Perfecting Preemption Threshold Scheduling for Object-Oriented Real-Time System Design: From The Perspective of Real-Time Synchronization

Size: px
Start display at page:

Download "Perfecting Preemption Threshold Scheduling for Object-Oriented Real-Time System Design: From The Perspective of Real-Time Synchronization"

Transcription

1 Perfectng Preempton Threshold Schedulng for Obect-Orented Real-Tme System Desgn: From The Perspectve of Real-Tme Synchronzaton Saehwa Km School of Electrcal Engneerng and Computer Scence Seoul Natonal Unversty Seoul 5-74 Korea Seongsoo Hong School of Electrcal Engneerng and Computer Scence Seoul Natonal Unversty Seoul 5-74 Korea Tae-Hyung Km Dept. of Computer Scence and Engneerng Hanyang Unversty Ansan Kyungg-Do Korea ABSTRACT In spte of the prolferaton of obect-orented desgn methodologes n contemporary software development ther applcaton to real-tme embedded systems has been lmted because of the practtoner s conservatve atttude toward handlng tmng constrants. In fact ths conservatve atttude s well-grounded because tradtonal prorty-based schedulng technques cannot be straghtforwardly ntegrated nto them. The automated mplementaton from the obect-orented real-tme desgns usually ncurs a large number of tasks whch under tradtonal prorty-based schedulng technques does not scale well due to excessve preempton overheads. Recently preempton threshold schedulng was ntroduced to reduce runtme mult-taskng overhead whle mprovng schedulablty by explotng non-preemptblty as much as possble. Unfortunately the preempton threshold schedulng cannot be drectly adopted nto the obect-orented desgn methods due to the lack of realtme synchronzaton. In ths paper we present the essental bass of real-tme synchronzaton for preempton threshold schedulng. Specfcally we ntegrate the prorty nhertance protocol the prorty celng protocol and the mmedate nhertance protocol nto preempton threshold schedulng. We also provde ther schedulablty analyses. Consequently the ntegrated scheme whch mnmzes worst-case context swtches s approprate for the automated mplementaton of real-tme obect-orented desgn models Categores and Subect Descrptors D.4. [Operatng Systems]: Process anagement schedulng. General Terms Theory Permsson to make dgtal or hard copes of all or part of ths work for personal or classroom use s granted wthout fee provded that copes are not made or dstrbuted for proft or commercal advantage and that copes bear ths notce and the full ctaton on the frst page. To copy otherwse or republsh to post on servers or to redstrbute to lsts requres pror specfc permsson and/or a fee. CTES 0 SCOPES 0 June 9-00 Berln Germany. Copyrght 00 AC /0/0006 $5.00. Keywords Preempton threshold schedulng real-tme synchronzaton prorty nhertance protocols prorty celng protocol obectorented real-tme system desgn.. INTRODUCTION Developng a real-tme embedded system s a sophstcated task. The monolthc approach to developng large and/or complex computer systems s comng to an end n the era of contemporary software development. Nonetheless ntrnsc dffcultes arsng from tmng constrants n the real-tme embedded applcatons stll te up practtoners wth a rather conservatve approach to developng embedded systems. any early real-tme system researchers have attached mportance to tmng analyss at desgn tme n order to derve a set of feasble tasks [ 3 4]. Ths has been qute successful and real-tme programmers are able to safely and relably derve a feasble set of tasks n advance. However as real-tme embedded systems get complex and sophstcated to meet the ncreased degree of safety relablty and performance requrements t becomes nevtable for real-tme system desgners to rely on systematc software desgn methodologes durng system development. Among a wde varety of software desgn methodologes obect-orented desgn methodologes have become domnant and popular snce they allow for easy software mantenance software reuse and component-based codng of complex real-tme systems that may evolve over tme. A recent trend n embedded systems development even tres to ht the lnes of hardware by harnessng software to mplement physcal layer functons n pursut of ts apparent advantages: re-usablty of hardware resources ease of upgrades flexblty to buld a dual- or multple-standards mplementaton and so on. Software defned rado (SDR s one of those representatve approaches. To cope wth the addtonal complexty the obect-orented desgn methodology should be seamlessly ntegrated nto tradtonal real-tme schedulablty analyss technques so that embedded system programmers can be equpped wth systematc software desgn methodologes and CASE tools as well. Tradtonal preemptve fxed prorty schedulng cannot be drectly used n ths ntegraton largely owng to ts excessve run-

2 tme overhead. Preempton threshold schedulng (PTS mproves both run-tme overhead and schedulablty but stll cannot be drectly used to ths end ether snce real-tme synchronzaton problems reman unsolved. In ths paper we present an essental bass for ther ntegraton by perfectng preempton threshold schedulng [ ] for real-tme synchronzaton.. Background Two prmary thrusts n embedded systems research are how to enhance schedulablty by devsng elaborate schedulng technques and how to ntegrate schedulablty analyss technques nto modern software engneerng desgn and mplementaton methodologes. Successfully appled schedulng technques have been based on preemptve fxed prorty schedulng n most real-tme systems developments snce u and ayland ntroduced t n ther poneerng paper []. However dynamc prorty schedulers can acheve hgher schedulablty than fxed ones and non-preemptve schedulers ncur less runtme overhead. Ths means that n fne-gran or medum-gran thread-based processng whch s the case especally n multthreaded mplementatons from real-tme obect-orented desgns the context swtchng overhead may overshadow the benefts of mult-threadng and preemptve schedulng. To allevate the run-tme overhead problem whle mprovng schedulablty ame at Express ogc Inc. ntroduced the noton of preempton threshold []. The preempton threshold regulates the degree of preemptveness n fxed-prorty schedulng. If the threshold of each task s the same as ts orgnal prorty then t s equvalent to the preemptve fxed prorty schedulng. If each task has the hghest threshold value n a system then t becomes a non-preemptve schedulng. As Wang and Saksena llustrated n [ ] the schedulablty of a task set under preemptve schedulng does not mply that the task set s also schedulable under non-preemptve schedulng and vce versa. Thus the preempton threshold schedulng s a good complement to preemptve fxed-prorty schedulng. It mproves schedulablty wthholds unnecessary preemptons reduces the number of tasks snce a group of non-preemptve tasks can be regarded as a unt and eventually helps allow for scalable real-tme system desgn. In the meantme there have been several research actvtes to ntegrate schedulablty analyss technques nto obect-orented desgn methodologes [5 6] based on the ROO (Real-tme Obect Orented odelng methodology [8]. The ntegraton s to provde many other advantages of obect-orented desgn methods and software engneerng tools for the automated mplementaton of real-tme embedded control systems from ROO-based desgn models. The most dffcult part of ths ntegraton s how to automatcally produce a thread-based mplementaton from a real-tme obect model wth tmng constrants. Saksena et al. ntate such a method usng one-toone mappng between obects and tasks for schedulablty [5] and mprove the performance usng preempton threshold schedulng to reduce the adverse effects of context swtchng n an automated mplementaton [].. Approaches and Contrbutons A real-tme system can be vewed as a collecton of concurrent obects that cooperate wth each other. Each obect may partcpate n multple system functons and as a result s subect to multple tmng constrants. In ths sense there are two extremes n mappng between obects and tasks. One s to map all obects nto a sngle task as n RoseRT for U-RT [9] and ObecTme for ROO [8]. Although ths s a practcal approach to reduce bng tme due to prorty nverson the system cannot be analyzed by the fxed-prorty schedulng theory. The alternatve s to map each obect nto a sngle task n order to perform tmng analyss wthn the obect-orented desgn technques as n [5]. Snce mult-taskng n ths mult-threaded mplementaton s expensve the preempton threshold schedulng s adopted n []. In our prevous work we presented a systematc schedulabltyaware method that can generate a mult-threaded mplementaton from a gven real-tme obect-orented desgn model [3 4]. Unlke the above mentoned approaches the mappng relatonshp between obects and tasks s not based to many-to-one or one-toone n our approach. Tasks are rather automatcally dentfed from a set of obects. Our method s a three-step process: ( dervng scenaros (end-to-end computaton unts ( dentfyng logcal threads and then (3 dervng physcal threads. ogcal threads are mapped from mutually exclusve scenaros and assgned prortes and preempton thresholds that guarantee schedulablty. To reduce the number of tasks those logcal threads are parttoned nto a mutually exclusve group of nonpreemptve ones. A group of non-preemptve logcal threads s n fact a physcal thread. Snce our approach s prmarly based upon real-tme synchronzaton under the preempton threshold schedulng t needs to be addressed n a comprehensve manner. Note that an obect-orented desgn produces a number of obect s for consstency of obect states and for mantenance of the run-tocompleton semantcs of a fnte state machne nsde each obect. Unfortunately real-tme synchronzaton under the preempton threshold schedulng has not been consdered yet. In ths paper we consder the problem of ntegratng three real-tme synchronzaton schemes nto the preempton threshold schedulng: they are the basc prorty nhertance protocol (BPI [5] the prorty celng protocol (PCP [5] and the mmedate prorty nhertance protocol (IIP [9]. In dong so we ntroduce the noton of effectve prorty nhertance and defne prorty celng and preempton threshold celng for reducng run-tme overhead. We also nvestgate ther schedulablty analyses. The remander of the paper s organzed as follows. Secton descrbes the task model wth proper defntons for the dscusson. In Secton 3 we dentfy the prorty nverson problem under PTS and present the revsed BPI protocol under PTS. In Sectons 4 and 5 we present the propertes of PCP and IIP under PTS and the schedulablty analyses of the proposed protocols respectvely. We conclude ths paper n Secton 6.. TASK ODE The task model used here s very smlar to that used n [ 5]. We assume a unprocessor envronment and allow only properly nested mutexes. We further assume a system wth a fxed set of tasks each of whch has a fxed perod known worst-case executon tme fxed prorty and preempton threshold. We denote a hgher prorty wth a larger value snce ths befts the ntutve meanng of beng a hgher threshold. The notatons and

