Future Generation Computer Systems. Self-healing network for scalable fault-tolerant runtime environments

Size: px
Start display at page:

Download "Future Generation Computer Systems. Self-healing network for scalable fault-tolerant runtime environments"

Transcription

1 Fuure Generaion Compuer Sysems 26 (2010) Conens liss available a ScienceDirec Fuure Generaion Compuer Sysems journal homepage: Self-healing nework for scalable faul-oleran runime environmens Thara Angskun, Graham Fagg, George Bosilca, Jelena Pješivac-Grbović, Jack Dongarra Deparmen of Compuer Science, The Universiy of Tennessee, 1122 Voluneer Blvd. Knoxville, TN 37996, USA a r i c l e i n f o a b s r a c Aricle hisory: Received 7 December 2006 Received in revised form 17 Sepember 2007 Acceped 29 April 2009 Available online 9 May 2009 Keywords: Faul olerance Rouing proocols Runime environmens Scalabiliy Self-healing The number of processors embedded on high performance compuing plaforms is growing daily o saisfy he user desire for solving larger and more complex problems. Scalable and faul-oleran runime environmens are needed o suppor and adap o he underlying libraries and hardware which require a high degree of scalabiliy in dynamic large-scale environmens. This paper presens a self-healing nework (SHN) for supporing scalable and faul-oleran runime environmens. The SHN is designed o suppor ransmission of messages across muliple nodes while also proecing agains recursive node and process failures. I will auomaically recover iself afer a failure occurs. SHN is implemened on op of a scalable faul-oleran proocol (SFTP). The experimenal resuls show ha boh he laes mulicas and broadcas rouing algorihms used in SHN are faser and more reliable han he original SFTP rouing algorihms Elsevier B.V. All righs reserved. 1. Inroducion Recenly, several high performance compuing plaforms have been insalled wih more han 10,000 CPUs, such as Blue-Gene/L a LLNL, BGW a IBM and Columbia a NASA [1]. However, as he number of componens increases, so does he probabiliy of failure. To saisfy he requiremens of such a dynamic environmen (where he available number of resources is flucuaing), a scalable and faul-oleran framework is needed. Many large-scale applicaions are implemened on op of message passing sysems for which he de faco sandard is he Message Passing Inerface (MPI) [2]. MPI implemenaions require suppor from parallel runime environmens, which are exensions of he operaing sysem services, and provide necessary funcionaliies (such as naming resoluion services) for boh he message passing libraries and applicaions. Alhough here are several exising parallel runime environmens for differen ypes of sysems, hey do no mee some of he major requiremens for MPI implemenaions: scalabiliy, porabiliy and performance. Typically, disribued OS and single sysem image sysems are no porable while he naure of Grid middle-ware has performance problems. The MPICH implemenaion [3] uses a parallel runime environmen called Muli-Purposed Daemon (MPD) [4] o provide scalabiliy and faul olerance hrough a ring opology for some operaions and a ree opology for ohers. Runime environmens of oher MPI implemenaions, such as Harness [5] of FT-MPI [6], Open RTE [7] Corresponding auhor. Tel.: address: angskun@cs.uk.edu (T. Angskun). of Open MPI [8] and LAM of LAM/MPI [9], do no currenly provide scalable or faul-oleran soluions for heir inernal communicaions. The scalabiliy and faul-olerance issues have been addressed in several neworking areas. However, hese approaches canno be used or hey are no efficien in he parallel runime environmens. Srucured peer-o-peer neworking based on disribued hash ables such as CAN [10], Chord [11], Pasry [12] and Tapesry [13] were designed for resource discovery. They are only opimized for unicas messages. Techniques used in sensor or large-scale ad hoc neworking based on gossiping (or he epidemic algorihm) [14,15] mainly focus on informaion aggregaion. A self-healing nework (SHN) ha can be used as a basis for consrucing a higher level, faul-oleran parallel runime environmen is described in his aricle. SHN was designed o suppor ransferring messages across muliple nodes efficienly, while proecing agains recursive node or process failures. SHN auomaically recovers iself o overcome he orphan siuaion (he siuaion where nodes are unreachable because he nework is bisecioned). SHN was buil on op of a scalable and faul-oleran proocol (SFTP) [16] and auomaically recovers iself afer a failure occurs. The SFTP is based on a k-ary sibling ree. The k-ary sibling ree opology is a k-ary ree, where k is he number of fan-ous (k 2), and he nodes on he same level (same deph on he ree) are linked ogeher using a ring opology. The ree is primarily designed o allow scalabiliy for broadcas and mulicas operaions, while he ring is used o provide a well-undersood secondary pah for ransmission when he ree is damaged during failure condiions. The proocol has been formally proven by SPIN [17] o work under boh normal and failure modes X/$ see fron maer 2009 Elsevier B.V. All righs reserved. doi: /j.fuure

2 480 T. Angskun e al. / Fuure Generaion Compuer Sysems 26 (2010) (a) Node 4 dies. (b) Nodes 4 and 5 die. Fig. 1. SHN afer recovery High possibiliy of orphan (a) Excepional case. (b) Rearranged nodes. Fig. 2. Orphan prevenion. The srucure of his paper is as follows. The nex secion inroduces he self-healing nework and is recovery algorihm. Secion 3 presens he rouing algorihm along wih some experimenal resuls, while Secion 4 analyzes and discusses he reliabiliy of self-healing nework, followed by conclusions and fuure work in Secion Self-healing nework (SHN) 2.1. Overview The self-healing nework (SHN) is designed o suppor generic runime environmens of MPI implemenaions. Currenly, he inegraion of SHN in a faul-oleran implemenaion of message passing inerface called FT-MPI, as well as in he modular MPI implemenaion called Open MPI, is in progress. The nework is designed o suppor various operaions needed by scalable and faul-oleran MPI runime environmens. The examples of hose operaions and he deails on heir usage are described below. Disribued direcory service Direcory service is a sorage ha mainains informaion used during execuion of an MPI job such as conac informaion of each process, coordinaor of recovery algorihm in FT-MPI, ec. SHN enables us o use he nework as a disribued direcory service by mapping necessary informaion o he logical node ID. Scalable and faul-oleran informaion managemen (updae, query) can be done wih unicas messages of SFTP rouing (similar o resource discovery in he srucured peero-peer neworking). Sandard I/O redirecion Alhough he MPI sandard does no define how an inpu and an oupu can be reaed, mos of he MPI implemenaion redirecs he sandard oupu and he sandard error o he user erminal (if no run under he bach scheduling). This operaion can be done using he k-ary ree as a main roue o forward he sandard oupu/error and using he ring in case of failures. Monioring framework A monioring framework provides informaion such as processes, nodes, messages for ool and applicaion developmen. Examples of hose ools are parallel debuggers, runime faul deecors, runime verificaion and load balancers, ec. To build a scalable and faul-oleran monioring framework, all of he communicaion underneah he framework can use muliple ypes of message ransmissions (unicas, mulicas and broadcas) provided by SHN. In general, SHN provides he capabiliy o send unicas, mulicas and broadcas messages from any nodes while proecing he effecive message delivery agains node and process failures SHN recovery There are some siuaions where nodes do no die bu become unreachable due o nework bisecioning. This siuaion can be prevened by self-recovery. When a node deecs ha a neighbor disappears, i will send a unicas message o esablish he connecion wih he nex neighbor in he ring in he direcion of he dead node. This procedure will be coninued unil he connecion wih one of he nodes in he ring can be esablished or unil he node idenifies iself as he las remaining node in he ring. If wo nodes ry o esablish a connecion a he same ime, he connecion which is iniiaed by he higher ID will be dropped. Fig. 1(a) illusraes an example where logical node 4 dies. All neighbors of node 4 will begin o recover he logical opology by re-esablishing heir connecions in he appropriae direcion. If node 5 also dies, he same recovery procedure will occur as shown in Fig. 1(b). There is an excepion when he number of nodes on he las level (highes deph) of he ree is a mos equal o k (k is he fan-ou as shown in Fig. 2(a)). In his case, he conac informaion of he nodes on he las level should be propagaed o he grandparen in order o avoid he nework bisecion if he paren disappears. Alernaively, if here are a leas wo nodes in he las level, hese nodes can be rearranged o reduce he possibiliy of orphan nodes as shown in he Fig. 2(b). However, he iniializaion phase of his opology is more complex. The simples soluion o preven his problem is o change he fan-ou for a paricular number of nodes such ha he number of nodes in he las level of he ree is always more han ha of fan-ou as shown in Fig. 3. The fan-ous beween he lower bound and he upper bound, excep hose excepional cases in he Fig. 3, are safe from he high possibiliy of an orphan problem. The experimens have been conduced on an AMD Ahlon TM 64 Processor GHz machine wih 1 GB of main memory, running on Linux kernel Rouing algorihm in SHN The SHN rouing algorihm is based on he SFTP rouing algorihm [16]. The iniial sysem proocol, unicas message proocol,

