snce t bases routng decsons on a network-wde load balancng crtera. On the other hand, WSP performs only lmted load balancng by selectng wdest shortest route among all canddates. The performance of MIRA and WSP are n general close to each other. There are notceable dfferences n the relatve performance of algorthms wth and wthout delay guarantees. Specfcally, we fnd that performance of LCBR s sgnfcantly better compared to WSP and MIRA n the presence of delay requrement than wthout delay requrement. Ths s explaned by the fact that tght delay requrements of a QVPN F N would requre a bandwdth reservaton ρ Nl at each lnk l that s larger than ts average bandwdth requrement ρ N. Thus wth delay-guaranteed QVPNs, there s a tendency towards hgher network-wde load mbalance and the benefts from LCBR s load balancng approach become more evdent. Another notceable trend s that the relatve dfference n performance of LCBR compared to WSP and MIRA s smaller for prmarybackup route selecton. The dsjonted-ness requrement on each prmary-backup route pars reduces the number of good qualty canddates avalable for LCBR to choose from. B. Network-Wde Load Balance We now show that LCBR ndeed mantans better networkwde load balance. Fgure 5 shows that the standard devaton n fnal percentage loads among all the lnks n the network s ndeed consstently smaller for LCBR compared to MIRA and WSP. The reason for LCBR s smaller load devaton s that t explctly mnmzes the load-balance metrc cost(g). Whle the cost(g) values of WSP and MIRA saturate after admttng fewer QVPNs, LCBR contnues to admt more QVPNs. C. Canddate Set Sze Fgure 6 shows that, wth ncreasng canddate set sze k, the number of QVPNs admtted wth bandwdth-delay guarantees ncreases rapdly for LCBR and then quckly saturates. Increasng k mples more choces for selectng a route leadng to more admtted QVPNs. However, for larger k values, the canddate set now ncludes longer routes whch can consume more resources and hence are rarely selected. WSP and MIRA are hardly affected by varyng k. Ths shows that LCBR mproves performance wthout devatng too much from the shortest route t tends to choose among 4 or 5 shortest canddates. WSP does not examne the entre canddate set; ncreasng k helps WSP only f t results n ncluson of more feasble routes wth shortest path length. Smlarly, ncreasng k helps MIRA only f t results n ncluson of routes wth fewer lnks that mpact the maxflow between varous source-destnaton pars. D. End-to-end Delay Fgure 7 shows that, wth ncreasng end-to-end delay, number of QVPNs admtted ncreases ntally and then tends to saturate for all the algorthms. Intally, each QVPN s delay-derved bandwdth requrement s tghter than the average rate requrement and hence delay s the tghter requrement. For larger delay requrement, the Average number of flows admtted End-to-end Delay (ms) Fg. 7. Effect of end-to-end delay on average number of admtted QVPNs. delay-derved bandwdth becomes smaller than the average rate requrement. In the saturaton regon, QVPNs are essentally beng servced at ther average rates. Also, LCBR admts more QVPNs than WSP and MIRA across all delay ranges. E. Computaton Cost The computaton cost for LCBR on a typcal operatonal network s qute small. For algorthm, offlne computaton tme, on a 1.8GHz Pentum 4 machne, vares from 3ms to 18ms as k ncreases from 2 to 10. For PB-LCBR, the correspondng varaton s from 10ms to 50ms. The onlne computaton cost s less than 2ms for and less than 4.5ms for PB- LCBR. The computaton cost s more for PB-LCBR because route selecton and QoS parttonng are performed for both prmary and backup routes. VI. RELATED WORK Tradtonal hop-by-hop shortest path (SP) algorthms [22] for best-effort traffc gnore the fact that lnks along a selected route mght already be congested whle non-congested alternatves may never be consdered. Wdest Shortest Path (WSP) [2] and several varatons [3], [22], [23] perform lmted load balancng and congeston management but fnally suffer the same fundamental problems as SP. QoS routng algorthms attempt to fnd a feasble short term route that satsfes ether sngle [1] [3] or multple [4] [8], [18], [24] QoS requrements, but not to optmze for long term traffc engneerng consderatons. On the other hand, traffc engneerng (TE) solutons attempt to acheve long term