3 ther descrptons used throughout the paper are summarzed n Table. Table. Summary of notatons for the task model. Notaton τ T C π γ p ϕ( Ψ Φ ξ z k β B S F R Descrpton A task The perod of task τ The worst-case executon tme of task τ The fxed-prorty of task τ The preempton threshold of task τ The effectve prorty of task τ A mutex (bnary semaphore Indvsble and operaton of mutex The celng of mutex The set of tasks that may use mutex The set of mutexes that task τ may use The set of mutexes that are currently ed by task τ The duraton of crtcal secton of task τ guarded by mutex k The bng tme of task τ due to synchronzaton The PTS bng tme of task τ The start tme of task τ The fnsh tme of task τ The response tme of task τ Under PTS each task has a preempton threshold n addton to ts regular prorty. Note that t s meanngful to assgn a task a preempton threshold that s no less than ts regular prorty snce a preempton threshold s used as an effectve run-tme prorty to control unnecessary preemptons. Snce the effectve prorty of a task s changed at run-tme due to prorty nhertance and task dspatchng under PTS t s desrable to precsely defne t. Conceptually the effectve prorty of a task s the prorty that s used by the kernel scheduler for selectng a task to be dspatched. Under PTS effectve prortes vary accordng to task states. We τ τ ={ ={ τ ={ ={ γ γ π π π π defne t n an operatonal manner as below. Effectve prorty p of τ = π f τ s released n ts perod and not yet dspatched; otherwse max(γ p p p such that τ τ τ are tasks bed by τ. In tradtonal prorty-based preemptve schedulng tasks may experence bng due to synchronzaton. Under PTS tasks may encounter another type of bng whch we name PTS bng. Task τ s sad to be n PTS bng f t s bed by a lower prorty task whose preempton threshold s hgher than π. We denote the duraton of PTS bng by B whle the duraton of other types of bng by β. 3. PREVENTING THE PRIORITY INVERSION PROBE Wthout usng prorty nhertance protocols catastrophc prorty nverson problems cannot be rectfed n a real-tme system that uses synchronzaton prmtves. The basc prorty nhertance protocol (BPI prevents preemptons that eventually cause prorty nverson. In ths secton we dentfy the prorty nverson problem under PTS formulate the BPI protocol under PTS and then descrbe ts propertes. 3. Prorty Inverson Problem under PTS The prorty nverson problem n tradtonal prorty-based schedulng occurs when a medum prorty task preempts a lower prorty task that bs a hgher prorty task. Under PTS we reformulate the orgnal problem nto the effectve prorty nverson problem by substtutng a prorty of a task wth ts effectve prorty. The ratonale behnd ths formulaton s obvous snce an effectve prorty takes the place of a prorty under PTS. In ths formulaton gven that task τ s bed by prorty nverson occurs ( when a medum prorty task wth γ π π preempts ; or ( when another task wth π π H γ preempts. These cases are llustrated n Fgure (a and (b τ τ ={ ={ τ ={ ={ γ γ π π π H π H γ γ τ preempt (attempt to bed by uncontrolled prorty nverson preempt preempt τ (attempt to bed by preempt uncontrolled prorty nverson t 0 t t t 3 t 4 (a Case. t 0 t t t 3 t 4 (b Case. Fgure. Prorty nverson problem under PTS.

4 respectvely. τ arrve (attempt to bed by τ preempt τ τ ={ ={ τ ={ ={ γ γ π π π π π H π γ H γ nhert γ drect bng arrve t 0 t t t 3 t 4 t 5 push-through bng t 6 push-through bng Fgure. Preventon of prorty nverson by BPI under PTS. 3. BPI under PTS We defne the BPI protocol under PTS usng effectve prortes as below. Protocol : BPI under PTS. Inhertance of effectve prortes. When task s bed by task wth π H > γ task nherts p H from task. Recovery of effectve prortes. When task exts from a crtcal secton task recovers p that t had before enterng that crtcal secton. We can easly see that ths defnton of BPI prevents preemptons that eventually cause prorty nverson. Fgure shows an example that follows Protocol. Snce nether does nor preempt τ prorty nverson s avoded. 3.3 Propertes of BPI under PTS BPI under PTS bears smlar propertes wth the orgnal BPI n [5]. In the orgnal protocol a task can be bed n one of three forms of bng: ( drect bng ( push-through bng and (3 transtve bng. In our protocol a task can be bed addtonally n PTS bng. Drect bng occurs when a hgher prorty task attempts to a mutex ed by a lower prorty task. It s to ensure the consstency of a non-preemptble shared resource. Push-through bng occurs under PTS when a medum prorty task attempts to preempt a lower prorty task that s bng a hgher prorty task as descrbed below. Push-through bng under PTS. Consder three tasks wth γ π γ π H and π γ H. If task s bed by task that already bed ths stuaton s referred to as push-through bng under PTS. In the above defnton task falls nto push-through bng snce the effectve prorty of task s greater than that of due to effectve prorty nhertance. Push-through bng s ntroduced to prevent preemptons that cause prorty nverson. In Fgure tasks and are bed n push-through bng durng the perod of (t 4 t 5 and (t 5 t 6 respectvely. Fnally transtve bng occurs when task s bed by whch n turn s bed by another task. Fgure 3 (a shows an example for transtve bng. Durng the perod of (t 6 t 7 s bed by whle bng. Therefore s ndrectly bed by n a transtve manner. As such transtve bng occurs when mutexes are accessed n a nested fashon. Addtonally a task can encounter chaned bng. Chaned ={ ={ P( V( ={ ={ P( V( ={ ={ γ π γ γ π π γ H π H (attempt to bed by transtve bng preempt preempt (attempt to bed by drect bng preempt (attempt to bed by celng bng t 0 t t t 3 t 4 t 5 t 6 t 7 t 8 (a t 8 t 0 t t t 3 t 4 t 5 t 6 t 7 t 8 t 9 t 0 t t t 3 (b Fgure 3. (a Transtve bng n BPI (b transtve bng preventon n PCP.

5 ={ ={ P( V( ={ ={ P( V( ={ ={ γ π γ γ π π γ H π H (attempt to bed by (attempt to bed by (attempt to bed by preempt preempt (attempt to bed by celng bng t 0 t t t 3 t 4 t 5 t 6 t 7 t 8 t 9 t 0 t 0 t t t 3 t 4 t 5 t 6 t 7 t 8 t 9 t 0 t t t 3 (a Fgure 4. (a Chaned bng n BPI (b chaned bng preventon n PCP. (b bng (or a chan of bng s sad to occur f a task s repeatedly bed when t enters ts crtcal sectons. An example for chaned bng s llustrated n Fgure 4 (a where task s bed durng the perod of (t 6 t 7 and bed agan durng (t 9 t 0. Note that chaned bng s not a form of bng but refers to a stuaton where a task s bed more than once. Note that the BPI protocol alone does not prevent dead. Dead may occur when multple tasks try to access nested mutexes n a crcular manner. Fgure 5 shows an example for dead. In ths example both tasks and need to acqure two mutexes and. As shown wats for to release and wats for to release smultaneously whle and are holdng and respectvely. Therefore a dead occurs at tme t PREVENTING DEADOCK TRANSITIVE BOCKING AND CHAINED BOCKING Although BPI prevents prorty nverson by means of pushthrough bng t may ncur dead and excessvely long bng delay due to transtve and chaned bng as dscussed n Secton 3.3. The prorty celng protocol (PCP was ntroduced to solve these problems [5]. In ths secton we defne two versons of PCP under PTS: one wth prorty celngs and the other wth preempton threshold celngs. We refer to the former as PC-PCP and the latter as PTC-PCP. Then we nvestgate ther propertes compare them and present schedulablty analyss. ={ ={ P( V( ={ ={ P( V( γ π γ H π H preempt (attempt to bed by (attempt to bed by preempt (try to bed by celng bng dead t 0 t t t 3 t 4 t 5 (a t 0 t t t 3 t 4 t 5 t 6 t 7 t 8 t 9 t 0 t (b Fgure 5. (a Dead n BPI (b dead preventon n PCP.