3 T. Angskun e al. / Fuure Generaion Compuer Sysems 26 (2010) Fig. 3. Safe fan-ou. and broadcas from a specific roo proocol are he same as he SFTP proocol. The new mulicas and broadcas rouing algorihms from any nodes in he nework, which are exensions of he SFTP rouing algorihm, have been added. Boh can be used before (including some node failures) and afer recovery of he logical opology. The SHN rouing algorihms can be described as follows Mulicas messages in SHN The mulicas from any nodes in SHN is he capabiliy o send messages o several desinaions (1 o m, where m < n). Unlike he IP mulicas, mulicas group managemen (group creaion and erminaion) is no required. The mulicas group members are embedded in he message header. Mulicas messages in SFTP are delivered by a sender o he firs desinaion in he desinaion liss. Then, he firs desinaion will forward he message o he nex desinaion and so on. If an inermediae node is one of he nodes in he desinaion lis, i will remove iself from he lis. The order of nodes in he desinaion lis is a descending order sored by he number of hops from a sender o hose desinaions (i.e., he larges number of hops firs). This rouing algorihm works fine if he desinaion nodes are consecuive or if hey are locaed in he same area of he ree. The new mulicas rouing algorihm in SHN is an enhancemen of he SFTP mulicas rouing algorihm. The mulicas message can be spli a an inermediae node, if he shores pahs o hose desinaion nodes are no in he same direcion from he inermediae node poin of view. However, if here are more han one shores pah o a desinaion, he inermediae node will choose he nex hop ha can go along wih oher desinaions. Fig. 4(a) shows an example of node 2 sending a mulicas message o nodes 7, 8 and 9 wih he new rouing algorihm. Fig. 5. Mulicas resul (wih 1 failure node). In case of failure, if a node deecs ha he nex hop for he mulicas messages has died, i auomaically reroues he mulicas messages using an alernae nex hop as shown in Fig. 4(b). When a node receives a mulicas message, i will firs deermine he header and choose he nex hop for each mulicas desinaion according o he shores pah o hem. The node will recreae he header corresponding o he direcion of each nex hop. Messages ha conain he larges number of hops will be forwarded firs, in order o increase he nework hroughpu (by allowing a large number of messages simulaneously ino he nework). Fig. 5 confirms ha he new mulicas rouing algorihm is faser han he original algorihm used in he SFTP. The experimenal resuls were obained from an average number of seps for sending mulicas messages o wo desinaions wih a dead node (fan-ou = 2). The wo desinaion nodes (D) were obained ( from combinaions of all possible nodes (N), i.e., source node D and he dead node were randomly seleced Broadcas messages in SHN N D ), where a Broadcas from any node rouing proocol is an enhancemen of broadcas rouing in SFTP. In SFTP, he broadcas is done by sending messages o a roo of he ree, which will hen forward he messages o he res of he ree. Only he ree opology of SFTP is used o preven a broadcas sorm and duplicae messages. The ring is used only in he case of failure. The firs obvious improvemen of his rouing proocol is o allow a node beween a source and a roo of he ree o send messages o heir children afer hey send he messages o heir paren (called up down), as shown in Fig. 6(a) wih node 4 as he roo. The second improvemen is using a logical spanning ree from he source as shown in Fig. 6(b). When a (a) Normal circumsances. (b) Failure circumsances. Fig. 4. Mulicas message ransmission.

4 482 T. Angskun e al. / Fuure Generaion Compuer Sysems 26 (2010) (a) Up down. (b) Spanning ree from source. Fig. 6. Broadcas message ransmission. Fig. 8. Bahub curve. Fig. 7. Broadcas resul. node receives a broadcas message, i will calculae he nex hops using spanning rees from he source node. There are wo seps involved in he compuaion of he nex hop. The firs sep is o creae a spanning ree using a source node as he roo node of he ree. The spanning-ree creaion algorihm is based on a modified version of he breah firs search wih a graph coloring algorihm. The second sep is o calculae he nex hop. The nex hop is chosen from children of each node according o he spanning ree ha has he highes cos among is children. The cos is compued from he number of seps used o send a message o all nodes in he children s subrees. In case of failure, a broadcas message is encapsulaed ino a mulicas message, and hen he message is sen from he paren of he failure node o is children in he spanning ree. Fig. 7 indicaes ha he up down algorihm is marginally faser han he original SFTP, while he new spanning-ree broadcas rouing algorihm is significanly faser han he SFTP broadcas rouing algorihm due o increased parallelism. The experimenal resuls were obained from an average number of seps for sending a broadcas message from every node (fan-ou = 2). 4. Reliabiliy analysis of SHN The reliabiliy of SHN has been analyzed using he discreeeven simulaion echnique [18]. The reliabiliy is defined as he abiliy o mainain an operaion over a period of ime, i.e., he reliabiliy R() = Pr (he nework is operaional in [0, ]). SHN is operaional if i can successfully deliver messages from any source o he alive desinaion(s), even when some nodes in he rouing pah die. The cumulaive disribuion funcion (cdf), F() can be defined as F() = 0 f ()d where f () is he probabiliy densiy funcion (pdf) ha is associaed wih he lifeime of he nework. There are several characerisics ha are commonly used in reliabiliy analysis. These characerisics can be deermined from he pdf and cdf, e.g., reliabiliy funcion, hazard funcion and mean ime beween failures. The reliabiliy funcion (or survival funcion), R(), is he probabiliy ha SHN survives up o ime. I can be defined as R() = 1 F(). The simulaion assumes ha here is no failure a he iniial ime, i.e., = 0, R(0) = 1. The Hazard funcion, h() is he failure rae of he nework. The h() is defined by h() = f () R(). The failure rae in pracice has a bahub shape [19]. The hazard funcion of SHN is also assumed o change as he bahub curve, which consiss of hree phases: decreasing failure rae (burn in), consan failure rae and increasing failure rae (wearing ou) as shown in Fig. 8. The mean ime beween failure (MTBF) is defined o be he average (or expeced) lifeime of he nework. The MTBF is defined by MTBF = R()d. 0 The probabiliy densiy funcion of SHN is assumed o follow he Weibull disribuion [20]. This disribuion has he capabiliy o model he bahub curve. The pdf of he Weibull disribuion is given by [ ( ) ] β f () = βα β β 1 exp α where α is he scale parameer and β is he shape parameer. The associae funcions of he Weibull disribuion can be summarized

5 T. Angskun e al. / Fuure Generaion Compuer Sysems 26 (2010) (a) β effec. (b) α effec. Fig. 9. Effecs of shape (β) and scale (α). Table 1 Associae characerisic funcions of disribuions. Characerisics General Weibull CDF, F() 0 f ()d 1 e ( α ) β Reliabiliy funcion, R() 1 F() e ( α ) β f () Hazard funcion, h() βα β β 1 R() ( ) MTBF R()d αγ β in Table 1. The Γ denoes a gamma funcion where Γ (n) is defined as Γ (n) = e x x n 1 dx. 0 If n is an ineger hen Γ (n) = (n 1)!. In he mos general form, he 3-parameer form [21] of he Weibull includes an addiional waiing ime parameer µ (someimes called a shif or locaion parameer). The formulas for he 3-parameer Weibull can be easily obained from hese formulas by subsiuing occurrences of by ( µ). The Weibull lifeime disribuion assumes ha he hazard funcion is ime dependen. The hazard funcion is dependen on he value of β as shown in Fig. 9(a). If β < 1, he hazard funcion is he decreasing funcion (infan moraliy or burn in), i.e., he older i is, he less likely i fails (he firs phase of he bahub curve). If β = 1, he age has no effec. I is he second phase of he bahub curve. If β > 1, he hazard funcion is he increasing funcion (wearing ou), i.e., he older i is, he more likely i is o fail. I is he hird phase of he bahub curve. If 1 < β < 2, he hazard funcion is concave (increasing a a decreasing rae). On he oher hand, he hazard funcion is convex (increasing a an increasing rae), if β > 2. Fig. 9(b) shows he effecs of he characerisic life (α) on he failure rae, which affecs he spread (scale) of he disribuion. The simulaion assumes ha MTBF of he nework is hree years (26,280 h). Several β and is corresponding α parameers have been esed as shown in Table 2. If β equals o 1, he hazard funcion is ime independen, i.e., he nework is equally likely o fail a any momen during is lifeime, regardless of how old i is. The failure rae is known o be a consan ( 1 ). This is a special case α where Weibull becomes he exponenial disribuion [22]. Fig. 10 illusraes he effec of β and is corresponding α parameers (as shown in Table 2) in he Weibull lifeime disribuion o he percen average of success of mulicas operaions. I shows ha he new mulicas rouing used in SHN is more reliable han he original SFTP rouing for every value of he β parameer. Fig. 11 Table 2 Weibull parameers (MTBF=26,280). β α h() R() , , e ( , e ( , , 111, e ( , , e ( , , e ( Fig. 10. Weibull on mulicas (MTBF=26,280). ) ,140 ) 26,280 29, , , illusraes he effec of β and is corresponding α parameers (as shown in Table 2) in he Weibull lifeime disribuion o he percen average of success of broadcas operaions. I shows ha he spanning-ree (from source) broadcas rouing algorihm used in SHN is he mos reliable rouing algorihm when compared wih he up down and he original SFTP rouing for all values of he β parameer. 5. Conclusions and fuure work The self-healing nework (SHN) for parallel runime environmens was designed and developed o suppor runime environmens of MPI implemenaions. SHN is implemened on op of a scalable faul-oleran proocol (SFTP). Simulaed performance resuls indicae ha he new broadcas and mulicas rouing algorihms of SHN are faser and more reliable han he original SFTP rouing algorihms. ) 1.5 ) 2.0 ) 2.5