6 network-wde resource management objectves. A class of TE schemes [13], [25], [26] manpulate lnk weghts to mantan load balance and mnmze congeston wth best-effort traffc, but do not support explct QoS guarantees. Mnmum Interference Routng Algorthm (MIRA) [11] s one of the frst explct routng TE schemes that selects the entre route for a QVPN n advance. MIRA uses a noton of mncut-based lnk crtcalty (descrbed n Secton V) whch s useful n avodng lnks that mpact the carryng capacty between large number of ngress-egress pars, but does not dentfy mportant lnks that may not be part of any mncuts. Profle-Based Routng (PBR) [14] uses measured traffc profles to solve a mult-commodty network flow problem and provde bandwdth guarantees. The key feature that dstngushes LCBR from the earler explct TE approaches s that LCBR s network cost metrc quantfes the mpact of all lnks on the networkwde load balance, whether or not they are part of any mncuts. VII. CONCLUSIONS In ths paper, we made the case that the key to maxmze network resource usage effcency under performance constrants s to acheve network-wde load balance every step of the way n the route selecton process. We have proposed the Lnk Crtcalty Based Routng (LCBR) algorthm, whch can select QoS guaranteed routes for wde-area QVPNs. To the best of our knowledge, our work s the frst to explctly quantfy the noton of network-wde load balance and use t systematcally n developng algorthms for traffc engneerng route selecton wth QoS guarantees. LCBR defnes a smple and yet effectve network-wde load balancng metrc that encourages the selecton of less crtcal lnks n the routng process. In our evaluatons, LCBR consstently supports more QVPNs under the same nput workloads, when compared aganst exstng traffc engneerng approaches. [12] H. Smt and T. L, IS-IS extensons for traffc engneerng, Internel Draft, August [13] A. Srdharan, R. Guern, and C. Dot, Achevng near optmal traffc engneerng solutons n current OSPF/ISIS networks, n Proc. of IEEE Infocom 2003, San Franssco, CA, March [14] S. Sur, M. Waldvogel, D. Bauer, and P.R. Warkhede, Profle-based routng and traffc engneerng, Computer Communcatons, vol. 26(4), pp , [15] K. Gopalan, T. Chueh, and Y.J. Ln, Load balancng routng wth bandwdth-delay guarantees, IEEE Communcatons Magazne, June [16] K. Gopalan and T. Chueh, Delay budget parttonng to maxmze network resource usage effcency, n Proc. IEEE INFOCOM 04, Hong Kong, Chna, March [17] R. Nagarajan, J. Kurose, and D. Towsley, Local allocaton of end-to-end qualty-of-servce n hgh-speed networks, n Proc. of Workshop on Perf. analyss of ATM Systems, Jan. 1993, pp [18] D. Eppsten, Fndng the k shortest paths, n Proc. of Symposum on Foundatons of Computer Scence, Nov. 1994, pp [19] J.W. Suurballe and R.E. Tarjan, A quck method for fndng shortest pars of dsjont paths, Networks, vol. 14, pp , [20] K. Dovrols and P. Ramanathan, Resource aggregaton for fault tolerance n ntegrated servces packet networks, ACM Computer Communcatons Revew, vol. 28(2), pp , Aprl [21] K. Gopalan, Effcent provsonng algorthms for network resource vrtualzaton wth QoS guarantees, Ph.D. Thess, Computer Scence Dept., Stony Brook Unversty, Aug [22] J. Moy, OSPF Verson 2, RFC 2328, Aprl [23] S. Blake, D. Black, D. Carlson, E. Daves, Z. Wang, and W. Wess, An archtecture for dfferentated servces, RFC 2475, Dec [24] R. Hassn, Approxmaton schemes for restrcted shortest path problem, Mathematcs of Operatons Research, vol. 17(1), pp , Feb [25] B. Fortz and M. Thorup, Optmzng OSPF/IS-IS weghts n a changng world, IEEE Journal on Selected Areas n Communcatons, vol. 20, pp , May [26] D. Awduche, A. Chu, A. Elwald, I. Wdjaja, and X. Xao, Overvew and prncples of nternet traffc engneerng, RFC 3272, May REFERENCES [1] R. Guern and A. Orda, QoS-based routng n networks wth naccurate nformaton, n Proc. of IEEE INFOCOM 97, Japan, Aprl [2] R. Guern, A. Orda, and D. Wllams, QoS routng mechansms and OSPF extensons, n Proc. of IEEE GLOBECOM 97, Phoenx, AZ, Nov. 1997, vol. 3, pp [3] S. Plotkn, Compettve routng of vrtual crcuts n ATM networks, IEEE J. Selected Areas n Comm., vol. 13(6), pp , [4] S. Chen and K. Nahrstedt, On fndng mult-constraned paths, n Proc. IEEE ICC 98, June [5] Z. Wang and J. Crowcroft, Bandwdth-delay based routng algorthms, n Proc. of IEEE GLOBECOM 95, Sngapore, Nov. 1995, pp [6] D. Raz and Y. Shavtt, Optmal partton of QoS requrements wth dscrete cost functons, n Proc. of INFOCOM, Israel, March [7] F. Ergun, R. Snha, and L. Zhang, QoS routng wth performance dependent costs, n Proc. of INFOCOM, Israel, March [8] D.H. Lorenz, A. Orda, D. Raz, and Y. Shavtt, Effcent QoS partton and routng of uncast and multcast, n Proc. of IWQoS 2000, Pttsburgh, PA, June 2000, pp [9] J. Boyle, V. Gll, A. Hannan, D. Cooper, D.Awduche, B. Chrstan, and W.S. La, Applcablty statement for traffc engneerng wth MPLS, RFC 3346, August [10] B. Fortz, J. Rexford, and M. Thorup, Traffc engneerng wth tradtonal IP routng protocols, IEEE Communcatons Magazne, Oct [11] M.S. Kodalam and T.V. Lakshman, Mnmum nterference routng wth applcatons to MPLS traffc engneerng, n Proc. of INFOCOM, Israel, March 2000, pp

Survey on Virtual Machine Placement Techniques in Cloud Computing Environment

Survey on Virtual Machine Placement Techniques in Cloud Computing Environment Survey on Vrtual Machne Placement Technques n Cloud Computng Envronment Rajeev Kumar Gupta and R. K. Paterya Department of Computer Scence & Engneerng, MANIT, Bhopal, Inda ABSTRACT In tradtonal data center

More information

PAS: A Packet Accounting System to Limit the Effects of DoS & DDoS. Debish Fesehaye & Klara Naherstedt University of Illinois-Urbana Champaign

PAS: A Packet Accounting System to Limit the Effects of DoS & DDoS. Debish Fesehaye & Klara Naherstedt University of Illinois-Urbana Champaign PAS: A Packet Accountng System to Lmt the Effects of DoS & DDoS Debsh Fesehaye & Klara Naherstedt Unversty of Illnos-Urbana Champagn DoS and DDoS DDoS attacks are ncreasng threats to our dgtal world. Exstng

More information

taposh_kuet20@yahoo.comcsedchan@cityu.edu.hk rajib_csedept@yahoo.co.uk, alam_shihabul@yahoo.com

taposh_kuet20@yahoo.comcsedchan@cityu.edu.hk rajib_csedept@yahoo.co.uk, alam_shihabul@yahoo.com G. G. Md. Nawaz Al 1,2, Rajb Chakraborty 2, Md. Shhabul Alam 2 and Edward Chan 1 1 Cty Unversty of Hong Kong, Hong Kong, Chna taposh_kuet20@yahoo.comcsedchan@ctyu.edu.hk 2 Khulna Unversty of Engneerng

More information

VoIP Playout Buffer Adjustment using Adaptive Estimation of Network Delays

VoIP Playout Buffer Adjustment using Adaptive Estimation of Network Delays VoIP Playout Buffer Adjustment usng Adaptve Estmaton of Network Delays Mroslaw Narbutt and Lam Murphy* Department of Computer Scence Unversty College Dubln, Belfeld, Dubln, IRELAND Abstract The poor qualty

More information

The Development of Web Log Mining Based on Improve-K-Means Clustering Analysis

The Development of Web Log Mining Based on Improve-K-Means Clustering Analysis The Development of Web Log Mnng Based on Improve-K-Means Clusterng Analyss TngZhong Wang * College of Informaton Technology, Luoyang Normal Unversty, Luoyang, 471022, Chna wangtngzhong2@sna.cn Abstract.

More information

Network Security Situation Evaluation Method for Distributed Denial of Service

Network Security Situation Evaluation Method for Distributed Denial of Service Network Securty Stuaton Evaluaton Method for Dstrbuted Denal of Servce Jn Q,2, Cu YMn,2, Huang MnHuan,2, Kuang XaoHu,2, TangHong,2 ) Scence and Technology on Informaton System Securty Laboratory, Bejng,

More information

What is Candidate Sampling

What is Candidate Sampling What s Canddate Samplng Say we have a multclass or mult label problem where each tranng example ( x, T ) conssts of a context x a small (mult)set of target classes T out of a large unverse L of possble

More information

Traffic State Estimation in the Traffic Management Center of Berlin

Traffic State Estimation in the Traffic Management Center of Berlin Traffc State Estmaton n the Traffc Management Center of Berln Authors: Peter Vortsch, PTV AG, Stumpfstrasse, D-763 Karlsruhe, Germany phone ++49/72/965/35, emal peter.vortsch@ptv.de Peter Möhl, PTV AG,

More information

Module 2 LOSSLESS IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur

Module 2 LOSSLESS IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur Module LOSSLESS IMAGE COMPRESSION SYSTEMS Lesson 3 Lossless Compresson: Huffman Codng Instructonal Objectves At the end of ths lesson, the students should be able to:. Defne and measure source entropy..