6 4. PCP under PTS wth Prorty Celngs The underlyng dea of the orgnal PCP algorthm s to allow a task to a mutex only f t can make sure that all mutexes that the task and ts hgher prorty tasks may use are not ed by lower prorty tasks. If not t forces the task to wat for those mutexes to be ed. To realze ths dea Rakumar et al. ntroduced the noton of a prorty celng and assocated t wth each mutex [5]. We defne PCP under PTS wth prorty celngs by combnng an offlne celng protocol and an onlne ng protocol n a smlar manner as n [6]. Protocol : PC-PCP. Offlne celng protocol CP. Each mutex s assgned a prorty celng whose value s gven by ϕ( = max{π τ Ψ } where Ψ s a set of tasks that may use mutex. Onlne ng protocol P. A system celng whch s a system-wde schedulng attrbute s dynamcally set to the maxmum of prorty celngs of all mutexes beng ed n the system. P. In order for task τ to enter ts crtcal secton π should be hgher than the system celng. P3. Inhertng effectve prortes If hgh prorty task s bed by low prorty task wth π H > γ task nherts p H from task as ts effectve prorty. P4. Recoverng effectve prortes When task exts from a crtcal secton task recovers p that t had before enterng that crtcal secton. In ths defnton of PCP under PTS effectve prortes are used for prorty nhertance (as specfed n P3 and P4 whle regular prortes are used for celngs of mutexes (as specfed n CP and P. It s qute straghtforward to use effectve prortes for prorty nhertance snce t s a mere adopton of BPI to solve the prorty nverson problemwe now show that PC-PCP prevents dead transtve bng and chaned bng. To begn wth we ntroduce a ng rule that can guarantee the avodance of dead transtve bng and chaned bng. Then we show our PCP defnton satsfes ths rule. ockng Rule. Task τ s allowed to a mutex only f all mutexes that may use are not ed by lower prorty tasks where π H π. Theorem. Conformng to ockng Rule guarantees the preventon of dead transtve bng and chaned bng. Proof. Conformng to ockng Rule ensures that a task cannot enter ts crtcal secton untl all mutexes t may use are ed. Therefore crcular watng cannot occur and thus dead s prevented. To show that conformng to ockng Rule guarantees the preventon of transtve and chaned bng we show that a task s bed at most once by at most one task under the ockng Rule. Suppose that task τ attempts to a mutex whle there s preempted task that has ed mutexes that may be used by hgher prorty or preempton threshold task where π H π. Then the ockng Rule forces task τ to get bed frst and wat for to those mutexes. Consequently there s always at most only one such task that s mutexes that may be used by where π H π. Accordngly whenever a task arrves t sees at most one (lower prorty task that has acqured mutexes t may use. Therefore t s guaranteed that a task s bed (to acqure ts mutexes at most once (when t frst tres to any mutex and by at most one lower prorty task. Therefore transtve and chaned bng cannot happen. Theorem. PC-PCP conforms to ockng Rule. Proof. Suppose that the PC-PCP does not conform to ockng Rule. Then task τ may a mutex when there s a ed mutex H that may be used by task where π H π. However accordng to CP of Protocol the prorty celng of H ϕ( H s equal to π H. ={ ={ H H H H ={ ={ ={ ={ γ π γ π π H γ π H γ γ γ H H arrve H H arrve (attempt to c bed by unnecessary bng t 0 t t t 3 t 4 t 5 t 6 t 7 t 0 t t t 3 t 4 t 5 t 6 t 7 (a Fgure 6. Examples: PCP under PTS usng (a prorty celngs and (b preempton threshold celngs. (b

7 When task τ attempts to any mutex the system celng equals ϕ( H = π H by P. Therefore τ should be bed snce π s not hgher than π H by P. Ths s a contradcton. Fgure 3 (b Fgure 4 (b and Fgure 5 (b llustrate how PCP prevents transtve bng chaned bng and dead respectvely. We have seen that n PCP a task can be bed only when t tres to enter ts crtcal secton for the frst tme: once a task successfully acqures any of ts mutexes t wll never get bed agan. Ths obvously mples that a task can be bed at most once by at most one lower prorty task. In PCP under PTS a task can encounter one of four forms of bng: ( drect bng ( push-through bng (3 celng bng [5] and (4 PTS bng. A task encounters celng bng when t attempts to enter a crtcal secton wth ts prorty not greater than the system celng. Fgure 3 (b Fgure 4 (b and Fgure 5 (b show examples of celng bng n PTS. Ths helps prevent dead transtve bng and chaned bng. 4. PCP under PTS wth Preempton Threshold Celngs It s not obvous why we used regular prortes to defne celngs of mutexes nstead of preempton threshold celng n Secton 4.. In fact we can use preempton threshold celngs for that purpose and defne another verson of PCP under PTS accordngly. The result s PTC-PCP. It s the same as PC-PCP of Protocol except that the prortes of CP and P of Protocol are replaced wth preempton thresholds. ockng Rule. Task τ s allowed to a mutex only f all mutexes that may use are not ed by lower prorty tasks where γ H γ. Theorem 3. Conformng to ockng Rule guarantees the preventon of dead transtve bng and chaned bng. Proof. Ths can be proved n a manner smlar to Theorem by substtutng prortes wth preempton thresholds. Theorem 4. PTC-PCP conforms to ockng Rule. Proof. Ths can be proved n a manner smlar to Theorem by substtutng prorty celngs of Theorem wth preempton threshold celngs and ockng Rule wth ockng Rule. 4.3 Comparson between PC-PCP and PTC- PCP Now we wll show that PC-PCP performs better than PTC-PCP wth respect to run-tme costs and the average response tme of a task set. The latter may ( ncur unnecessary context swtches and ( lead to longer response tmes due to unnecessary bngs. In most cases of practcal nterest a task has a preempton threshold that s no less than ts fxed prorty so as to avod unnecessary context swtches. In [] a task s even assgned the maxmum possble preempton threshold. Ths s ndeed a desrable heurstc to reduce unnecessary preemptons n preemptve fxed-prorty schedulng. However ths heurstc may not work well n PTC-PCP. In PTC- PCP wth ths heurstcs the system ends up wth a large number of tasks assgned the maxmum preempton threshold whch often forces hgh prorty tasks to serously suffer from celng bng. Fgure 6 demonstrates the stuaton. Suppose there are three tasks and wth γ π π H and γ =γ H. At t task acqures and then task arrves and preempts task at t. At t 3 t attempts to acqure H. At that pont n tme the system celng s equal to ϕ(. If we adopt prorty celngs (PC-PCP as depcted n Fgure 6 (a the system celng at t 3 s π. Snce π H > π task acqures H and contnues to run at t 3. On the other hand f we adopt preempton threshold celngs (PTC-PCP as depcted n Fgure 6 (b the system celng at t 3 s γ. Snce γ H s not greater than γ task gets bed at t 3 and resumes at t 4. Thus two extra context swtches occur at t 3 and t 4. oreover the response tme of task becomes longer n Fgure 6 (b than n Fgure 6 (a due to unnecessary bng at t 3. When runnng task τ tres to enter ts crtcal secton n PTS there s a case where s bed unnecessarly only n PTC-PCP: when the preempted task wth γ π H has acqured mutex l whch may be requested by τ l wth π l γ H and γ l γ H. Ths type of celng bng does not contrbute to preventng dead transtve bng or chaned bng snce ths bng s for mutexes that may be requested by lower prorty tasks. For every such case there are two addtonal context swtches n PTC-PCP as compared to PC-PCP. Note that such cases tend to be very frequent due to maxmum preempton threshold assgnment polcy n PTS. From ths observaton we recommend PC-PCP for obvous reasons: to avod unnecessary context swtches and to reduce the response tmes of hgher prorty tasks. 4.4 Schedulablty Analyss of PC-PCP For the schedulablty analyss of PC-PCP we adopt the worstcase response tme analyss n [0]. Under PTS the sets of nterferng hgher prorty tasks before and after a task gets the CPU are dfferent. Therefore for each task τ we should analyze ts start tme S and fnsh tme F separately. The equatons for the response tme analyss of PCP under PTS for task τ are as follows. β = max B γ π γ π > π 0 f Φ = φ { z k :: ϕ( k π } otherwse { C + β } C + β = max ( n n S + S = max { B β } + + C T (3 π > π 0 f B = 0 or β 0 δ = (4 otherwse (