6 484 T. Angskun e al. / Fuure Generaion Compuer Sysems 26 (2010) (a) SFTP. (b) Up down. (c) Spanning ree. Fig. 11. Weibull on broadcas (MTBF=26,280). There are several improvemens ha we plan for he near fuure. Making he proocol aware of he underlying nework opology (in boh he LAN and WAN environmens) will grealy improve he overall performance of boh he broadcas and mulicas message disribuion. This is equivalen o adding a funcion cos on each possible pah and inegraing his funcion cos wih he compuaion of he shores pah. In he longer erm, we hope ha SHN will become he basic message disribuion of he runime environmen wihin he FT-MPI and Open MPI runime sysems. References [1] J.J. Dongarra, H. Meuer, E. Srohmaier, TOP500 supercompuer sies, Supercompuer 13 (1) (1997) [2] MPI Forum, MPI: A message-passing inerface sandard, Tech. Rep., [3] W. Gropp, E. Lusk, N. Doss, A. Skjellum, A high-performance, porable implemenaion of MPI message passing inerface sandard, Parallel Compuing 22 (6) (1996) [4] R. Buler, W. Gropp, E.L. Lusk, A scalable process-managemen environmen for parallel program, in: Recen Advances in PVM and MPI, in: LNCS, vol. 1908, Springer, 2000, pp [5] G.E. Fagg, J.J. Dongarra, HARNESS faul oleran MPI design, usage and performance issues, Fuure Generaion Compuer Sysems 18 (8) (2002) [6] M. Beck, J.J. Dongarra, G.E. Fagg, G.A. Geis, P. Gray, J. Kohl, M. Migliardi, K. Moore, T. Moore, P. Papadopoulous, S.L. Sco, V. Sunderam, HARNESS: A nex generaion disribued virual machine, Fuure Generaion Compuer Sysems 15 (5 6) (1999) [7] R.H. Casain, T.S. Woodall, D.J. Daniel, J.M. Squyres, B. Barre, G.E. Fagg, The open run-ime environmen (openre): A ransparen muli-cluser environmen for high-performance compuing, in: Recen Advances in PVM and MPI, in: LNCS, vol. 3666, Springer, 2005, pp [8] E. Gabriel, G.E. Fagg, G. Bosilca, T. Angskun, J.J. Dongarra, J.M. Squyres, V. Sahay, P. Kambadur, B. Barre, A. Lumsdaine, R.H. Casain, D.J. Daniel, R.L. Graham, T.S. Woodall, Open MPI: Goals, concep, and design of a nex generaion MPI implemenaion, in: Recen Advances in PVM and MPI, in: LNCS, vol. 3241, Springer, 2004, pp [9] G. Burns, R. Daoud, J. Vaigl, LAM: An open cluser environmen for MPI, in: Proceedings Supercompuing Symposium, 1994, pp [10] S. Ranasamy, P. Francis, M. Handley, R. Karp, S. Shenker, A scalable conen addressable nework, Tech. Rep. TR , Berkeley, CA, [11] I. Soica, R. Morris, D. Karger, F. Kaashoek, H. Balakrishnan, Chord: A scalable peer-o-peer lookup service for inerne applicaions, in: Proceedings of he 2001 ACM SIGCOMM Conference, 2001, pp [12] A. Rowsron, P. Druschel, Pasry: Scalable, decenralized objec locaion, and rouing for large-scale peer-o-peer sysems, Lecure Noes in Compuer Science 2218 (2001) [13] B.Y. Zhao, J.D. Kubiaowicz, A.D. Joseph, Tapesry: An infrasrucure for fauloleran wide-area locaion and rouing, Tech. Rep. UCB/CSD , UC Berkeley, April [14] I. Gupa, R. van Renesse, K. Birman, Scalable faul-oleran aggregaion in large process groups, in: Proceedings of The Inernaional Conference on Dependable Sysems and Neworks, DSN, 2001, pp [15] R.V. Renesse, Y. Minsky, M. Hayden, A gossip-syle failure deecion service, Tech. Rep. TR , 28, [16] T. Angskun, G.E. Fagg, G. Bosilca, J.P. vac Grbovic, J. Dongarra, Scalable faul oleran proocol for parallel runime environmens, in: Recen Advances in PVM and MPI, in: LNCS, vol. 4192, Springer, 2006, pp [17] G.J. Holzmann, The model checker SPIN, IEEE Transacions on Sofware Engineering 23 (5) (1997) [18] L.M. Leemis, S.K. Park, Discree-Even Simulaion: A Firs Course, Prenice Hall, [19] M. Xie, C.D. Lai, Reliabiliy analysis using an addiive Weibull model wih bahub-shaped failure rae funcion, Reliabiliy Engineering and Sysem Safey 52 (1) (1995) [20] A. Plai, The Weibull disribuion wih ables, Indusrial Qualiy Conrol 19 (5) (1962) [21] G.W. Cran, Momen esimaors for he 3-parameer Weibull disribuion, IEEE Transacions on Reliabiliy 37 (4) (1988) [22] N. Balakrishnan, A.P. Basu, The Exponenial Disribuion: Theory, Mehods and Applicaions, Gordon and Breach Publishers, Thara Angskun received his Bachelor and Maser degree in Compuer Engineering from Kasesar Universiy, Thailand. Currenly, he is a Ph.D. suden and Graduae Research Assisan a he Innovaive Compuing Laboraory, Deparmen of Compuer Science, Universiy of Tennessee, Knoxville. His major research ineress are in parallel and disribued environmens, message passing, high performance compuing, compuer neworking, cluser and grid compuing. He is a developer of several projecs including OpenSCE, KSIX, ACI, CAMETA, Harness/FT-MPI and Open MPI. Graham Fagg received his B.Sc. in Compuer Science and Cyberneics from he Universiy of Reading (UK) in 1991 and a Ph.D. in Compuer Science in Currenly he is a Research Associae Professor a he Universiy of Tennessee. His curren research ineress include disribued scheduling, resource managemen, performance predicion, profiling, benchmarking, cluser managemen ools, parallel and disribued IO and high speed neworking. He is currenly involved in he developmen of a number of disribued, meacompuing and GRID middleware sysems including SNIPE/2, HARNESS, faul-oleran MPI (FT-MPI) and Open MPI. He is a member of he IEEE. George Bosilca is a Senior Research Associae a he Innovaive Compuing Laboraory (ICL). He is also an Adjunc Professor in Compuer Science Deparmen a he Universiy of Tennessee, Knoxville. He received a Ph.D. degree in parallel archiecures from he Universiy de Paris XI. He was he main developer of he channel memory subsysem for MPICH-V. Dr. Bosilca is currenly working on he Open MPI projec.

7 T. Angskun e al. / Fuure Generaion Compuer Sysems 26 (2010) Jelena Pjesivac-Grbovic is a Graduae Research Assisan a he Innovaive Compuing Laboraory a he Universiy of Tennessee a Knoxville, working oward Ph.D. degree in Compuer Science. She received M.S. in Compuer Science from he Universiy of Tennessee a Knoxville, and B.S. degrees in Compuer Science and Physics from Ramapo College of New Jersey. Her research ineress are parallel communicaion libraries and compuer archiecures, scienific and grid compuing, and modeling of biophysical sysems. She is a developer on Harness/FT-MPI projec. Jack Dongarra holds an appoinmen as Universiy Disinguished Professor of Compuer Science in he Compuer Science Deparmen a he Universiy of Tennessee and holds he ile of Disinguished Research Saff in he Compuer Science and Mahemaics Division a Oak Ridge Naional Laboraory (ORNL), and is an Adjunc Professor in he Compuer Science Deparmen a Rice Universiy. He specializes in numerical algorihms in linear algebra, parallel compuing, use of advanced-compuer archiecures, programming mehodology, and ools for parallel compuers. He is a Fellow of he AAAS, ACM, and he IEEE and a member of he Naional Academy of Engineering.