More information

Fault tolerance in cloud technologies presented as a service

Fault tolerance in cloud technologies presented as a service Internatonal Scentfc Conference Computer Scence 2015 Pavel Dzhunev, PhD student Fault tolerance n cloud technologes presented as a servce INTRODUCTION Improvements n technques for vrtualzaton and performance

More information

A New Paradigm for Load Balancing in Wireless Mesh Networks

A New Paradigm for Load Balancing in Wireless Mesh Networks A New Paradgm for Load Balancng n Wreless Mesh Networks Abstract: Obtanng maxmum throughput across a network or a mesh through optmal load balancng s known to be an NP-hard problem. Desgnng effcent load

More information

DEFINING %COMPLETE IN MICROSOFT PROJECT

DEFINING %COMPLETE IN MICROSOFT PROJECT CelersSystems DEFINING %COMPLETE IN MICROSOFT PROJECT PREPARED BY James E Aksel, PMP, PMI-SP, MVP For Addtonal Informaton about Earned Value Management Systems and reportng, please contact: CelersSystems,

More information

An Alternative Way to Measure Private Equity Performance

An Alternative Way to Measure Private Equity Performance An Alternatve Way to Measure Prvate Equty Performance Peter Todd Parlux Investment Technology LLC Summary Internal Rate of Return (IRR) s probably the most common way to measure the performance of prvate

More information

Enabling P2P One-view Multi-party Video Conferencing

Enabling P2P One-view Multi-party Video Conferencing Enablng P2P One-vew Mult-party Vdeo Conferencng Yongxang Zhao, Yong Lu, Changja Chen, and JanYn Zhang Abstract Mult-Party Vdeo Conferencng (MPVC) facltates realtme group nteracton between users. Whle P2P

More information

A Replication-Based and Fault Tolerant Allocation Algorithm for Cloud Computing

A Replication-Based and Fault Tolerant Allocation Algorithm for Cloud Computing A Replcaton-Based and Fault Tolerant Allocaton Algorthm for Cloud Computng Tork Altameem Dept of Computer Scence, RCC, Kng Saud Unversty, PO Box: 28095 11437 Ryadh-Saud Araba Abstract The very large nfrastructure

More information

IMPACT ANALYSIS OF A CELLULAR PHONE

IMPACT ANALYSIS OF A CELLULAR PHONE 4 th ASA & μeta Internatonal Conference IMPACT AALYSIS OF A CELLULAR PHOE We Lu, 2 Hongy L Bejng FEAonlne Engneerng Co.,Ltd. Bejng, Chna ABSTRACT Drop test smulaton plays an mportant role n nvestgatng

More information

benefit is 2, paid if the policyholder dies within the year, and probability of death within the year is ).

benefit is 2, paid if the policyholder dies within the year, and probability of death within the year is ). REVIEW OF RISK MANAGEMENT CONCEPTS LOSS DISTRIBUTIONS AND INSURANCE Loss and nsurance: When someone s subject to the rsk of ncurrng a fnancal loss, the loss s generally modeled usng a random varable or

More information

Availability-Based Path Selection and Network Vulnerability Assessment

Availability-Based Path Selection and Network Vulnerability Assessment Avalablty-Based Path Selecton and Network Vulnerablty Assessment Song Yang, Stojan Trajanovsk and Fernando A. Kupers Delft Unversty of Technology, The Netherlands {S.Yang, S.Trajanovsk, F.A.Kupers}@tudelft.nl

More information

On Multi-controller Placement Optimization in Software Defined Networking - based WANs

On Multi-controller Placement Optimization in Software Defined Networking - based WANs On Mult-controller Placement Optmzaton n Software Defned Networkng - based WANs Eugen Borcoc, Radu Badea, Serban Georgca Obreja, Marus Vochn Unversty POLITEHNICA of Bucharest - UPB Bucharest, Romana eugen.borcoc@elcom.pub.ro,

More information

An Interest-Oriented Network Evolution Mechanism for Online Communities

An Interest-Oriented Network Evolution Mechanism for Online Communities An Interest-Orented Network Evoluton Mechansm for Onlne Communtes Cahong Sun and Xaopng Yang School of Informaton, Renmn Unversty of Chna, Bejng 100872, P.R. Chna {chsun,yang}@ruc.edu.cn Abstract. Onlne

More information

A New Quality of Service Metric for Hard/Soft Real-Time Applications

A New Quality of Service Metric for Hard/Soft Real-Time Applications A New Qualty of Servce Metrc for Hard/Soft Real-Tme Applcatons Shaoxong Hua and Gang Qu Electrcal and Computer Engneerng Department and Insttute of Advanced Computer Study Unversty of Maryland, College

More information

Network Aware Load-Balancing via Parallel VM Migration for Data Centers

Network Aware Load-Balancing via Parallel VM Migration for Data Centers Network Aware Load-Balancng va Parallel VM Mgraton for Data Centers Kun-Tng Chen 2, Chen Chen 12, Po-Hsang Wang 2 1 Informaton Technology Servce Center, 2 Department of Computer Scence Natonal Chao Tung

More information

Open Access A Load Balancing Strategy with Bandwidth Constraint in Cloud Computing. Jing Deng 1,*, Ping Guo 2, Qi Li 3, Haizhu Chen 1

Open Access A Load Balancing Strategy with Bandwidth Constraint in Cloud Computing. Jing Deng 1,*, Ping Guo 2, Qi Li 3, Haizhu Chen 1 Send Orders for Reprnts to reprnts@benthamscence.ae The Open Cybernetcs & Systemcs Journal, 2014, 8, 115-121 115 Open Access A Load Balancng Strategy wth Bandwdth Constrant n Cloud Computng Jng Deng 1,*,

More information

A Performance Analysis of View Maintenance Techniques for Data Warehouses

A Performance Analysis of View Maintenance Techniques for Data Warehouses A Performance Analyss of Vew Mantenance Technques for Data Warehouses Xng Wang Dell Computer Corporaton Round Roc, Texas Le Gruenwald The nversty of Olahoma School of Computer Scence orman, OK 739 Guangtao