8 F n+ = S + C + π > γ F T n S + C + β δ T (5 R = F (6 In PCP a task may be bed once n one of three forms of bng: push-through drect and celng bng. The worst case bng duraton for these types of bng for task τ (β s formulated n Equaton ( where z k s the duraton of the crtcal secton of task τ guarded by mutex k. Note that f the set of mutexes that task τ may use s empty (Φ =0 the bng tme under PCP s zero. Addtonally n PTS a task can experence PTS bng. The worst-case duraton of PTS bng for task τ s formulated n Equaton (. The start tme of task τ (S s formulated n Equaton (3: a task may be bed once n ether PTS bng or push-through bng before ts executon snce ( both bngs are caused by the same runnng task when the task arrves and ( the sets of tasks that can cause PTS bng and push-through bngs are mutually exclusve as shown n Equatons ( and (. The fnsh tme of task τ (F s formulated n Equaton (5 where δ s calculated from Equaton (4: task τ can be bed once agan durng ts executon n ether drect or celng bng. As shown n Equaton (4 ths type of bng (β δ becomes zero ( f PTS bng cannot occur (B = 0 or ( f the PCP bng tme of the task causng the PTS bng s not zero (β 0. Note that f task τ has encountered push-through bng or the task that caused PTS bng ( has encountered PCP bng (β 0 drect or celng bng cannot occur. 5. INIIZING CONTEXT SWITCHES In PCP a task can be bed once when t attempts to enter a crtcal secton. Alternatvely f we gve a task a chance to be bed when t s released for the frst tme n ts perod t wll never be bed agan durng ts executon. If we choose to use ths approach we can reduce two context swtches assocated wth each bng of a task n PCP. Ths alternatve whch s already wdely used n practce s called the mmedate nhertance protocol (IIP or the PCP emulaton protocol wth flag _POSIX_THREAD_PRIO_PROTECT n IEEE POSIX 003.c [9]. In ths secton we defne two versons of IIP under PTS: one wth prorty celngs and the other wth preempton threshold celngs. We refer to the former as PC-IIP and the latter as PTC-IIP. Then we nvestgate ther propertes compare them and present schedulablty analyss. 5. IIP under PTS wth Prorty Celngs The underlyng dea of the orgnal IIP algorthm s to allow a task to start ts executon only f t can make sure that all mutexes that the task and ts hgher prorty tasks may use are not ed by lower prorty tasks. If not t forces the task to wat for those mutexes to be ed. IIP bears many smlartes and benefts wth PCP. It can prevent prorty nverson dead transtve bng and chaned bng. oreover t can reduce the number of context swtches wthout sacrfcng the worst-case response tme. As n PCP we can defne two versons of IIP: PC-IIP and PTC- IIP. PC-IIP uses effectve prortes for prorty nhertance and regular prortes for prorty celngs of mutexes. Its offlne celng protocol and onlne ng protocol are presented below. Protocol 3: PC-IIP. Offlne celng protocol CP. Each mutex s assgned a prorty celng gven by ϕ( = max{π τ Ψ } where Ψ s a set of tasks that may use mutex. Onlne ng protocol P. Inhertng celng If task τ c acqures a mutex ts effectve prorty s set to p c = max{p c ϕ( ξ c } where ξ c s a set of mutexes that are currently ed by τ c. P. Recoverng effectve prortes When task τ c exts from a crtcal secton task τ c recovers p c that t had before enterng that crtcal secton. Whle the offlne celng protocol (CP s the same as that of PC- PCP the ng protocol (P and P s much more smplfed. Ths s because IIP forces prorty nhertance whenever a task s a mutex whle PCP does so only when a task bs a hgher prorty task. It s obvous that our PC-IIP defnton prevents prorty nverson dead transtve bng and chaned bng snce IIP s merely a specal case of PCP. ater we compare IIP wth PCP and dscuss the advantages of IIP over PCP. arrve bed by ={ ={ H H H H ={ ={ ={ ={ γ γ π π π H π H γ γ γ γ H H H H unnecessary bng t 0 t t t 3 t 4 t 5 t 6 t 7 Fgure 7. Example: IIP under PTS wth preempton threshold celng. 5. IIP under PTS wth Preempton Threshold Celngs As mentoned above the IIP under PTS can be defned wth ether prorty celngs (PC-IIP or preempton threshold celngs (PTC- IIP. We compare these two versons of IIP defntons usng the same example task set shown n Secton 4.3. Fgure 6 (a can be reused to llustrate IIP wth prorty celngs. Fgure 7 shows an example of PTC-IIP. As n PCP under PTS PTC-IIP yelds extra

9 bngs thus makes the response tmes of hgher prorty tasks longer. On the other hand note that the context-swtchng overhead remans the same n the two versons snce bng n IIP does not accompany context swtches. As n PCP we recommend PC-IIP over PTC-IIP. 5.3 IIP vs. PCP The most strkng dfference between IIP and PCP s that a task n IIP can encounter only celng bng: there s nether drect bng nor push-through bng n IIP. The number of context swtches ncurred by IIP can be as low as a half of the number ncurred by PCP. The best effcency under IIP occurs when every task n the system enters at least one crtcal secton. On the other hand f qute a few tasks do not enter a crtcal secton at all IIP may yeld extra bngs. In ether case the number of context swtches n IIP s always lower than n PCP. Ths s because bng n IIP does not accompany context swtches snce t always occurs before a task starts executon. However t s also true that the extra bngs encountered by hgh prorty tasks ncrease ther average response tmes n IIP. 5.4 Schedulablty Analyss of PC-IIP We provde the schedulablty analyss for PC-IIP. In IIP under PTS a task can be bed only once before ts start tme. Snce a task can be bed even f t does not enter any crtcal secton before ts start tme the equaton for bng tme β n IIP s formulated as Equaton (7. On the other hand snce a task s not bed durng ts executon the PTS bng n IIP under PTS s formulated as Equaton (8 and the fnsh tme s formulated as Equaton (9. Wthout these the formulatons for S and R n Equatons (4 and (6 n Secton 4.3 are adopted as they are. β = max { z :: ϕ( π } γ π k k (7 { } B = C max γ π > π (8 n+ = S + C + π > γ F F T n S + C T (9 6. CONCUSIONS In spte of the prolferaton of obect-orented desgn methodologes n contemporary software development ther applcaton to real-tme embedded systems has been lmted for the lack of proper real-tme schedulng theory that can be seamlessly ntegrated nto these methods. Specfcally popular preemptve fxed prorty schedulng cannot be drectly used n real-tme obect-orented desgn methodologes largely owng to ts excessve run-tme overhead. The preempton threshold schedulng has been recently suggested to ths end snce t mproves both run-tme overhead and schedulablty. Unfortunately t lacks real-tme synchronzaton capabltes. To solve ths problem n ths paper we have adopted three wellknown real-tme synchronzaton protocols and ntegrated them nto the preempton threshold schedulng. They are the basc prorty nhertance protocol (BPI the prorty celng protocol (PCP and the mmedate nhertance protocol (IIP. We have also presented ther schedulablty analyses. Snce each task under PTS has two schedulng attrbutes t s not ntutve durng the ntegraton whch schedulng attrbute should be used for prorty nhertance and for prorty celngs of mutexes. To clarfy ths problem we have ntroduced the noton of the effectve prorty whch s conceptually a task prorty that s used by the kernel scheduler for selectng a task to be dspatched. Wth ths we have dentfed the prorty nverson problem under PTS and presented the BPI protocol under PTS that avods such prorty nverson. Snce BPI alone cannot prevent dead we have also presented PCP and IIP under PTS. In these protocols we used effectve prortes for prorty nhertance and regular prortes for prorty celngs of mutexes. We have shown that two protocols wth prorty celngs yeld the smaller number of context swtches and shorter response tmes for hgher prorty tasks than those wth preempton threshold celngs. We have also shown that IIP further reduces the number context swtches compared to PCP. Currently we are mplementng a RoseRT-based CASE tool that s capable of dervng tasks from an obect-orented desgn model. We are ntegratng nto the CASE tool the preempton threshold schedulng and the IIP algorthm proposed n ths paper. We are also conductng extensve experments to show the vablty of our approach. The results look promsng. 7. ACKNOWEDGEENTS The authors would lke to thank Jamson Allen asse for hs frutful comments on drafts of ths artcle. The comments from the anonymous revewers further mproved the qualty. The work reported n ths paper s supported n part by OST under the Natonal Research aboratory (NR grant 000-N-N-0-C-36 by Automaton and Systems Research Insttute (ASRI and by Automatc Control Research Center (ACRC. 8. REFERENCES []. Saksena and Y. Wang. Scalable real-tme system desgn usng preempton thresholds In Proceedngs of IEEE Real- Tme Systems Symposum pp [] Y. Wang and. Saksena. Schedulng fxed prorty tasks wth preempton threshold. In Proceedngs of IEEE Real- Tme Computng Systems and Applcatons Symposum pp [3] S. Km S. Cho and S. Hong. Schedulablty-aware mappng of real-tme obect-orented models to mult-threaded mplementatons. In Proceedngs of Internatonal Conference on Real-Tme Computng Systems and Applcatons pp [4] S. Km S. Hong and N. Chang Scenaro-based mplementaton archtecture for real-tme obect-orented models In Proceedngs of IEEE Internatonal Workshop on Obect-orented Real-tme Dependable Systems 00. [5] R. Rakumar. Sha and J. ehoczky. Prorty nhertance protocols: an approach to real-tme synchronzaton. In IEEE Transactons on Software Engneerng vol. 39 pp