SELF-HEALING NETWORK FOR SCALABLE FAULT TOLERANT RUNTIME ENVIRONMENTS

SELF-HEALING NETWORK FOR SCALABLE FAULT TOLERANT RUNTIME ENVIRONMENTS SELF-HEALING NETWORK FOR SCALABLE FAULT TOLERANT RUNTIME ENVIRONMENTS Thara Angskun, Graham Fagg, George Bosilca, Jelena Pješivac Grbović, and Jack Dongarra,2,3 University of Tennessee, 2 Oak Ridge National

More information

Implementing Ray Casting in Tetrahedral Meshes with Programmable Graphics Hardware (Technical Report)

Implementing Ray Casting in Tetrahedral Meshes with Programmable Graphics Hardware (Technical Report) Implemening Ray Casing in Terahedral Meshes wih Programmable Graphics Hardware (Technical Repor) Marin Kraus, Thomas Erl March 28, 2002 1 Inroducion Alhough cell-projecion, e.g., [3, 2], and resampling,

More information

Simple Network Management Based on PHP and SNMP

Simple Network Management Based on PHP and SNMP Simple Nework Managemen Based on PHP and SNMP Krasimir Trichkov, Elisavea Trichkova bsrac: This paper aims o presen simple mehod for nework managemen based on SNMP - managemen of Cisco rouer. The paper

More information

Scalable Fault Tolerant Protocol for Parallel Runtime Environments

Scalable Fault Tolerant Protocol for Parallel Runtime Environments Scalable Fault Tolerant Protocol for Parallel Runtime Environments Thara Angskun 1, Graham Fagg 1, George Bosilca 1, Jelena Pješivac Grbović 1, and Jack Dongarra 2 1 Department of Computer Science, The

More information

COSC 3213: Computer Networks I Chapter 6 Handout # 7

COSC 3213: Computer Networks I Chapter 6 Handout # 7 COSC 3213: Compuer Neworks I Chaper 6 Handou # 7 Insrucor: Dr. Marvin Mandelbaum Deparmen of Compuer Science York Universiy F05 Secion A Medium Access Conrol (MAC) Topics: 1. Muliple Access Communicaions:

More information

A time-space consistency solution for hardware-in-the-loop simulation system

A time-space consistency solution for hardware-in-the-loop simulation system Inernaional Conference on Advanced Elecronic Science and Technology (AEST 206) A ime-space consisency soluion for hardware-in-he-loop simulaion sysem Zexin Jiang a Elecric Power Research Insiue of Guangdong

More information

Test - Accredited Configuration Engineer (ACE) Exam - PAN-OS 6.0 Version

Test - Accredited Configuration Engineer (ACE) Exam - PAN-OS 6.0 Version Tes - Accredied Configuraion Engineer (ACE) Exam - PAN-OS 6.0 Version ACE Exam Quesion 1 of 50. Which of he following saemens is NOT abou Palo Alo Neworks firewalls? Sysem defauls may be resored by performing

More information

A Tool for Multi-Hour ATM Network Design considering Mixed Peer-to-Peer and Client-Server based Services

A Tool for Multi-Hour ATM Network Design considering Mixed Peer-to-Peer and Client-Server based Services A Tool for Muli-Hour ATM Nework Design considering Mied Peer-o-Peer and Clien-Server based Services Conac Auhor Name: Luis Cardoso Company / Organizaion: Porugal Telecom Inovação Complee Mailing Address:

More information

Analysis of Various Types of Bugs in the Object Oriented Java Script Language Coding

Analysis of Various Types of Bugs in the Object Oriented Java Script Language Coding Indian Journal of Science and Technology, Vol 8(21), DOI: 10.17485/ijs/2015/v8i21/69958, Sepember 2015 ISSN (Prin) : 0974-6846 ISSN (Online) : 0974-5645 Analysis of Various Types of Bugs in he Objec Oriened

More information

User Adjustable Process Scheduling Mechanism for a Multiprocessor Embedded System

User Adjustable Process Scheduling Mechanism for a Multiprocessor Embedded System Proceedings of he 6h WSEAS Inernaional Conference on Applied Compuer Science, Tenerife, Canary Islands, Spain, December 16-18, 2006 346 User Adjusable Process Scheduling Mechanism for a Muliprocessor Embedded

More information

The Impact of Product Development on the Lifecycle of Defects

The Impact of Product Development on the Lifecycle of Defects The Impac of Produc Developmen on he Lifecycle of Rudolf Ramler Sofware Compeence Cener Hagenberg Sofware Park 21 A-4232 Hagenberg, Ausria +43 7236 3343 872 rudolf.ramler@scch.a ABSTRACT This paper invesigaes

More information

Restorable Dynamic Quality of Service Routing

Restorable Dynamic Quality of Service Routing QOS ROUTING Resorable Dynamic Qualiy of Service Rouing Murali Kodialam and T. V. Lakshman, Lucen Technologies ABSTRACT The focus of qualiy-of-service rouing has been on he rouing of a single pah saisfying

More information

Selective Offloading in Mobile Edge Computing for the Green Internet of Things

Selective Offloading in Mobile Edge Computing for the Green Internet of Things EDGE COMPUTING FOR THE INTERNET OF THINGS Selecive Offloading in Mobile Edge Compuing for he Green Inerne of Things Xinchen Lyu, Hui Tian, Li Jiang, Alexey Vinel, Sabia Maharjan, Sein Gjessing, and Yan

More information

MORPHOLOGICAL SEGMENTATION OF IMAGE SEQUENCES

MORPHOLOGICAL SEGMENTATION OF IMAGE SEQUENCES MORPHOLOGICAL SEGMENTATION OF IMAGE SEQUENCES B. MARCOTEGUI and F. MEYER Ecole des Mines de Paris, Cenre de Morphologie Mahémaique, 35, rue Sain-Honoré, F 77305 Fonainebleau Cedex, France Absrac. In image

More information

Web System for the Remote Control and Execution of an IEC Application

Web System for the Remote Control and Execution of an IEC Application Web Sysem for he Remoe Conrol and Execuion of an IEC 61499 Applicaion Oana ROHAT, Dan POPESCU Faculy of Auomaion and Compuer Science, Poliehnica Universiy, Splaiul Independenței 313, Bucureși, 060042,

More information

Nonparametric CUSUM Charts for Process Variability

Nonparametric CUSUM Charts for Process Variability Journal of Academia and Indusrial Research (JAIR) Volume 3, Issue June 4 53 REEARCH ARTICLE IN: 78-53 Nonparameric CUUM Chars for Process Variabiliy D.M. Zombade and V.B. Ghue * Dep. of aisics, Walchand

More information

Chapter 8 LOCATION SERVICES

Chapter 8 LOCATION SERVICES Disribued Compuing Group Chaper 8 LOCATION SERVICES Mobile Compuing Winer 2005 / 2006 Overview Mobile IP Moivaion Daa ransfer Encapsulaion Locaion Services & Rouing Classificaion of locaion services Home

More information

MOBILE COMPUTING 3/18/18. Wi-Fi IEEE. CSE 40814/60814 Spring 2018

MOBILE COMPUTING 3/18/18. Wi-Fi IEEE. CSE 40814/60814 Spring 2018 MOBILE COMPUTING CSE 40814/60814 Spring 2018 Wi-Fi Wi-Fi: name is NOT an abbreviaion play on Hi-Fi (high fideliy) Wireless Local Area Nework (WLAN) echnology WLAN and Wi-Fi ofen used synonymous Typically

More information

MOBILE COMPUTING. Wi-Fi 9/20/15. CSE 40814/60814 Fall Wi-Fi:

MOBILE COMPUTING. Wi-Fi 9/20/15. CSE 40814/60814 Fall Wi-Fi: MOBILE COMPUTING CSE 40814/60814 Fall 2015 Wi-Fi Wi-Fi: name is NOT an abbreviaion play on Hi-Fi (high fideliy) Wireless Local Area Nework (WLAN) echnology WLAN and Wi-Fi ofen used synonymous Typically

More information

Less Pessimistic Worst-Case Delay Analysis for Packet-Switched Networks

Less Pessimistic Worst-Case Delay Analysis for Packet-Switched Networks Less Pessimisic Wors-Case Delay Analysis for Packe-Swiched Neworks Maias Wecksén Cenre for Research on Embedded Sysems P O Box 823 SE-31 18 Halmsad maias.wecksen@hh.se Magnus Jonsson Cenre for Research

More information

Assignment 2. Due Monday Feb. 12, 10:00pm.