More information

Virtual Network Embedding with Coordinated Node and Link Mapping

Virtual Network Embedding with Coordinated Node and Link Mapping Vrtual Network Embeddng wth Coordnated Node and Lnk Mappng N. M. Mosharaf Kabr Chowdhury Cherton School of Computer Scence Unversty of Waterloo Waterloo, Canada Emal: nmmkchow@uwaterloo.ca Muntasr Rahan

More information

Project Networks With Mixed-Time Constraints

Project Networks With Mixed-Time Constraints Project Networs Wth Mxed-Tme Constrants L Caccetta and B Wattananon Western Australan Centre of Excellence n Industral Optmsaton (WACEIO) Curtn Unversty of Technology GPO Box U1987 Perth Western Australa

More information

Efficient Striping Techniques for Variable Bit Rate Continuous Media File Servers æ

Efficient Striping Techniques for Variable Bit Rate Continuous Media File Servers æ Effcent Strpng Technques for Varable Bt Rate Contnuous Meda Fle Servers æ Prashant J. Shenoy Harrck M. Vn Department of Computer Scence, Department of Computer Scences, Unversty of Massachusetts at Amherst

More information

On the Optimal Control of a Cascade of Hydro-Electric Power Stations

On the Optimal Control of a Cascade of Hydro-Electric Power Stations On the Optmal Control of a Cascade of Hydro-Electrc Power Statons M.C.M. Guedes a, A.F. Rbero a, G.V. Smrnov b and S. Vlela c a Department of Mathematcs, School of Scences, Unversty of Porto, Portugal;

More information

The Greedy Method. Introduction. 0/1 Knapsack Problem

The Greedy Method. Introduction. 0/1 Knapsack Problem The Greedy Method Introducton We have completed data structures. We now are gong to look at algorthm desgn methods. Often we are lookng at optmzaton problems whose performance s exponental. For an optmzaton

More information

A Design Method of High-availability and Low-optical-loss Optical Aggregation Network Architecture

A Design Method of High-availability and Low-optical-loss Optical Aggregation Network Architecture A Desgn Method of Hgh-avalablty and Low-optcal-loss Optcal Aggregaton Network Archtecture Takehro Sato, Kuntaka Ashzawa, Kazumasa Tokuhash, Dasuke Ish, Satoru Okamoto and Naoak Yamanaka Dept. of Informaton

More information

M3S MULTIMEDIA MOBILITY MANAGEMENT AND LOAD BALANCING IN WIRELESS BROADCAST NETWORKS

M3S MULTIMEDIA MOBILITY MANAGEMENT AND LOAD BALANCING IN WIRELESS BROADCAST NETWORKS M3S MULTIMEDIA MOBILITY MANAGEMENT AND LOAD BALANCING IN WIRELESS BROADCAST NETWORKS Bogdan Cubotaru, Gabrel-Mro Muntean Performance Engneerng Laboratory, RINCE School of Electronc Engneerng Dubln Cty

More information

A hybrid global optimization algorithm based on parallel chaos optimization and outlook algorithm

A hybrid global optimization algorithm based on parallel chaos optimization and outlook algorithm Avalable onlne www.ocpr.com Journal of Chemcal and Pharmaceutcal Research, 2014, 6(7):1884-1889 Research Artcle ISSN : 0975-7384 CODEN(USA) : JCPRC5 A hybrd global optmzaton algorthm based on parallel

More information

Efficient Bandwidth Management in Broadband Wireless Access Systems Using CAC-based Dynamic Pricing

Efficient Bandwidth Management in Broadband Wireless Access Systems Using CAC-based Dynamic Pricing Effcent Bandwdth Management n Broadband Wreless Access Systems Usng CAC-based Dynamc Prcng Bader Al-Manthar, Ndal Nasser 2, Najah Abu Al 3, Hossam Hassanen Telecommuncatons Research Laboratory School of

More information

CHOLESTEROL REFERENCE METHOD LABORATORY NETWORK. Sample Stability Protocol

CHOLESTEROL REFERENCE METHOD LABORATORY NETWORK. Sample Stability Protocol CHOLESTEROL REFERENCE METHOD LABORATORY NETWORK Sample Stablty Protocol Background The Cholesterol Reference Method Laboratory Network (CRMLN) developed certfcaton protocols for total cholesterol, HDL

More information

ANALYZING THE RELATIONSHIPS BETWEEN QUALITY, TIME, AND COST IN PROJECT MANAGEMENT DECISION MAKING

ANALYZING THE RELATIONSHIPS BETWEEN QUALITY, TIME, AND COST IN PROJECT MANAGEMENT DECISION MAKING ANALYZING THE RELATIONSHIPS BETWEEN QUALITY, TIME, AND COST IN PROJECT MANAGEMENT DECISION MAKING Matthew J. Lberatore, Department of Management and Operatons, Vllanova Unversty, Vllanova, PA 19085, 610-519-4390,

More information

Formulating & Solving Integer Problems Chapter 11 289

Formulating & Solving Integer Problems Chapter 11 289 Formulatng & Solvng Integer Problems Chapter 11 289 The Optonal Stop TSP If we drop the requrement that every stop must be vsted, we then get the optonal stop TSP. Ths mght correspond to a ob sequencng

More information

Resource Scheduling in Desktop Grid by Grid-JQA

Resource Scheduling in Desktop Grid by Grid-JQA The 3rd Internatonal Conference on Grd and Pervasve Computng - Worshops esource Schedulng n Destop Grd by Grd-JQA L. Mohammad Khanl M. Analou Assstant professor Assstant professor C.S. Dept.Tabrz Unversty

More information

J. Parallel Distrib. Comput. Environment-conscious scheduling of HPC applications on distributed Cloud-oriented data centers

J. Parallel Distrib. Comput. Environment-conscious scheduling of HPC applications on distributed Cloud-oriented data centers J. Parallel Dstrb. Comput. 71 (2011) 732 749 Contents lsts avalable at ScenceDrect J. Parallel Dstrb. Comput. ournal homepage: www.elsever.com/locate/pdc Envronment-conscous schedulng of HPC applcatons

More information

Efficient Project Portfolio as a tool for Enterprise Risk Management

Efficient Project Portfolio as a tool for Enterprise Risk Management Effcent Proect Portfolo as a tool for Enterprse Rsk Management Valentn O. Nkonov Ural State Techncal Unversty Growth Traectory Consultng Company January 5, 27 Effcent Proect Portfolo as a tool for Enterprse

More information

An Analysis of Central Processor Scheduling in Multiprogrammed Computer Systems

An Analysis of Central Processor Scheduling in Multiprogrammed Computer Systems STAN-CS-73-355 I SU-SE-73-013 An Analyss of Central Processor Schedulng n Multprogrammed Computer Systems (Dgest Edton) by Thomas G. Prce October 1972 Techncal Report No. 57 Reproducton n whole or n part