10 [6]. Chen and K. n. Dynamc prorty celngs: A concurrencly control protocol for real-tme systems. In Real- Tme Systems Journal vol. pp [7] T. P. Baker. A stack-based resource allocaton polcy for real-tme processes. In Proceedngs of IEEE Real-Tme Systems Symposum pp [8] T. P. Baker. Stack-based schedulng of real-tme processes. In Real-Tme Systems Journal vol. 3 num. pp [9] Insttute for Electrcal and Electronc Engneers. In IEEE Std. 003.c-995 POSIX Part : System applcaton program nterface - amendment :threads extenson 995. [0] K. Tndell A. Burns and A. Wellngs. An extendble approach for analyzng fxed prorty hard real-tme tasks. In Real-Tme Systems Journal vol. 6 pp [] C. u and J. ayland Schedulng algorthm for multprogrammng n a hard real-tme envronment. In Journal of the AC vol. 0( pp Jan [] W. ame Preempton-Threshold Whte Paper Express ogc Inc. Avalable at [3]. Harbour. Klen and J. ehoczky Fxed Prorty Schedulng of Perodc Tasks wth Varyng Executon Prorty In Proceedngs of IEEE Real-Tme Systems Symposum pp. 6-8 Dec. 99. [4] J. ehoczky. Sha and Y. Dng The Rate onotonc Schedulng Algorthm: Exact Characterzaton and Average Case Behavor. In Proceedngs of IEEE Real-Tme Systems Symposum pp Dec [5]. Saksena P. Freeman and P. Rodzewcz Gudelnes for Automated Implementaton of Executable Obect Orented odels for Real-Tme Embedded Control Systems In Proceedngs of IEEE Real-Tme Systems Symposum pp Dec [6]. Saksena A. Ptak P. Freeman and P. Rodzewcz Schedulablty Analyss for Automated Implementatons of Real-Tme Obect-Orented odels In Proceedngs of IEEE Real-Tme Systems Symposum pp. 9-0 Dec [7] H. Gomma Software Desgn ethods for Concurrent and Real-Tme Systems Addson-Wesley 993. [8] B. Selc G. Gullekson and P. T. Ward Real-Tme Obect- Orented odelng. John Wesley and Sons 994. [9] Ratonal Software

AADL : about scheduling analysis

AADL : about scheduling analysis AADL : about schedulng analyss Schedulng analyss, what s t? Embedded real-tme crtcal systems have temporal constrants to meet (e.g. deadlne). Many systems are bult wth operatng systems provdng multtaskng

More information

Real-time Scheduling

Real-time Scheduling Real-tme Schedulng COE718: Embedded System Desgn http://www.ee.ryerson.ca/~courses/coe718/ Dr. Gul N. Khan http://www.ee.ryerson.ca/~gnkhan Electrcal and Computer Engneerng Ryerson Unversty Overvew RTX

More information

Integrating Real-Time Synchronization Schemes into Preemption Threshold Scheduling

Integrating Real-Time Synchronization Schemes into Preemption Threshold Scheduling Integrating Real-Time Synchronization Schemes into Preemption Threshold Scheduling Saehwa Kim, Seongsoo Hong and Tae-Hyung Kim {ksaehwa, sshong}@redwood.snu.ac.kr, tkim@cse.hanyang.ac.kr Abstract Preemption

More information

Multitasking and Real-time Scheduling

Multitasking and Real-time Scheduling Multtaskng and Real-tme Schedulng EE8205: Embedded Computer Systems http://www.ee.ryerson.ca/~courses/ee8205/ Dr. Gul N. Khan http://www.ee.ryerson.ca/~gnkhan Electrcal and Computer Engneerng Ryerson Unversty

More information

Lecture 7 Real Time Task Scheduling. Forrest Brewer

Lecture 7 Real Time Task Scheduling. Forrest Brewer Lecture 7 Real Tme Task Schedulng Forrest Brewer Real Tme ANSI defnes real tme as A Real tme process s a process whch delvers the results of processng n a gven tme span A data may requre processng at a

More information

Parallelism for Nested Loops with Non-uniform and Flow Dependences

Parallelism for Nested Loops with Non-uniform and Flow Dependences Parallelsm for Nested Loops wth Non-unform and Flow Dependences Sam-Jn Jeong Dept. of Informaton & Communcaton Engneerng, Cheonan Unversty, 5, Anseo-dong, Cheonan, Chungnam, 330-80, Korea. seong@cheonan.ac.kr

More information

Maintaining temporal validity of real-time data on non-continuously executing resources

Maintaining temporal validity of real-time data on non-continuously executing resources Mantanng temporal valdty of real-tme data on non-contnuously executng resources Tan Ba, Hong Lu and Juan Yang Hunan Insttute of Scence and Technology, College of Computer Scence, 44, Yueyang, Chna Wuhan

More information

A mathematical programming approach to the analysis, design and scheduling of offshore oilfields

A mathematical programming approach to the analysis, design and scheduling of offshore oilfields 17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 A mathematcal programmng approach to the analyss, desgn and

More information

Compiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz

Compiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz Compler Desgn Sprng 2014 Regster Allocaton Sample Exercses and Solutons Prof. Pedro C. Dnz USC / Informaton Scences Insttute 4676 Admralty Way, Sute 1001 Marna del Rey, Calforna 90292 pedro@s.edu Regster

More information

An Optimal Algorithm for Prufer Codes *

An Optimal Algorithm for Prufer Codes * J. Software Engneerng & Applcatons, 2009, 2: 111-115 do:10.4236/jsea.2009.22016 Publshed Onlne July 2009 (www.scrp.org/journal/jsea) An Optmal Algorthm for Prufer Codes * Xaodong Wang 1, 2, Le Wang 3,

More information

Module Management Tool in Software Development Organizations

Module Management Tool in Software Development Organizations Journal of Computer Scence (5): 8-, 7 ISSN 59-66 7 Scence Publcatons Management Tool n Software Development Organzatons Ahmad A. Al-Rababah and Mohammad A. Al-Rababah Faculty of IT, Al-Ahlyyah Amman Unversty,

More information

A Binarization Algorithm specialized on Document Images and Photos

A Binarization Algorithm specialized on Document Images and Photos A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a

More information

6.854 Advanced Algorithms Petar Maymounkov Problem Set 11 (November 23, 2005) With: Benjamin Rossman, Oren Weimann, and Pouya Kheradpour

6.854 Advanced Algorithms Petar Maymounkov Problem Set 11 (November 23, 2005) With: Benjamin Rossman, Oren Weimann, and Pouya Kheradpour 6.854 Advanced Algorthms Petar Maymounkov Problem Set 11 (November 23, 2005) Wth: Benjamn Rossman, Oren Wemann, and Pouya Kheradpour Problem 1. We reduce vertex cover to MAX-SAT wth weghts, such that the