Assignment 2. Due Monday Feb. 12, 10:00pm. Faculy of rs and Science Universiy of Torono CSC 358 - Inroducion o Compuer Neworks, Winer 218, LEC11 ssignmen 2 Due Monday Feb. 12, 1:pm. 1 Quesion 1 (2 Poins): Go-ack n RQ In his quesion, we review how

More information

Scheduling. Scheduling. EDA421/DIT171 - Parallel and Distributed Real-Time Systems, Chalmers/GU, 2011/2012 Lecture #4 Updated March 16, 2012

Scheduling. Scheduling. EDA421/DIT171 - Parallel and Distributed Real-Time Systems, Chalmers/GU, 2011/2012 Lecture #4 Updated March 16, 2012 EDA421/DIT171 - Parallel and Disribued Real-Time Sysems, Chalmers/GU, 2011/2012 Lecure #4 Updaed March 16, 2012 Aemps o mee applicaion consrains should be done in a proacive way hrough scheduling. Schedule

More information

Shortest Path Algorithms. Lecture I: Shortest Path Algorithms. Example. Graphs and Matrices. Setting: Dr Kieran T. Herley.

Shortest Path Algorithms. Lecture I: Shortest Path Algorithms. Example. Graphs and Matrices. Setting: Dr Kieran T. Herley. Shores Pah Algorihms Background Seing: Lecure I: Shores Pah Algorihms Dr Kieran T. Herle Deparmen of Compuer Science Universi College Cork Ocober 201 direced graph, real edge weighs Le he lengh of a pah

More information

FIELD PROGRAMMABLE GATE ARRAY (FPGA) AS A NEW APPROACH TO IMPLEMENT THE CHAOTIC GENERATORS

FIELD PROGRAMMABLE GATE ARRAY (FPGA) AS A NEW APPROACH TO IMPLEMENT THE CHAOTIC GENERATORS FIELD PROGRAMMABLE GATE ARRAY (FPGA) AS A NEW APPROACH TO IMPLEMENT THE CHAOTIC GENERATORS Mohammed A. Aseeri and M. I. Sobhy Deparmen of Elecronics, The Universiy of Ken a Canerbury Canerbury, Ken, CT2

More information

An Adaptive Spatial Depth Filter for 3D Rendering IP

An Adaptive Spatial Depth Filter for 3D Rendering IP JOURNAL OF SEMICONDUCTOR TECHNOLOGY AND SCIENCE, VOL.3, NO. 4, DECEMBER, 23 175 An Adapive Spaial Deph Filer for 3D Rendering IP Chang-Hyo Yu and Lee-Sup Kim Absrac In his paper, we presen a new mehod

More information

Communication Networks

Communication Networks Communicaion Neworks Chaper 10 Wireless Local Area Neworks According o IEEE 802.11 Communicaion Neworks: 10. IEEE 802.11 651 10. WLANs According o IEEE 802.11 Overview Organizaion of a WLAN according o

More information

MIC2569. Features. General Description. Applications. Typical Application. CableCARD Power Switch

MIC2569. Features. General Description. Applications. Typical Application. CableCARD Power Switch CableCARD Power Swich General Descripion is designed o supply power o OpenCable sysems and CableCARD hoss. These CableCARDs are also known as Poin of Disribuion (POD) cards. suppors boh Single and Muliple

More information

Let s get physical - EDA Tools for Mobility

Let s get physical - EDA Tools for Mobility Le s ge physical - EDA Tools for Mobiliy Aging and Reliabiliy Communicaion Mobile and Green Mobiliy - Smar and Safe Frank Oppenheimer OFFIS Insiue for Informaion Technology OFFIS a a glance Applicaion-oriened

More information

This is the published version of a paper presented at The 2013 IEEE International Conference on Internet of Things, Beijing, China, August 2013.

This is the published version of a paper presented at The 2013 IEEE International Conference on Internet of Things, Beijing, China, August 2013. hp://www.diva-poral.org This is he published version of a paper presened a The 2013 IEEE Inernaional Conference on Inerne of Things, Beijing, China, 20-23 Augus 2013. Ciaion for he original published paper:

More information

Voltair Version 2.5 Release Notes (January, 2018)

Voltair Version 2.5 Release Notes (January, 2018) Volair Version 2.5 Release Noes (January, 2018) Inroducion 25-Seven s new Firmware Updae 2.5 for he Volair processor is par of our coninuing effors o improve Volair wih new feaures and capabiliies. For

More information

Low-Cost WLAN based. Dr. Christian Hoene. Computer Science Department, University of Tübingen, Germany

Low-Cost WLAN based. Dr. Christian Hoene. Computer Science Department, University of Tübingen, Germany Low-Cos WLAN based Time-of-fligh fligh Trilaeraion Precision Indoor Personnel Locaion and Tracking for Emergency Responders Third Annual Technology Workshop, Augus 5, 2008 Worceser Polyechnic Insiue, Worceser,

More information

Scalable Fault Tolerant Protocol for Parallel Runtime Environments

Scalable Fault Tolerant Protocol for Parallel Runtime Environments Scalable Fault Tolerant Protocol for Parallel Runtime Environments Thara Angskun, Graham E. Fagg, George Bosilca, Jelena Pješivac Grbović, and Jack J. Dongarra Dept. of Computer Science, 1122 Volunteer

More information

Network management and QoS provisioning - QoS in Frame Relay. . packet switching with virtual circuit service (virtual circuits are bidirectional);

Network management and QoS provisioning - QoS in Frame Relay. . packet switching with virtual circuit service (virtual circuits are bidirectional); QoS in Frame Relay Frame relay characerisics are:. packe swiching wih virual circui service (virual circuis are bidirecional);. labels are called DLCI (Daa Link Connecion Idenifier);. for connecion is

More information

Parallel and Distributed Systems for Constructive Neural Network Learning*

Parallel and Distributed Systems for Constructive Neural Network Learning* Parallel and Disribued Sysems for Consrucive Neural Nework Learning* J. Flecher Z. Obradovi School of Elecrical Engineering and Compuer Science Washingon Sae Universiy Pullman WA 99164-2752 Absrac A consrucive

More information

PART 1 REFERENCE INFORMATION CONTROL DATA 6400 SYSTEMS CENTRAL PROCESSOR MONITOR

PART 1 REFERENCE INFORMATION CONTROL DATA 6400 SYSTEMS CENTRAL PROCESSOR MONITOR . ~ PART 1 c 0 \,).,,.,, REFERENCE NFORMATON CONTROL DATA 6400 SYSTEMS CENTRAL PROCESSOR MONTOR n CONTROL DATA 6400 Compuer Sysems, sysem funcions are normally handled by he Monior locaed in a Peripheral

More information

Coded Caching with Multiple File Requests

Coded Caching with Multiple File Requests Coded Caching wih Muliple File Requess Yi-Peng Wei Sennur Ulukus Deparmen of Elecrical and Compuer Engineering Universiy of Maryland College Park, MD 20742 ypwei@umd.edu ulukus@umd.edu Absrac We sudy a

More information

STEREO PLANE MATCHING TECHNIQUE

STEREO PLANE MATCHING TECHNIQUE STEREO PLANE MATCHING TECHNIQUE Commission III KEY WORDS: Sereo Maching, Surface Modeling, Projecive Transformaion, Homography ABSTRACT: This paper presens a new ype of sereo maching algorihm called Sereo

More information

Evaluation and Improvement of Multicast Service in b

Evaluation and Improvement of Multicast Service in b Evaluaion and Improvemen of Mulicas Service in 802.11b Chrisian Bravo 1 and Agusín González 2 1 Universidad Federico Sana María, Deparmen of Elecronics. Valparaíso, Chile chbravo@elo.ufsm.cl 2 Universidad

More information

Distributed Task Negotiation in Modular Robots

Distributed Task Negotiation in Modular Robots Disribued Task Negoiaion in Modular Robos Behnam Salemi, eer Will, and Wei-Min Shen USC Informaion Sciences Insiue and Compuer Science Deparmen Marina del Rey, USA, {salemi, will, shen}@isi.edu Inroducion

More information

Improving the Efficiency of Dynamic Service Provisioning in Transport Networks with Scheduled Services