More information

The OC Curve of Attribute Acceptance Plans

The OC Curve of Attribute Acceptance Plans The OC Curve of Attrbute Acceptance Plans The Operatng Characterstc (OC) curve descrbes the probablty of acceptng a lot as a functon of the lot s qualty. Fgure 1 shows a typcal OC Curve. 10 8 6 4 1 3 4

More information

Forecasting the Direction and Strength of Stock Market Movement

Forecasting the Direction and Strength of Stock Market Movement Forecastng the Drecton and Strength of Stock Market Movement Jngwe Chen Mng Chen Nan Ye cjngwe@stanford.edu mchen5@stanford.edu nanye@stanford.edu Abstract - Stock market s one of the most complcated systems

More information

A Dynamic Energy-Efficiency Mechanism for Data Center Networks

A Dynamic Energy-Efficiency Mechanism for Data Center Networks A Dynamc Energy-Effcency Mechansm for Data Center Networks Sun Lang, Zhang Jnfang, Huang Daochao, Yang Dong, Qn Yajuan A Dynamc Energy-Effcency Mechansm for Data Center Networks 1 Sun Lang, 1 Zhang Jnfang,

More information

A 2 -MAC: An Adaptive, Anycast MAC Protocol for Wireless Sensor Networks

A 2 -MAC: An Adaptive, Anycast MAC Protocol for Wireless Sensor Networks : An Adaptve, Anycast MAC Protocol for Wreless Sensor Networks Hwee-Xan Tan and Mun Choon Chan Department of Computer Scence, School of Computng, Natonal Unversty of Sngapore {hweexan, chanmc}@comp.nus.edu.sg

More information

Traffic Engineering without Congestion Hot Spots in MPLS

Traffic Engineering without Congestion Hot Spots in MPLS Vol. 02, No. 07, 200, 2227-2233 Traffc Engneerng wthout Congeston Hot Spots n MPLS A.Padmaprya(Author) Assstant Professor Department of Computer Scence & Engneerng Alagappa Unversty, Karakud, Taml Nadu,

More information

Effective Network Defense Strategies against Malicious Attacks with Various Defense Mechanisms under Quality of Service Constraints

Effective Network Defense Strategies against Malicious Attacks with Various Defense Mechanisms under Quality of Service Constraints Effectve Network Defense Strateges aganst Malcous Attacks wth Varous Defense Mechansms under Qualty of Servce Constrants Frank Yeong-Sung Ln Department of Informaton Natonal Tawan Unversty Tape, Tawan,

More information

Fair Virtual Bandwidth Allocation Model in Virtual Data Centers

Fair Virtual Bandwidth Allocation Model in Virtual Data Centers Far Vrtual Bandwdth Allocaton Model n Vrtual Data Centers Yng Yuan, Cu-rong Wang, Cong Wang School of Informaton Scence and Engneerng ortheastern Unversty Shenyang, Chna School of Computer and Communcaton

More information

An Evaluation of the Extended Logistic, Simple Logistic, and Gompertz Models for Forecasting Short Lifecycle Products and Services

An Evaluation of the Extended Logistic, Simple Logistic, and Gompertz Models for Forecasting Short Lifecycle Products and Services An Evaluaton of the Extended Logstc, Smple Logstc, and Gompertz Models for Forecastng Short Lfecycle Products and Servces Charles V. Trappey a,1, Hsn-yng Wu b a Professor (Management Scence), Natonal Chao

More information

J. Parallel Distrib. Comput.

J. Parallel Distrib. Comput. J. Parallel Dstrb. Comput. 71 (2011) 62 76 Contents lsts avalable at ScenceDrect J. Parallel Dstrb. Comput. journal homepage: www.elsever.com/locate/jpdc Optmzng server placement n dstrbuted systems n

More information

Network Services Definition and Deployment in a Differentiated Services Architecture

Network Services Definition and Deployment in a Differentiated Services Architecture etwork Servces Defnton and Deployment n a Dfferentated Servces Archtecture E. kolouzou, S. Manats, P. Sampatakos,. Tsetsekas, I. S. Veners atonal Techncal Unversty of Athens, Department of Electrcal and

More information

A Novel Auction Mechanism for Selling Time-Sensitive E-Services

A Novel Auction Mechanism for Selling Time-Sensitive E-Services A ovel Aucton Mechansm for Sellng Tme-Senstve E-Servces Juong-Sk Lee and Boleslaw K. Szymansk Optmaret Inc. and Department of Computer Scence Rensselaer Polytechnc Insttute 110 8 th Street, Troy, Y 12180,

More information

How To Calculate The Accountng Perod Of Nequalty

How To Calculate The Accountng Perod Of Nequalty Inequalty and The Accountng Perod Quentn Wodon and Shlomo Ytzha World Ban and Hebrew Unversty September Abstract Income nequalty typcally declnes wth the length of tme taen nto account for measurement.

More information

Logical Development Of Vogel s Approximation Method (LD-VAM): An Approach To Find Basic Feasible Solution Of Transportation Problem

Logical Development Of Vogel s Approximation Method (LD-VAM): An Approach To Find Basic Feasible Solution Of Transportation Problem INTERNATIONAL JOURNAL OF SCIENTIFIC & TECHNOLOGY RESEARCH VOLUME, ISSUE, FEBRUARY ISSN 77-866 Logcal Development Of Vogel s Approxmaton Method (LD- An Approach To Fnd Basc Feasble Soluton Of Transportaton

More information

Robust Design of Public Storage Warehouses. Yeming (Yale) Gong EMLYON Business School

Robust Design of Public Storage Warehouses. Yeming (Yale) Gong EMLYON Business School Robust Desgn of Publc Storage Warehouses Yemng (Yale) Gong EMLYON Busness School Rene de Koster Rotterdam school of management, Erasmus Unversty Abstract We apply robust optmzaton and revenue management

More information

Politecnico di Torino. Porto Institutional Repository

Politecnico di Torino. Porto Institutional Repository Poltecnco d Torno Porto Insttutonal Repostory [Artcle] A cost-effectve cloud computng framework for acceleratng multmeda communcaton smulatons Orgnal Ctaton: D. Angel, E. Masala (2012). A cost-effectve

More information

An agent architecture for network support of distributed simulation systems

An agent architecture for network support of distributed simulation systems An agent archtecture for network support of dstrbuted smulaton systems Robert Smon, Mark Pullen and Woan Sun Chang Department of Computer Scence George Mason Unversty Farfax, VA, 22032 U.S.A. smon, mpullen,