More information

For instance, ; the five basic number-sets are increasingly more n A B & B A A = B (1)

For instance, ; the five basic number-sets are increasingly more n A B & B A A = B (1) Secton 1.2 Subsets and the Boolean operatons on sets If every element of the set A s an element of the set B, we say that A s a subset of B, or that A s contaned n B, or that B contans A, and we wrte A

More information

Mathematics 256 a course in differential equations for engineering students

Mathematics 256 a course in differential equations for engineering students Mathematcs 56 a course n dfferental equatons for engneerng students Chapter 5. More effcent methods of numercal soluton Euler s method s qute neffcent. Because the error s essentally proportonal to the

More information

Virtual Machine Migration based on Trust Measurement of Computer Node

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

More information

Course Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms

Course Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms Course Introducton Course Topcs Exams, abs, Proects A quc loo at a few algorthms 1 Advanced Data Structures and Algorthms Descrpton: We are gong to dscuss algorthm complexty analyss, algorthm desgn technques

More information

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization Problem efntons and Evaluaton Crtera for Computatonal Expensve Optmzaton B. Lu 1, Q. Chen and Q. Zhang 3, J. J. Lang 4, P. N. Suganthan, B. Y. Qu 6 1 epartment of Computng, Glyndwr Unversty, UK Faclty

More information

Burst Round Robin as a Proportional-Share Scheduling Algorithm

Burst Round Robin as a Proportional-Share Scheduling Algorithm Burst Round Robn as a Proportonal-Share Schedulng Algorthm Tarek Helmy * Abdelkader Dekdouk ** * College of Computer Scence & Engneerng, Kng Fahd Unversty of Petroleum and Mnerals, Dhahran 31261, Saud

More information

Support Vector Machines

Support Vector Machines /9/207 MIST.6060 Busness Intellgence and Data Mnng What are Support Vector Machnes? Support Vector Machnes Support Vector Machnes (SVMs) are supervsed learnng technques that analyze data and recognze patterns.

More information

Related-Mode Attacks on CTR Encryption Mode

Related-Mode Attacks on CTR Encryption Mode Internatonal Journal of Network Securty, Vol.4, No.3, PP.282 287, May 2007 282 Related-Mode Attacks on CTR Encrypton Mode Dayn Wang, Dongda Ln, and Wenlng Wu (Correspondng author: Dayn Wang) Key Laboratory

More information

Hybrid Job Scheduling Mechanism Using a Backfill-based Multi-queue Strategy in Distributed Grid Computing

Hybrid Job Scheduling Mechanism Using a Backfill-based Multi-queue Strategy in Distributed Grid Computing IJCSNS Internatonal Journal of Computer Scence and Network Securty, VOL.12 No.9, September 2012 39 Hybrd Job Schedulng Mechansm Usng a Backfll-based Mult-queue Strategy n Dstrbuted Grd Computng Ken Park

More information

Video Proxy System for a Large-scale VOD System (DINA)

Video Proxy System for a Large-scale VOD System (DINA) Vdeo Proxy System for a Large-scale VOD System (DINA) KWUN-CHUNG CHAN #, KWOK-WAI CHEUNG *# #Department of Informaton Engneerng *Centre of Innovaton and Technology The Chnese Unversty of Hong Kong SHATIN,

More information

Scheduling. In general, a scheduling scheme provides two features: An algorithm for ordering the use of system resources (in particular the CPUs)

Scheduling. In general, a scheduling scheme provides two features: An algorithm for ordering the use of system resources (in particular the CPUs) Schedulng Goal To understand the role that schedulng and schedulablty analyss plays n predctng that real-tme applcatons meet ther deadlnes Topcs Smple process model The cyclc executve approach Process-based

More information

On-line Scheduling Algorithm with Precedence Constraint in Embeded Real-time System

On-line Scheduling Algorithm with Precedence Constraint in Embeded Real-time System 00 rd Internatonal Conference on Coputer and Electrcal Engneerng (ICCEE 00 IPCSIT vol (0 (0 IACSIT Press, Sngapore DOI: 077/IPCSIT0VNo80 On-lne Schedulng Algorth wth Precedence Constrant n Ebeded Real-te

More information

Cluster Analysis of Electrical Behavior

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

More information

Concurrent Apriori Data Mining Algorithms

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

More information

Verification by testing

Verification by testing Real-Tme Systems Specfcaton Implementaton System models Executon-tme analyss Verfcaton Verfcaton by testng Dad? How do they know how much weght a brdge can handle? They drve bgger and bgger trucks over

More information

Virtual Memory. Background. No. 10. Virtual Memory: concept. Logical Memory Space (review) Demand Paging(1) Virtual Memory

Virtual Memory. Background. No. 10. Virtual Memory: concept. Logical Memory Space (review) Demand Paging(1) Virtual Memory Background EECS. Operatng System Fundamentals No. Vrtual Memory Prof. Hu Jang Department of Electrcal Engneerng and Computer Scence, York Unversty Memory-management methods normally requres the entre process

More information

Real-time Fault-tolerant Scheduling Algorithm for Distributed Computing Systems

Real-time Fault-tolerant Scheduling Algorithm for Distributed Computing Systems Real-tme Fault-tolerant Schedulng Algorthm for Dstrbuted Computng Systems Yun Lng, Y Ouyang College of Computer Scence and Informaton Engneerng Zheang Gongshang Unversty Postal code: 310018 P.R.CHINA {ylng,

More information

A Novel Fault Tolerant Scheduling Technique In Real-Time Heterogeneous Distributed Systems Using Distributed Recovery Block

A Novel Fault Tolerant Scheduling Technique In Real-Time Heterogeneous Distributed Systems Using Distributed Recovery Block Proceedngs of Natonal Conference VISION 07 on Hgh Performance Computng 2 nd Aprl 07 Government College of Engneerng, Department of CSE, Trunelvel, Taml Nadu, INDIA A Novel Fault Tolerant Schedulng Technque

More information

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

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

More information

Assignment # 2. Farrukh Jabeen Algorithms 510 Assignment #2 Due Date: June 15, 2009.

Assignment # 2. Farrukh Jabeen Algorithms 510 Assignment #2 Due Date: June 15, 2009. Farrukh Jabeen Algorthms 51 Assgnment #2 Due Date: June 15, 29. Assgnment # 2 Chapter 3 Dscrete Fourer Transforms Implement the FFT for the DFT. Descrbed n sectons 3.1 and 3.2. Delverables: 1. Concse descrpton

More information

Learning the Kernel Parameters in Kernel Minimum Distance Classifier

Learning the Kernel Parameters in Kernel Minimum Distance Classifier Learnng the Kernel Parameters n Kernel Mnmum Dstance Classfer Daoqang Zhang 1,, Songcan Chen and Zh-Hua Zhou 1* 1 Natonal Laboratory for Novel Software Technology Nanjng Unversty, Nanjng 193, Chna Department

More information

A Resources Virtualization Approach Supporting Uniform Access to Heterogeneous Grid Resources 1

A Resources Virtualization Approach Supporting Uniform Access to Heterogeneous Grid Resources 1 A Resources Vrtualzaton Approach Supportng Unform Access to Heterogeneous Grd Resources 1 Cunhao Fang 1, Yaoxue Zhang 2, Song Cao 3 1 Tsnghua Natonal Labatory of Inforamaton Scence and Technology 2 Department

More information

Efficient Broadcast Disks Program Construction in Asymmetric Communication Environments

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

More information

Real-Time Systems. Real-Time Systems. Verification by testing. Verification by testing

Real-Time Systems. Real-Time Systems. Verification by testing. Verification by testing EDA222/DIT161 Real-Tme Systems, Chalmers/GU, 2014/2015 Lecture #8 Real-Tme Systems Real-Tme Systems Lecture #8 Specfcaton Professor Jan Jonsson Implementaton System models Executon-tme analyss Department

More information

Scheduling Remote Access to Scientific Instruments in Cyberinfrastructure for Education and Research

Scheduling Remote Access to Scientific Instruments in Cyberinfrastructure for Education and Research Schedulng Remote Access to Scentfc Instruments n Cybernfrastructure for Educaton and Research Je Yn 1, Junwe Cao 2,3,*, Yuexuan Wang 4, Lanchen Lu 1,3 and Cheng Wu 1,3 1 Natonal CIMS Engneerng and Research

More information

An Investigation into Server Parameter Selection for Hierarchical Fixed Priority Pre-emptive Systems

An Investigation into Server Parameter Selection for Hierarchical Fixed Priority Pre-emptive Systems An Investgaton nto Server Parameter Selecton for Herarchcal Fxed Prorty Pre-emptve Systems R.I. Davs and A. Burns Real-Tme Systems Research Group, Department of omputer Scence, Unversty of York, YO10 5DD,

More information

Assembler. Building a Modern Computer From First Principles.