Improving the Efficiency of Dynamic Service Provisioning in Transport Networks with Scheduled Services Improving he Efficiency of Dynamic Service Provisioning in Transpor Neworks wih Scheduled Services Ralf Hülsermann, Monika Jäger and Andreas Gladisch Technologiezenrum, T-Sysems, Goslarer Ufer 35, D-1585

More information

/85/ $ IEEE

/85/ $ IEEE 174 Saged Circui Swiching MAURICIO ARANGO, HUSSEIN BADR, AND DAVID GELERNTER Absrac -Saged circui swiching (S) is a message-swiching echnique ha combines a new proocol wih new communicaion hardware. Proocol

More information

An efficient approach to improve throughput for TCP vegas in ad hoc network

An efficient approach to improve throughput for TCP vegas in ad hoc network Inernaional Research Journal of Engineering and Technology (IRJET) e-issn: 395-0056 Volume: 0 Issue: 03 June-05 www.irje.ne p-issn: 395-007 An efficien approach o improve hroughpu for TCP vegas in ad hoc

More information

Performance Evaluation of Implementing Calls Prioritization with Different Queuing Disciplines in Mobile Wireless Networks

Performance Evaluation of Implementing Calls Prioritization with Different Queuing Disciplines in Mobile Wireless Networks Journal of Compuer Science 2 (5): 466-472, 2006 ISSN 1549-3636 2006 Science Publicaions Performance Evaluaion of Implemening Calls Prioriizaion wih Differen Queuing Disciplines in Mobile Wireless Neworks

More information

Learning in Games via Opponent Strategy Estimation and Policy Search

Learning in Games via Opponent Strategy Estimation and Policy Search Learning in Games via Opponen Sraegy Esimaion and Policy Search Yavar Naddaf Deparmen of Compuer Science Universiy of Briish Columbia Vancouver, BC yavar@naddaf.name Nando de Freias (Supervisor) Deparmen

More information

4 Error Control. 4.1 Issues with Reliable Protocols

4 Error Control. 4.1 Issues with Reliable Protocols 4 Error Conrol Jus abou all communicaion sysems aemp o ensure ha he daa ges o he oher end of he link wihou errors. Since i s impossible o build an error-free physical layer (alhough some shor links can

More information

Chapter 4 Sequential Instructions

Chapter 4 Sequential Instructions Chaper 4 Sequenial Insrucions The sequenial insrucions of FBs-PLC shown in his chaper are also lised in secion 3.. Please refer o Chaper, "PLC Ladder diagram and he Coding rules of Mnemonic insrucion",

More information

Sam knows that his MP3 player has 40% of its battery life left and that the battery charges by an additional 12 percentage points every 15 minutes.

Sam knows that his MP3 player has 40% of its battery life left and that the battery charges by an additional 12 percentage points every 15 minutes. 8.F Baery Charging Task Sam wans o ake his MP3 player and his video game player on a car rip. An hour before hey plan o leave, he realized ha he forgo o charge he baeries las nigh. A ha poin, he plugged

More information

Y. Tsiatouhas. VLSI Systems and Computer Architecture Lab

Y. Tsiatouhas. VLSI Systems and Computer Architecture Lab CMOS INEGRAED CIRCUI DESIGN ECHNIQUES Universiy of Ioannina Clocking Schemes Dep. of Compuer Science and Engineering Y. siaouhas CMOS Inegraed Circui Design echniques Overview 1. Jier Skew hroughpu Laency

More information

Michiel Helder and Marielle C.T.A Geurts. Hoofdkantoor PTT Post / Dutch Postal Services Headquarters

Michiel Helder and Marielle C.T.A Geurts. Hoofdkantoor PTT Post / Dutch Postal Services Headquarters SHORT TERM PREDICTIONS A MONITORING SYSTEM by Michiel Helder and Marielle C.T.A Geurs Hoofdkanoor PTT Pos / Duch Posal Services Headquarers Keywords macro ime series shor erm predicions ARIMA-models faciliy

More information

Lecture 18: Mix net Voting Systems

Lecture 18: Mix net Voting Systems 6.897: Advanced Topics in Crypography Apr 9, 2004 Lecure 18: Mix ne Voing Sysems Scribed by: Yael Tauman Kalai 1 Inroducion In he previous lecure, we defined he noion of an elecronic voing sysem, and specified

More information

A Matching Algorithm for Content-Based Image Retrieval

A Matching Algorithm for Content-Based Image Retrieval A Maching Algorihm for Conen-Based Image Rerieval Sue J. Cho Deparmen of Compuer Science Seoul Naional Universiy Seoul, Korea Absrac Conen-based image rerieval sysem rerieves an image from a daabase using

More information

Time-Shifted Streaming in a Tree-Based Peer-to-Peer System Jeonghun Noh ASSIA inc., Redwood City, CA, USA

Time-Shifted Streaming in a Tree-Based Peer-to-Peer System Jeonghun Noh ASSIA inc., Redwood City, CA, USA 22 JOURNAL OF COMMUNICATIONS, VOL. 7, NO. 3, MARCH 212 -Shifed Sreaming in a Tree-Based Peer-o-Peer Sysem Jeonghun Noh ASSIA inc., Redwood Ciy, CA, USA Email: jnoh@assia-inc.com Bernd Girod Informaion

More information

PERFORMANCE OF TCP CONGESTION CONTROL IN UAV NETWORKS OF VARIOUS RADIO PROPAGATION MODELS

PERFORMANCE OF TCP CONGESTION CONTROL IN UAV NETWORKS OF VARIOUS RADIO PROPAGATION MODELS PERFORMANCE OF TCP CONGESTION CONTROL IN UAV NETWORKS OF VARIOUS RADIO PROPAGATION MODELS Jun Peng Elecrical Engineering Deparmen, Universiy of Texas Rio Grande Valley, Edinburg, Texas, USA ABSTRACT Unmanned

More information

Optimal Crane Scheduling

Optimal Crane Scheduling Opimal Crane Scheduling Samid Hoda, John Hooker Laife Genc Kaya, Ben Peerson Carnegie Mellon Universiy Iiro Harjunkoski ABB Corporae Research EWO - 13 November 2007 1/16 Problem Track-mouned cranes move

More information

Architectural Design and Implementation of 2008 ZigBee Specification on TKNlS.4 Layer

Architectural Design and Implementation of 2008 ZigBee Specification on TKNlS.4 Layer rchiecural Design and mplemenaion of 2008 ZigBee Specificaion on TKNlS.4 Layer nh-vu Dinh-Due Universiy of nfonnaion Technology Ho Chi Minh ciy, Vie Nam anhvu@ui.edu.vn Ngoe-nLam HCMC Universiy of Technology

More information

Provided by the author(s) and NUI Galway in accordance with publisher policies. Please cite the published version when available.

Provided by the author(s) and NUI Galway in accordance with publisher policies. Please cite the published version when available. Provided by he auhor(s) and NUI Galway in accordance wih publisher policies. Please cie he published version when available. Tile Conneciviy soluions o link a blueooh camera o he inerne Auhor(s) Ionas,

More information

Vulnerability Evaluation of Multimedia Subsystem Based on Complex Network

Vulnerability Evaluation of Multimedia Subsystem Based on Complex Network JOURAL OF MULTIMDIA, VOL. 8, O. 4, AUGUST 23 439 Vulnerabiliy valuaion of Mulimedia Subsysem Based on Complex ewor Xiaoling Tang Insiue of Higher ducaion Research, Jilin Business and Technology College,

More information

M(t)/M/1 Queueing System with Sinusoidal Arrival Rate

M(t)/M/1 Queueing System with Sinusoidal Arrival Rate 20 TUTA/IOE/PCU Journal of he Insiue of Engineering, 205, (): 20-27 TUTA/IOE/PCU Prined in Nepal M()/M/ Queueing Sysem wih Sinusoidal Arrival Rae A.P. Pan, R.P. Ghimire 2 Deparmen of Mahemaics, Tri-Chandra

More information

Dynamic Route Planning and Obstacle Avoidance Model for Unmanned Aerial Vehicles

Dynamic Route Planning and Obstacle Avoidance Model for Unmanned Aerial Vehicles Volume 116 No. 24 2017, 315-329 ISSN: 1311-8080 (prined version); ISSN: 1314-3395 (on-line version) url: hp://www.ijpam.eu ijpam.eu Dynamic Roue Planning and Obsacle Avoidance Model for Unmanned Aerial

More information

An Improved Square-Root Nyquist Shaping Filter

An Improved Square-Root Nyquist Shaping Filter An Improved Square-Roo Nyquis Shaping Filer fred harris San Diego Sae Universiy fred.harris@sdsu.edu Sridhar Seshagiri San Diego Sae Universiy Seshigar.@engineering.sdsu.edu Chris Dick Xilinx Corp. chris.dick@xilinx.com