More information

行 政 院 國 家 科 學 委 員 會 補 助 專 題 研 究 計 畫 成 果 報 告 期 中 進 度 報 告

行 政 院 國 家 科 學 委 員 會 補 助 專 題 研 究 計 畫 成 果 報 告 期 中 進 度 報 告 行 政 院 國 家 科 學 委 員 會 補 助 專 題 研 究 計 畫 成 果 報 告 期 中 進 度 報 告 畫 類 別 : 個 別 型 計 畫 半 導 體 產 業 大 型 廠 房 之 設 施 規 劃 計 畫 編 號 :NSC 96-2628-E-009-026-MY3 執 行 期 間 : 2007 年 8 月 1 日 至 2010 年 7 月 31 日 計 畫 主 持 人 : 巫 木 誠 共 同

More information

An Intelligent Policy System for Channel Allocation of Information Appliance

An Intelligent Policy System for Channel Allocation of Information Appliance Tamkang Journal of Scence and Engneerng, Vol. 5, No., pp. 63-68 (2002) 63 An Intellgent Polcy System for Channel Allocaton of Informaton Applance Cheng-Yuan Ku, Chang-Jnn Tsao 2 and Davd Yen 3 Department

More information

An MILP model for planning of batch plants operating in a campaign-mode

An MILP model for planning of batch plants operating in a campaign-mode An MILP model for plannng of batch plants operatng n a campagn-mode Yanna Fumero Insttuto de Desarrollo y Dseño CONICET UTN yfumero@santafe-concet.gov.ar Gabrela Corsano Insttuto de Desarrollo y Dseño

More information

A Secure Password-Authenticated Key Agreement Using Smart Cards

A Secure Password-Authenticated Key Agreement Using Smart Cards A Secure Password-Authentcated Key Agreement Usng Smart Cards Ka Chan 1, Wen-Chung Kuo 2 and Jn-Chou Cheng 3 1 Department of Computer and Informaton Scence, R.O.C. Mltary Academy, Kaohsung 83059, Tawan,

More information

APPLICATION OF PROBE DATA COLLECTED VIA INFRARED BEACONS TO TRAFFIC MANEGEMENT

APPLICATION OF PROBE DATA COLLECTED VIA INFRARED BEACONS TO TRAFFIC MANEGEMENT APPLICATION OF PROBE DATA COLLECTED VIA INFRARED BEACONS TO TRAFFIC MANEGEMENT Toshhko Oda (1), Kochro Iwaoka (2) (1), (2) Infrastructure Systems Busness Unt, Panasonc System Networks Co., Ltd. Saedo-cho

More information

An Adaptive Cross-layer Bandwidth Scheduling Strategy for the Speed-Sensitive Strategy in Hierarchical Cellular Networks

An Adaptive Cross-layer Bandwidth Scheduling Strategy for the Speed-Sensitive Strategy in Hierarchical Cellular Networks An Adaptve Cross-layer Bandwdth Schedulng Strategy for the Speed-Senstve Strategy n erarchcal Cellular Networks Jong-Shn Chen #1, Me-Wen #2 Department of Informaton and Communcaton Engneerng ChaoYang Unversty

More information

A generalized hierarchical fair service curve algorithm for high network utilization and link-sharing

A generalized hierarchical fair service curve algorithm for high network utilization and link-sharing Computer Networks 43 (2003) 669 694 www.elsever.com/locate/comnet A generalzed herarchcal far servce curve algorthm for hgh network utlzaton and lnk-sharng Khyun Pyun *, Junehwa Song, Heung-Kyu Lee Department

More information

Number of Levels Cumulative Annual operating Income per year construction costs costs ($) ($) ($) 1 600,000 35,000 100,000 2 2,200,000 60,000 350,000

Number of Levels Cumulative Annual operating Income per year construction costs costs ($) ($) ($) 1 600,000 35,000 100,000 2 2,200,000 60,000 350,000 Problem Set 5 Solutons 1 MIT s consderng buldng a new car park near Kendall Square. o unversty funds are avalable (overhead rates are under pressure and the new faclty would have to pay for tself from

More information

LAMOR: Lifetime-Aware Multipath Optimized Routing Algorithm for Video Transmission over Ad Hoc Networks

LAMOR: Lifetime-Aware Multipath Optimized Routing Algorithm for Video Transmission over Ad Hoc Networks LAMOR: Lfetme-Aware Multpath Optmzed Routng Algorthm for Vdeo ransmsson over Ad Hoc Networks 1 Lansheng an, Lng Xe, Kng-m Ko, Mng Le and Moshe Zukerman Abstract Multpath routng s a key technque to support

More information

Causal, Explanatory Forecasting. Analysis. Regression Analysis. Simple Linear Regression. Which is Independent? Forecasting

Causal, Explanatory Forecasting. Analysis. Regression Analysis. Simple Linear Regression. Which is Independent? Forecasting Causal, Explanatory Forecastng Assumes cause-and-effect relatonshp between system nputs and ts output Forecastng wth Regresson Analyss Rchard S. Barr Inputs System Cause + Effect Relatonshp The job of

More information

QoS-Aware Active Queue Management for Multimedia Services over the Internet

QoS-Aware Active Queue Management for Multimedia Services over the Internet QoS-Aware Actve Queue Management for Multmeda Servces over the Internet I-Shyan Hwang, *Bor-Junn Hwang, Pen-Mng Chang, Cheng-Yu Wang Abstract Recently, the multmeda servces such as IPTV, vdeo conference

More information

Capacity Reservation for Time-Sensitive Service Providers: An Application in Seaport Management

Capacity Reservation for Time-Sensitive Service Providers: An Application in Seaport Management Capacty Reservaton for Tme-Senstve Servce Provders: An Applcaton n Seaport Management L. Jeff Hong Department of Industral Engneerng and Logstcs Management The Hong Kong Unversty of Scence and Technology

More information

Study on Model of Risks Assessment of Standard Operation in Rural Power Network

Study on Model of Risks Assessment of Standard Operation in Rural Power Network Study on Model of Rsks Assessment of Standard Operaton n Rural Power Network Qngj L 1, Tao Yang 2 1 Qngj L, College of Informaton and Electrcal Engneerng, Shenyang Agrculture Unversty, Shenyang 110866,

More information

Institute of Informatics, Faculty of Business and Management, Brno University of Technology,Czech Republic

Institute of Informatics, Faculty of Business and Management, Brno University of Technology,Czech Republic Lagrange Multplers as Quanttatve Indcators n Economcs Ivan Mezník Insttute of Informatcs, Faculty of Busness and Management, Brno Unversty of TechnologCzech Republc Abstract The quanttatve role of Lagrange