Assembler. Building a Modern Computer From First Principles. Assembler Buldng a Modern Computer From Frst Prncples www.nand2tetrs.org Elements of Computng Systems, Nsan & Schocken, MIT Press, www.nand2tetrs.org, Chapter 6: Assembler slde Where we are at: Human Thought

More information

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration Improvement of Spatal Resoluton Usng BlockMatchng Based Moton Estmaton and Frame Integraton Danya Suga and Takayuk Hamamoto Graduate School of Engneerng, Tokyo Unversty of Scence, 6-3-1, Nuku, Katsuska-ku,

More information

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data A Fast Content-Based Multmeda Retreval Technque Usng Compressed Data Borko Furht and Pornvt Saksobhavvat NSF Multmeda Laboratory Florda Atlantc Unversty, Boca Raton, Florda 3343 ABSTRACT In ths paper,

More information

TPL-Aware Displacement-driven Detailed Placement Refinement with Coloring Constraints

TPL-Aware Displacement-driven Detailed Placement Refinement with Coloring Constraints TPL-ware Dsplacement-drven Detaled Placement Refnement wth Colorng Constrants Tao Ln Iowa State Unversty tln@astate.edu Chrs Chu Iowa State Unversty cnchu@astate.edu BSTRCT To mnmze the effect of process

More information

Priority-Based Scheduling Algorithm for Downlink Traffics in IEEE Networks

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

More information

A comparison of MPCP and MSRP when sharing resources in the Janus multiple-processor on a chip platform

A comparison of MPCP and MSRP when sharing resources in the Janus multiple-processor on a chip platform A comparson of MPCP and MSRP when sharng resources n the Janus multple-processor on a chp platform Paolo Ga, Marco D Natale, Guseppe Lpar, Scuola Superore Sant Anna, Psa, Italy {pj,marco,lpar}@sssup.t

More information

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching A Fast Vsual Trackng Algorthm Based on Crcle Pxels Matchng Zhqang Hou hou_zhq@sohu.com Chongzhao Han czhan@mal.xjtu.edu.cn Ln Zheng Abstract: A fast vsual trackng algorthm based on crcle pxels matchng

More information

Chapter 1. Introduction

Chapter 1. Introduction Chapter 1 Introducton 1.1 Parallel Processng There s a contnual demand for greater computatonal speed from a computer system than s currently possble (.e. sequental systems). Areas need great computatonal

More information

CE 221 Data Structures and Algorithms

CE 221 Data Structures and Algorithms CE 1 ata Structures and Algorthms Chapter 4: Trees BST Text: Read Wess, 4.3 Izmr Unversty of Economcs 1 The Search Tree AT Bnary Search Trees An mportant applcaton of bnary trees s n searchng. Let us assume

More information

Solving two-person zero-sum game by Matlab

Solving two-person zero-sum game by Matlab Appled Mechancs and Materals Onlne: 2011-02-02 ISSN: 1662-7482, Vols. 50-51, pp 262-265 do:10.4028/www.scentfc.net/amm.50-51.262 2011 Trans Tech Publcatons, Swtzerland Solvng two-person zero-sum game by

More information

The Codesign Challenge

The Codesign Challenge ECE 4530 Codesgn Challenge Fall 2007 Hardware/Software Codesgn The Codesgn Challenge Objectves In the codesgn challenge, your task s to accelerate a gven software reference mplementaton as fast as possble.

More information

A Frame Packing Mechanism Using PDO Communication Service within CANopen

A Frame Packing Mechanism Using PDO Communication Service within CANopen 28 A Frame Packng Mechansm Usng PDO Communcaton Servce wthn CANopen Mnkoo Kang and Kejn Park Dvson of Industral & Informaton Systems Engneerng, Ajou Unversty, Suwon, Gyeongg-do, South Korea Summary The

More information

A Model Based on Multi-agent for Dynamic Bandwidth Allocation in Networks Guang LU, Jian-Wen QI

A Model Based on Multi-agent for Dynamic Bandwidth Allocation in Networks Guang LU, Jian-Wen QI 216 Jont Internatonal Conference on Artfcal Intellgence and Computer Engneerng (AICE 216) and Internatonal Conference on etwork and Communcaton Securty (CS 216) ISB: 978-1-6595-362-5 A Model Based on Mult-agent

More information

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

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

More information

Available online at Available online at Advanced in Control Engineering and Information Science

Available online at   Available online at   Advanced in Control Engineering and Information Science Avalable onlne at wwwscencedrectcom Avalable onlne at wwwscencedrectcom Proceda Proceda Engneerng Engneerng 00 (2011) 15000 000 (2011) 1642 1646 Proceda Engneerng wwwelsevercom/locate/proceda Advanced

More information

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

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

More information

Efficient Distributed File System (EDFS)

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

More information

Simulation Based Analysis of FAST TCP using OMNET++

Simulation Based Analysis of FAST TCP using OMNET++ Smulaton Based Analyss of FAST TCP usng OMNET++ Umar ul Hassan 04030038@lums.edu.pk Md Term Report CS678 Topcs n Internet Research Sprng, 2006 Introducton Internet traffc s doublng roughly every 3 months

More information

Distributed Resource Scheduling in Grid Computing Using Fuzzy Approach

Distributed Resource Scheduling in Grid Computing Using Fuzzy Approach Dstrbuted Resource Schedulng n Grd Computng Usng Fuzzy Approach Shahram Amn, Mohammad Ahmad Computer Engneerng Department Islamc Azad Unversty branch Mahallat, Iran Islamc Azad Unversty branch khomen,

More information

FAHP and Modified GRA Based Network Selection in Heterogeneous Wireless Networks

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

More information

A New Transaction Processing Model Based on Optimistic Concurrency Control

A New Transaction Processing Model Based on Optimistic Concurrency Control A New Transacton Processng Model Based on Optmstc Concurrency Control Wang Pedong,Duan Xpng,Jr. Abstract-- In ths paper, to support moblty and dsconnecton of moble clents effectvely n moble computng envronment,

More information

Wishing you all a Total Quality New Year!

Wishing you all a Total Quality New Year! Total Qualty Management and Sx Sgma Post Graduate Program 214-15 Sesson 4 Vnay Kumar Kalakband Assstant Professor Operatons & Systems Area 1 Wshng you all a Total Qualty New Year! Hope you acheve Sx sgma

More information

Petri Net Based Software Dependability Engineering

Petri Net Based Software Dependability Engineering Proc. RELECTRONIC 95, Budapest, pp. 181-186; October 1995 Petr Net Based Software Dependablty Engneerng Monka Hener Brandenburg Unversty of Technology Cottbus Computer Scence Insttute Postbox 101344 D-03013

More information

Chapter 6 Programmng the fnte element method Inow turn to the man subject of ths book: The mplementaton of the fnte element algorthm n computer programs. In order to make my dscusson as straghtforward

More information

Ontology Generator from Relational Database Based on Jena

Ontology Generator from Relational Database Based on Jena Computer and Informaton Scence Vol. 3, No. 2; May 2010 Ontology Generator from Relatonal Database Based on Jena Shufeng Zhou (Correspondng author) College of Mathematcs Scence, Laocheng Unversty No.34

More information

A Generic and Compositional Framework for Multicore Response Time Analysis

A Generic and Compositional Framework for Multicore Response Time Analysis A Generc and Compostonal Framework for Multcore Response Tme Analyss Sebastan Altmeyer Unversty of Luxembourg Unversty of Amsterdam Clare Maza Grenoble INP Vermag Robert I. Davs Unversty of York INRIA,

More information

Helsinki University Of Technology, Systems Analysis Laboratory Mat Independent research projects in applied mathematics (3 cr)

Helsinki University Of Technology, Systems Analysis Laboratory Mat Independent research projects in applied mathematics (3 cr) Helsnk Unversty Of Technology, Systems Analyss Laboratory Mat-2.08 Independent research projects n appled mathematcs (3 cr) "! #$&% Antt Laukkanen 506 R ajlaukka@cc.hut.f 2 Introducton...3 2 Multattrbute