More information

SEINA: A Stealthy and Effective Internal Attack in Hadoop Systems

SEINA: A Stealthy and Effective Internal Attack in Hadoop Systems SEINA: A Sealhy and Effecive Inernal Aack in Hadoop Sysems Jiayin Wang, Teng Wang, Zhengyu Yang, Ying ao, Ningfang i, and Bo Sheng Deparmen of Compuer Science, Universiy of assachuses Boson, 1 orrissey

More information

COMP26120: Algorithms and Imperative Programming

COMP26120: Algorithms and Imperative Programming COMP26120 ecure C3 1/48 COMP26120: Algorihms and Imperaive Programming ecure C3: C - Recursive Daa Srucures Pee Jinks School of Compuer Science, Universiy of Mancheser Auumn 2011 COMP26120 ecure C3 2/48

More information

Image Based Computer-Aided Manufacturing Technology

Image Based Computer-Aided Manufacturing Technology Sensors & Transducers 03 by IFSA hp://www.sensorsporal.com Image Based Compuer-Aided Manufacuring Technology Zhanqi HU Xiaoqin ZHANG Jinze LI Wei LI College of Mechanical Engineering Yanshan Universiy

More information

EP2200 Queueing theory and teletraffic systems

EP2200 Queueing theory and teletraffic systems EP2200 Queueing heory and eleraffic sysems Vikoria Fodor Laboraory of Communicaion Neworks School of Elecrical Engineering Lecure 1 If you wan o model neworks Or a comple daa flow A queue's he key o help

More information

source managemen, naming, proecion, and service provisions. This paper concenraes on he basic processor scheduling aspecs of resource managemen. 2 The

source managemen, naming, proecion, and service provisions. This paper concenraes on he basic processor scheduling aspecs of resource managemen. 2 The Virual Compuers A New Paradigm for Disribued Operaing Sysems Banu Ozden y Aaron J. Goldberg Avi Silberschaz z 600 Mounain Ave. AT&T Bell Laboraories Murray Hill, NJ 07974 Absrac The virual compuers (VC)

More information

Delayed reservation decision in optical burst switching networks with optical buffers. Title. Li, GM; Li, VOK; Li, CY; Wai, PKA

Delayed reservation decision in optical burst switching networks with optical buffers. Title. Li, GM; Li, VOK; Li, CY; Wai, PKA Tile Delayed reservaion decision in opical burs swiching neworks wih opical buffers Auhor(s) Li, GM; Li, VOK; Li, CY; Wai, PKA Ciaion The 3rd nernaional Conference on Communicaions and Neworking in China

More information

MoBAN: A Configurable Mobility Model for Wireless Body Area Networks

MoBAN: A Configurable Mobility Model for Wireless Body Area Networks MoBAN: A Configurable Mobiliy Model for Wireless Body Area Neworks Majid Nabi 1, Marc Geilen 1, Twan Basen 1,2 1 Deparmen of Elecrical Engineering, Eindhoven Universiy of Technology, he Neherlands 2 Embedded

More information

SREM: A Novel Multicast Routing Algorithm - Comprehensive Cost Analysis

SREM: A Novel Multicast Routing Algorithm - Comprehensive Cost Analysis REM: A Novel Mulicas Rouing Algorihm - Comprehensive Cos Analysis Yewen Cao and Khalid Al-Begain chool of Compuing, Universiy of Glamorgan, CF37 DL, Wales, U.K E-mail:{ycao,kbegain}@glam.ac.uk Absrac Ever-increasing

More information

Software Architecture of the Light Weight Kernel, Catamount

Software Architecture of the Light Weight Kernel, Catamount Sofware Archiecure of he Ligh Weigh Kernel Caamoun Suzanne M. Kelly Sandia Naional Laboraories* Scalable Sysems Inegraion Deparmen PO BOX 5800 Albuquerque NM 87185-0817 smkelly@sandia.gov Ron Brighwell

More information

Reliability Analysis of Self-Healing Network using Discrete-Event Simulation

Reliability Analysis of Self-Healing Network using Discrete-Event Simulation Reliability Analysis of Self-Healing Network using Discrete-Event Simulation Thara Angskun, George Bosilca, Graham Fagg, Jelena Pješivac Grbović, and Jack Dongarra Department of Computer Science, The University

More information

NRMI: Natural and Efficient Middleware

NRMI: Natural and Efficient Middleware NRMI: Naural and Efficien Middleware Eli Tilevich and Yannis Smaragdakis Cener for Experimenal Research in Compuer Sysems (CERCS), College of Compuing, Georgia Tech {ilevich, yannis}@cc.gaech.edu Absrac

More information

A Scheme to Reduce Packet Loss during PMIPv6 Handover considering Authentication

A Scheme to Reduce Packet Loss during PMIPv6 Handover considering Authentication Inernaional Conference on Compuaional Sciences and Is Applicaions ICCSA 2008 A Scheme o Reduce Loss during P Handover considering Auhenicaion Seonggeun Ryu, Gye-Young Kim, Byunggi Kim, and Youngsong Mun

More information

FUZZY HUMAN/MACHINE RELIABILITY USING VHDL

FUZZY HUMAN/MACHINE RELIABILITY USING VHDL FUZZY HUMN/MCHINE RELIBILITY USING VHDL Carlos. Graciós M. 1, lejandro Díaz S. 2, Efrén Gorroiea H. 3 (1) Insiuo Tecnológico de Puebla v. Tecnológico 420. Col. Maravillas, C. P. 72220, Puebla, Pue. México

More information

MAXIMIZING LIFETIME OF POWERLIMITED NETWORK WITH ACTIVE MINIMUM SPANNING TREE AGGREGATION

MAXIMIZING LIFETIME OF POWERLIMITED NETWORK WITH ACTIVE MINIMUM SPANNING TREE AGGREGATION Published in he Proceedings of he 7 h Parallel and Disribued Compuing and Sysems, MAXIMIZING LIFETIME OF POWERLIMITED NETWORK WITH ACTIVE MINIMUM SPANNING TREE AGGREGATION Javed I. Khan and Asrar U. Haque

More information

Improved TLD Algorithm for Face Tracking

Improved TLD Algorithm for Face Tracking Absrac Improved TLD Algorihm for Face Tracking Huimin Li a, Chaojing Yu b and Jing Chen c Chongqing Universiy of Poss and Telecommunicaions, Chongqing 400065, China a li.huimin666@163.com, b 15023299065@163.com,

More information

Design and Application of Computer-aided English Online Examination System NONG DeChang 1, a

Design and Application of Computer-aided English Online Examination System NONG DeChang 1, a 3rd Inernaional Conference on Maerials Engineering, Manufacuring Technology and Conrol (ICMEMTC 2016) Design and Applicaion of Compuer-aided English Online Examinaion Sysem NONG DeChang 1, a 1,2 Guangxi

More information

Opportunistic Flooding in Low-Duty-Cycle Wireless Sensor Networks with Unreliable Links

Opportunistic Flooding in Low-Duty-Cycle Wireless Sensor Networks with Unreliable Links 1 in Low-uy-ycle Wireless Sensor Neworks wih Unreliable Links Shuo uo, Suden Member, IEEE, Liang He, Member, IEEE, Yu u, Member, IEEE, o Jiang, Suden Member, IEEE, and Tian He, Member, IEEE bsrac looding

More information

Video Content Description Using Fuzzy Spatio-Temporal Relations

Video Content Description Using Fuzzy Spatio-Temporal Relations Proceedings of he 4s Hawaii Inernaional Conference on Sysem Sciences - 008 Video Conen Descripion Using Fuzzy Spaio-Temporal Relaions rchana M. Rajurkar *, R.C. Joshi and Sananu Chaudhary 3 Dep of Compuer

More information

Chapter 3 MEDIA ACCESS CONTROL

Chapter 3 MEDIA ACCESS CONTROL Chaper 3 MEDIA ACCESS CONTROL Overview Moivaion SDMA, FDMA, TDMA Aloha Adapive Aloha Backoff proocols Reservaion schemes Polling Disribued Compuing Group Mobile Compuing Summer 2003 Disribued Compuing

More information

CAMERA CALIBRATION BY REGISTRATION STEREO RECONSTRUCTION TO 3D MODEL

CAMERA CALIBRATION BY REGISTRATION STEREO RECONSTRUCTION TO 3D MODEL CAMERA CALIBRATION BY REGISTRATION STEREO RECONSTRUCTION TO 3D MODEL Klečka Jan Docoral Degree Programme (1), FEEC BUT E-mail: xkleck01@sud.feec.vubr.cz Supervised by: Horák Karel E-mail: horak@feec.vubr.cz