More information

Performance Analysis of Energy Consumption of Smartphone Running Mobile Hotspot Application

Performance Analysis of Energy Consumption of Smartphone Running Mobile Hotspot Application Internatonal Journal of mart Grd and lean Energy Performance Analyss of Energy onsumpton of martphone Runnng Moble Hotspot Applcaton Yun on hung a chool of Electronc Engneerng, oongsl Unversty, 511 angdo-dong,

More information

Analysis of Energy-Conserving Access Protocols for Wireless Identification Networks

Analysis of Energy-Conserving Access Protocols for Wireless Identification Networks From the Proceedngs of Internatonal Conference on Telecommuncaton Systems (ITC-97), March 2-23, 1997. 1 Analyss of Energy-Conservng Access Protocols for Wreless Identfcaton etworks Imrch Chlamtac a, Chara

More information

How To Improve Delay Throughput In Wireless Networks With Multipath Routing And Channel Codeing

How To Improve Delay Throughput In Wireless Networks With Multipath Routing And Channel Codeing Delay-Throughput Enhancement n Wreless Networs wth Mult-path Routng and Channel Codng Kevan Ronas, Student Member, IEEE, Amr-Hamed Mohsenan-Rad, Member, IEEE, Vncent W.S. Wong, Senor Member, IEEE, Sathsh

More information

QOS DISTRIBUTION MONITORING FOR PERFORMANCE MANAGEMENT IN MULTIMEDIA NETWORKS

QOS DISTRIBUTION MONITORING FOR PERFORMANCE MANAGEMENT IN MULTIMEDIA NETWORKS QOS DISTRIBUTION MONITORING FOR PERFORMANCE MANAGEMENT IN MULTIMEDIA NETWORKS Yumng Jang, Chen-Khong Tham, Ch-Chung Ko Department Electrcal Engneerng Natonal Unversty Sngapore 119260 Sngapore Emal: {engp7450,

More information

Self-Motivated Relay Selection for a Generalized Power Line Monitoring Network

Self-Motivated Relay Selection for a Generalized Power Line Monitoring Network Self-Motvated Relay Selecton for a Generalzed Power Lne Montorng Network Jose Cordova and Xn Wang 1, Dong-Lang Xe 2, Le Zuo 3 1 Department of Electrcal and Computer Engneerng, State Unversty of New York

More information

Outsourcing inventory management decisions in healthcare: Models and application

Outsourcing inventory management decisions in healthcare: Models and application European Journal of Operatonal Research 154 (24) 271 29 O.R. Applcatons Outsourcng nventory management decsons n healthcare: Models and applcaton www.elsever.com/locate/dsw Lawrence Ncholson a, Asoo J.

More information

Activity Scheduling for Cost-Time Investment Optimization in Project Management

Activity Scheduling for Cost-Time Investment Optimization in Project Management PROJECT MANAGEMENT 4 th Internatonal Conference on Industral Engneerng and Industral Management XIV Congreso de Ingenería de Organzacón Donosta- San Sebastán, September 8 th -10 th 010 Actvty Schedulng

More information

Agile Traffic Merging for Data Center Networks. Qing Yi and Suresh Singh Portland State University, Oregon June 10 th, 2014

Agile Traffic Merging for Data Center Networks. Qing Yi and Suresh Singh Portland State University, Oregon June 10 th, 2014 Agle Traffc Mergng for Data Center Networks Qng Y and Suresh Sngh Portland State Unversty, Oregon June 10 th, 2014 Agenda Background and motvaton Power optmzaton model Smulated greedy algorthm Traffc mergng

More information

How To Understand The Results Of The German Meris Cloud And Water Vapour Product

How To Understand The Results Of The German Meris Cloud And Water Vapour Product Ttel: Project: Doc. No.: MERIS level 3 cloud and water vapour products MAPP MAPP-ATBD-ClWVL3 Issue: 1 Revson: 0 Date: 9.12.1998 Functon Name Organsaton Sgnature Date Author: Bennartz FUB Preusker FUB Schüller

More information

Joint Scheduling of Processing and Shuffle Phases in MapReduce Systems

Joint Scheduling of Processing and Shuffle Phases in MapReduce Systems Jont Schedulng of Processng and Shuffle Phases n MapReduce Systems Fangfe Chen, Mural Kodalam, T. V. Lakshman Department of Computer Scence and Engneerng, The Penn State Unversty Bell Laboratores, Alcatel-Lucent

More information

QoS-based Scheduling of Workflow Applications on Service Grids

QoS-based Scheduling of Workflow Applications on Service Grids QoS-based Schedulng of Workflow Applcatons on Servce Grds Ja Yu, Rakumar Buyya and Chen Khong Tham Grd Computng and Dstrbuted System Laboratory Dept. of Computer Scence and Software Engneerng The Unversty

More information

Dynamic Pricing for Smart Grid with Reinforcement Learning

Dynamic Pricing for Smart Grid with Reinforcement Learning Dynamc Prcng for Smart Grd wth Renforcement Learnng Byung-Gook Km, Yu Zhang, Mhaela van der Schaar, and Jang-Won Lee Samsung Electroncs, Suwon, Korea Department of Electrcal Engneerng, UCLA, Los Angeles,

More information

INVESTIGATION OF VEHICULAR USERS FAIRNESS IN CDMA-HDR NETWORKS

INVESTIGATION OF VEHICULAR USERS FAIRNESS IN CDMA-HDR NETWORKS 21 22 September 2007, BULGARIA 119 Proceedngs of the Internatonal Conference on Informaton Technologes (InfoTech-2007) 21 st 22 nd September 2007, Bulgara vol. 2 INVESTIGATION OF VEHICULAR USERS FAIRNESS

More information

Allocating Collaborative Profit in Less-than-Truckload Carrier Alliance

Allocating Collaborative Profit in Less-than-Truckload Carrier Alliance J. Servce Scence & Management, 2010, 3: 143-149 do:10.4236/jssm.2010.31018 Publshed Onlne March 2010 (http://www.scrp.org/journal/jssm) 143 Allocatng Collaboratve Proft n Less-than-Truckload Carrer Allance

More information

Master s Thesis. Configuring robust virtual wireless sensor networks for Internet of Things inspired by brain functional networks

Master s Thesis. Configuring robust virtual wireless sensor networks for Internet of Things inspired by brain functional networks Master s Thess Ttle Confgurng robust vrtual wreless sensor networks for Internet of Thngs nspred by bran functonal networks Supervsor Professor Masayuk Murata Author Shnya Toyonaga February 10th, 2014

More information

Reinforcement Learning for Quality of Service in Mobile Ad Hoc Network (MANET)