More information

Partition-based document identifier assignment (PBDIA) algorithm. (long queries)

Partition-based document identifier assignment (PBDIA) algorithm. (long queries) ( ) Pariion-based documen idenifier assignmen (PBDIA) algorihm PBDIA (long queries) (parallel IR) :,,,, d-gap Compressing an invered file can grealy improve query performance of an informaion rerieval

More information

Packet Scheduling in a Low-Latency Optical Interconnect with Electronic Buffers

Packet Scheduling in a Low-Latency Optical Interconnect with Electronic Buffers Packe cheduling in a Low-Laency Opical Inerconnec wih Elecronic Buffers Lin Liu Zhenghao Zhang Yuanyuan Yang Dep Elecrical & Compuer Engineering Compuer cience Deparmen Dep Elecrical & Compuer Engineering

More information

Robust Multi-view Face Detection Using Error Correcting Output Codes

Robust Multi-view Face Detection Using Error Correcting Output Codes Robus Muli-view Face Deecion Using Error Correcing Oupu Codes Hongming Zhang,2, Wen GaoP P, Xilin Chen 2, Shiguang Shan 2, and Debin Zhao Deparmen of Compuer Science and Engineering, Harbin Insiue of Technolog

More information

A Progressive-ILP Based Routing Algorithm for Cross-Referencing Biochips

A Progressive-ILP Based Routing Algorithm for Cross-Referencing Biochips 16.3 A Progressive-ILP Based Rouing Algorihm for Cross-Referencing Biochips Ping-Hung Yuh 1, Sachin Sapanekar 2, Chia-Lin Yang 1, Yao-Wen Chang 3 1 Deparmen of Compuer Science and Informaion Engineering,

More information

CENG 477 Introduction to Computer Graphics. Modeling Transformations

CENG 477 Introduction to Computer Graphics. Modeling Transformations CENG 477 Inroducion o Compuer Graphics Modeling Transformaions Modeling Transformaions Model coordinaes o World coordinaes: Model coordinaes: All shapes wih heir local coordinaes and sies. world World

More information

A Routing Algorithm for Flip-Chip Design

A Routing Algorithm for Flip-Chip Design A Rouing Algorihm for Flip-hip Design Jia-Wei Fang, I-Jye Lin, and Yao-Wen hang, Graduae Insiue of Elecronics Engineering, Naional Taiwan Universiy, Taipei Deparmen of Elecrical Engineering, Naional Taiwan

More information

Location. Electrical. Loads. 2-wire mains-rated. 0.5 mm² to 1.5 mm² Max. length 300 m (with 1.5 mm² cable). Example: Belden 8471

Location. Electrical. Loads. 2-wire mains-rated. 0.5 mm² to 1.5 mm² Max. length 300 m (with 1.5 mm² cable). Example: Belden 8471 Produc Descripion Insallaion and User Guide Transiser Dimmer (454) The DIN rail mouned 454 is a 4channel ransisor dimmer. I can operae in one of wo modes; leading edge or railing edge. All 4 channels operae

More information

Open Access Research on an Improved Medical Image Enhancement Algorithm Based on P-M Model. Luo Aijing 1 and Yin Jin 2,* u = div( c u ) u

Open Access Research on an Improved Medical Image Enhancement Algorithm Based on P-M Model. Luo Aijing 1 and Yin Jin 2,* u = div( c u ) u Send Orders for Reprins o reprins@benhamscience.ae The Open Biomedical Engineering Journal, 5, 9, 9-3 9 Open Access Research on an Improved Medical Image Enhancemen Algorihm Based on P-M Model Luo Aijing

More information

Image segmentation. Motivation. Objective. Definitions. A classification of segmentation techniques. Assumptions for thresholding

Image segmentation. Motivation. Objective. Definitions. A classification of segmentation techniques. Assumptions for thresholding Moivaion Image segmenaion Which pixels belong o he same objec in an image/video sequence? (spaial segmenaion) Which frames belong o he same video sho? (emporal segmenaion) Which frames belong o he same

More information

Towards a Realistic Model for Failure Propagation in Interdependent Networks

Towards a Realistic Model for Failure Propagation in Interdependent Networks Towards a Realisic Model for Failure Propagaion in Inerdependen Neworks Agosino Suraro, Simone Silvesri, Mauro Coni, Sajal K. Das Deparmen of Mahemaics, Universiy of Padua, email: agosino.suraro@sudeni.unipd.i,

More information

Wireless LANs: MAC. Wireless LAN: MAC. IEEE protocol stack. Module W.lan MAC

Wireless LANs: MAC. Wireless LAN: MAC. IEEE protocol stack. Module W.lan MAC Wireless LANs: W.lan.3-2 Wireless LAN: 802.11MAC Dr.M.Y.Wu@CSE Shanghai Jiaoong Universiy Shanghai, China Module W.lan.3 Dr.W.Shu@ECE Universiy of New Mexico Albuquerque, NM, USA managemen PANs & Blueooh:

More information

IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, VOL. XX, NO. XX, XX XXXX 1

IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, VOL. XX, NO. XX, XX XXXX 1 This is he auhor's version of an aricle ha has been published in his journal. Changes were made o his version by he publisher prior o publicaion. IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, VOL. XX,

More information

Service Oriented Solution Modeling and Variation Propagation Analysis based on Architectural Building Blocks

Service Oriented Solution Modeling and Variation Propagation Analysis based on Architectural Building Blocks Carnegie Mellon Universiy From he SelecedWorks of Jia Zhang Ocober, 203 Service Oriened Soluion Modeling and Variaion Propagaion Analysis based on Archiecural uilding locks Liang-Jie Zhang Jia Zhang Available

More information

Algorithm for image reconstruction in multi-slice helical CT

Algorithm for image reconstruction in multi-slice helical CT Algorihm for image reconsrucion in muli-slice helical CT Kasuyuki Taguchi a) and Hiroshi Aradae Medical Engineering Laboraory, Toshiba Corporaion, 1385 Shimoishigami, Oawara, Tochigi 324-855, Japan Received

More information

Automatic Calculation of Coverage Profiles for Coverage-based Testing

Automatic Calculation of Coverage Profiles for Coverage-based Testing Auomaic Calculaion of Coverage Profiles for Coverage-based Tesing Raimund Kirner 1 and Waler Haas 1 Vienna Universiy of Technology, Insiue of Compuer Engineering, Vienna, Ausria, raimund@vmars.uwien.ac.a

More information

I. INTRODUCTION. Keywords -- Web Server, Perceived User Latency, HTTP, Local Measuring. interchangeably.

I. INTRODUCTION. Keywords -- Web Server, Perceived User Latency, HTTP, Local Measuring. interchangeably. Evaluaing Web User Perceived Laency Using Server Side Measuremens Marik Marshak 1 and Hanoch Levy School of Compuer Science Tel Aviv Universiy, Tel-Aviv, Israel mmarshak@emc.com, hanoch@pos.au.ac.il 1

More information

Computational Geometry in Wireless Networks - Routing. Presented by Heather M. Michaud

Computational Geometry in Wireless Networks - Routing. Presented by Heather M. Michaud Compaional Geomery in Wireless Neworks - Roing Presened by Heaher M. Michad 1 Ad Hoc Wireless Neworks No fixed pre-exising infrasrcre Nodes can be saic or mobile Assme nodes don move dring roing or opology

More information

Difficulty-aware Hybrid Search in Peer-to-Peer Networks

Difficulty-aware Hybrid Search in Peer-to-Peer Networks Difficuly-aware Hybrid Search in Peer-o-Peer Neworks Hanhua Chen, Hai Jin, Yunhao Liu, Lionel M. Ni School of Compuer Science and Technology Huazhong Univ. of Science and Technology {chenhanhua, hjin}@hus.edu.cn

More information

Rule-Based Multi-Query Optimization

Rule-Based Multi-Query Optimization Rule-Based Muli-Query Opimizaion Mingsheng Hong Dep. of Compuer cience Cornell Universiy mshong@cs.cornell.edu Johannes Gehrke Dep. of Compuer cience Cornell Universiy johannes@cs.cornell.edu Mirek Riedewald

More information

MATH Differential Equations September 15, 2008 Project 1, Fall 2008 Due: September 24, 2008

MATH Differential Equations September 15, 2008 Project 1, Fall 2008 Due: September 24, 2008 MATH 5 - Differenial Equaions Sepember 15, 8 Projec 1, Fall 8 Due: Sepember 4, 8 Lab 1.3 - Logisics Populaion Models wih Harvesing For his projec we consider lab 1.3 of Differenial Equaions pages 146 o

More information