Reinforcement Learning for Quality of Service in Mobile Ad Hoc Network (MANET) Renforcement Learnng for Qualty of Servce n Moble Ad Hoc Network (MANET) *T.KUMANAN AND **K.DURAISWAMY *Meenaksh College of Engneerng West K.K Nagar, Cheena-78 **Dean/academc,K.S.R College of Technology,Truchengode

More information

1. Fundamentals of probability theory 2. Emergence of communication traffic 3. Stochastic & Markovian Processes (SP & MP)

1. Fundamentals of probability theory 2. Emergence of communication traffic 3. Stochastic & Markovian Processes (SP & MP) 6.3 / -- Communcaton Networks II (Görg) SS20 -- www.comnets.un-bremen.de Communcaton Networks II Contents. Fundamentals of probablty theory 2. Emergence of communcaton traffc 3. Stochastc & Markovan Processes

More information

denote the location of a node, and suppose node X . This transmission causes a successful reception by node X for any other node

denote the location of a node, and suppose node X . This transmission causes a successful reception by node X for any other node Fnal Report of EE359 Class Proect Throughput and Delay n Wreless Ad Hoc Networs Changhua He changhua@stanford.edu Abstract: Networ throughput and pacet delay are the two most mportant parameters to evaluate

More information

A DYNAMIC CRASHING METHOD FOR PROJECT MANAGEMENT USING SIMULATION-BASED OPTIMIZATION. Michael E. Kuhl Radhamés A. Tolentino-Peña

A DYNAMIC CRASHING METHOD FOR PROJECT MANAGEMENT USING SIMULATION-BASED OPTIMIZATION. Michael E. Kuhl Radhamés A. Tolentino-Peña Proceedngs of the 2008 Wnter Smulaton Conference S. J. Mason, R. R. Hll, L. Mönch, O. Rose, T. Jefferson, J. W. Fowler eds. A DYNAMIC CRASHING METHOD FOR PROJECT MANAGEMENT USING SIMULATION-BASED OPTIMIZATION

More information

Checkng and Testng in Nokia RMS Process

Checkng and Testng in Nokia RMS Process An Integrated Schedulng Mechansm for Fault-Tolerant Modular Avoncs Systems Yann-Hang Lee Mohamed Youns Jeff Zhou CISE Department Unversty of Florda Ganesvlle, FL 326 yhlee@cse.ufl.edu Advanced System Technology

More information

How To Solve An Onlne Control Polcy On A Vrtualzed Data Center

How To Solve An Onlne Control Polcy On A Vrtualzed Data Center Dynamc Resource Allocaton and Power Management n Vrtualzed Data Centers Rahul Urgaonkar, Ulas C. Kozat, Ken Igarash, Mchael J. Neely urgaonka@usc.edu, {kozat, garash}@docomolabs-usa.com, mjneely@usc.edu

More information

Minimal Coding Network With Combinatorial Structure For Instantaneous Recovery From Edge Failures

Minimal Coding Network With Combinatorial Structure For Instantaneous Recovery From Edge Failures Mnmal Codng Network Wth Combnatoral Structure For Instantaneous Recovery From Edge Falures Ashly Joseph 1, Mr.M.Sadsh Sendl 2, Dr.S.Karthk 3 1 Fnal Year ME CSE Student Department of Computer Scence Engneerng

More information

Multi-Resource Fair Allocation in Heterogeneous Cloud Computing Systems

Multi-Resource Fair Allocation in Heterogeneous Cloud Computing Systems 1 Mult-Resource Far Allocaton n Heterogeneous Cloud Computng Systems We Wang, Student Member, IEEE, Ben Lang, Senor Member, IEEE, Baochun L, Senor Member, IEEE Abstract We study the mult-resource allocaton

More information

Can Auto Liability Insurance Purchases Signal Risk Attitude?

Can Auto Liability Insurance Purchases Signal Risk Attitude? Internatonal Journal of Busness and Economcs, 2011, Vol. 10, No. 2, 159-164 Can Auto Lablty Insurance Purchases Sgnal Rsk Atttude? Chu-Shu L Department of Internatonal Busness, Asa Unversty, Tawan Sheng-Chang

More information

Cost Minimization using Renewable Cooling and Thermal Energy Storage in CDNs

Cost Minimization using Renewable Cooling and Thermal Energy Storage in CDNs Cost Mnmzaton usng Renewable Coolng and Thermal Energy Storage n CDNs Stephen Lee College of Informaton and Computer Scences UMass, Amherst stephenlee@cs.umass.edu Rahul Urgaonkar IBM Research rurgaon@us.bm.com

More information

Feature selection for intrusion detection. Slobodan Petrović NISlab, Gjøvik University College

Feature selection for intrusion detection. Slobodan Petrović NISlab, Gjøvik University College Feature selecton for ntruson detecton Slobodan Petrovć NISlab, Gjøvk Unversty College Contents The feature selecton problem Intruson detecton Traffc features relevant for IDS The CFS measure The mrmr measure

More information

Frequency Selective IQ Phase and IQ Amplitude Imbalance Adjustments for OFDM Direct Conversion Transmitters

Frequency Selective IQ Phase and IQ Amplitude Imbalance Adjustments for OFDM Direct Conversion Transmitters Frequency Selectve IQ Phase and IQ Ampltude Imbalance Adjustments for OFDM Drect Converson ransmtters Edmund Coersmeer, Ernst Zelnsk Noka, Meesmannstrasse 103, 44807 Bochum, Germany edmund.coersmeer@noka.com,

More information

Calculation of Sampling Weights

Calculation of Sampling Weights Perre Foy Statstcs Canada 4 Calculaton of Samplng Weghts 4.1 OVERVIEW The basc sample desgn used n TIMSS Populatons 1 and 2 was a two-stage stratfed cluster desgn. 1 The frst stage conssted of a sample

More information

Towards bandwidth guaranteed energy efficient data center networking

Towards bandwidth guaranteed energy efficient data center networking Wang et al. Journal of Cloud Computng: Advances, Systems and Applcatons (2015) 4:9 DOI 10.1186/s13677-015-0035-7 RESEARCH Open Access Towards bandwdth guaranteed energy effcent data center networkng Tng

More information

AN APPROACH TO WIRELESS SCHEDULING CONSIDERING REVENUE AND USERS SATISFACTION

AN APPROACH TO WIRELESS SCHEDULING CONSIDERING REVENUE AND USERS SATISFACTION The Medterranean Journal of Computers and Networks, Vol. 2, No. 1, 2006 57 AN APPROACH TO WIRELESS SCHEDULING CONSIDERING REVENUE AND USERS SATISFACTION L. Bada 1,*, M. Zorz 2 1 Department of Engneerng,

More information