More information

MODULE DESIGN BASED ON INTERFACE INTEGRATION TO MAXIMIZE PRODUCT VARIETY AND MINIMIZE FAMILY COST

MODULE DESIGN BASED ON INTERFACE INTEGRATION TO MAXIMIZE PRODUCT VARIETY AND MINIMIZE FAMILY COST INTERNATIONAL CONFERENCE ON ENGINEERING DESIGN, ICED 07 28-31 AUGUST 2007, CITE DES SCIENCES ET DE L'INDUSTRIE, PARIS, FRANCE MODULE DESIGN BASED ON INTERFACE INTEGRATION TO MAIMIZE PRODUCT VARIETY AND

More information

Space-Optimal, Wait-Free Real-Time Synchronization

Space-Optimal, Wait-Free Real-Time Synchronization 1 Space-Optmal, Wat-Free Real-Tme Synchronzaton Hyeonjoong Cho, Bnoy Ravndran ECE Dept., Vrgna Tech Blacksburg, VA 24061, USA {hjcho,bnoy}@vt.edu E. Douglas Jensen The MITRE Corporaton Bedford, MA 01730,

More information

Load Balancing for Hex-Cell Interconnection Network

Load Balancing for Hex-Cell Interconnection Network Int. J. Communcatons, Network and System Scences,,, - Publshed Onlne Aprl n ScRes. http://www.scrp.org/journal/jcns http://dx.do.org/./jcns.. Load Balancng for Hex-Cell Interconnecton Network Saher Manaseer,

More information

A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS

A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS Proceedngs of the Wnter Smulaton Conference M E Kuhl, N M Steger, F B Armstrong, and J A Jones, eds A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS Mark W Brantley Chun-Hung

More information

Distributed Middlebox Placement Based on Potential Game

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

More information

Performance Study of Parallel Programming on Cloud Computing Environments Using MapReduce

Performance Study of Parallel Programming on Cloud Computing Environments Using MapReduce Performance Study of Parallel Programmng on Cloud Computng Envronments Usng MapReduce Wen-Chung Shh, Shan-Shyong Tseng Department of Informaton Scence and Applcatons Asa Unversty Tachung, 41354, Tawan

More information

UB at GeoCLEF Department of Geography Abstract

UB at GeoCLEF Department of Geography   Abstract UB at GeoCLEF 2006 Mguel E. Ruz (1), Stuart Shapro (2), June Abbas (1), Slva B. Southwck (1) and Davd Mark (3) State Unversty of New York at Buffalo (1) Department of Lbrary and Informaton Studes (2) Department

More information

Hermite Splines in Lie Groups as Products of Geodesics

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

More information

TN348: Openlab Module - Colocalization

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

More information

Parallel matrix-vector multiplication

Parallel matrix-vector multiplication Appendx A Parallel matrx-vector multplcaton The reduced transton matrx of the three-dmensonal cage model for gel electrophoress, descrbed n secton 3.2, becomes excessvely large for polymer lengths more

More information

The Greedy Method. Outline and Reading. Change Money Problem. Greedy Algorithms. Applications of the Greedy Strategy. The Greedy Method Technique

The Greedy Method. Outline and Reading. Change Money Problem. Greedy Algorithms. Applications of the Greedy Strategy. The Greedy Method Technique //00 :0 AM Outlne and Readng The Greedy Method The Greedy Method Technque (secton.) Fractonal Knapsack Problem (secton..) Task Schedulng (secton..) Mnmum Spannng Trees (secton.) Change Money Problem Greedy

More information

An Online Delay Efficient Multi-Class Packet Scheduler for Heterogeneous M2M Uplink Traffic

An Online Delay Efficient Multi-Class Packet Scheduler for Heterogeneous M2M Uplink Traffic An Onlne Delay Effcent Mult-Class Packet Scheduler for Heterogeneous M2M Uplnk Traffc Akshay Kumar, Ahmed Abdelhad and Charles Clancy Hume Center, Vrgna Tech Emal:{akshay2, aabdelhad, tcc}@vt.edu Abstract

More information

The Shortest Path of Touring Lines given in the Plane

The Shortest Path of Touring Lines given in the Plane Send Orders for Reprnts to reprnts@benthamscence.ae 262 The Open Cybernetcs & Systemcs Journal, 2015, 9, 262-267 The Shortest Path of Tourng Lnes gven n the Plane Open Access Ljuan Wang 1,2, Dandan He

More information

Mixed-Criticality Scheduling on Multiprocessors using Task Grouping

Mixed-Criticality Scheduling on Multiprocessors using Task Grouping Mxed-Crtcalty Schedulng on Multprocessors usng Task Groupng Jankang Ren Lnh Th Xuan Phan School of Software Technology, Dalan Unversty of Technology, Chna Computer and Informaton Scence Department, Unversty

More information

Solution Brief: Creating a Secure Base in a Virtual World

Solution Brief: Creating a Secure Base in a Virtual World Soluton Bref: Creatng a Secure Base n a Vrtual World Soluton Bref: Creatng a Secure Base n a Vrtual World Abstract The adopton rate of Vrtual Machnes has exploded at most organzatons, drven by the mproved

More information

Machine Learning: Algorithms and Applications

Machine Learning: Algorithms and Applications 14/05/1 Machne Learnng: Algorthms and Applcatons Florano Zn Free Unversty of Bozen-Bolzano Faculty of Computer Scence Academc Year 011-01 Lecture 10: 14 May 01 Unsupervsed Learnng cont Sldes courtesy of

More information

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

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

More information

Run-Time Operator State Spilling for Memory Intensive Long-Running Queries

Run-Time Operator State Spilling for Memory Intensive Long-Running Queries Run-Tme Operator State Spllng for Memory Intensve Long-Runnng Queres Bn Lu, Yal Zhu, and lke A. Rundenstener epartment of Computer Scence, Worcester Polytechnc Insttute Worcester, Massachusetts, USA {bnlu,

More information

TECHNIQUE OF FORMATION HOMOGENEOUS SAMPLE SAME OBJECTS. Muradaliyev A.Z.

TECHNIQUE OF FORMATION HOMOGENEOUS SAMPLE SAME OBJECTS. Muradaliyev A.Z. TECHNIQUE OF FORMATION HOMOGENEOUS SAMPLE SAME OBJECTS Muradalyev AZ Azerbajan Scentfc-Research and Desgn-Prospectng Insttute of Energetc AZ1012, Ave HZardab-94 E-mal:aydn_murad@yahoocom Importance of

More information

Needed Information to do Allocation

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

More information

Memory Modeling in ESL-RTL Equivalence Checking

Memory Modeling in ESL-RTL Equivalence Checking 11.4 Memory Modelng n ESL-RTL Equvalence Checkng Alfred Koelbl 2025 NW Cornelus Pass Rd. Hllsboro, OR 97124 koelbl@synopsys.com Jerry R. Burch 2025 NW Cornelus Pass Rd. Hllsboro, OR 97124 burch@synopsys.com

More information

Two-Stage Data Distribution for Distributed Surveillance Video Processing with Hybrid Storage Architecture

Two-Stage Data Distribution for Distributed Surveillance Video Processing with Hybrid Storage Architecture Two-Stage Data Dstrbuton for Dstrbuted Survellance Vdeo Processng wth Hybrd Storage Archtecture Yangyang Gao, Hatao Zhang, Bngchang Tang, Yanpe Zhu, Huadong Ma Bejng Key Lab of Intellgent Telecomm. Software

More information

Remote Sensing Image Retrieval Algorithm based on MapReduce and Characteristic Information

Remote Sensing Image Retrieval Algorithm based on MapReduce and Characteristic Information Remote Sensng Image Retreval Algorthm based on MapReduce and Characterstc Informaton Zhang Meng 1, 1 Computer School, Wuhan Unversty Hube, Wuhan430097 Informaton Center, Wuhan Unversty Hube, Wuhan430097

More information

An Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation

An Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation 17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 An Iteratve Soluton Approach to Process Plant Layout usng Mxed

More information

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS ARPN Journal of Engneerng and Appled Scences 006-017 Asan Research Publshng Network (ARPN). All rghts reserved. NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS Igor Grgoryev, Svetlana

More information

Application of VCG in Replica Placement Strategy of Cloud Storage

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

More information

Assembler. Shimon Schocken. Spring Elements of Computing Systems 1 Assembler (Ch. 6) Compiler. abstract interface.

Assembler. Shimon Schocken. Spring Elements of Computing Systems 1 Assembler (Ch. 6) Compiler. abstract interface. IDC Herzlya Shmon Schocken Assembler Shmon Schocken Sprng 2005 Elements of Computng Systems 1 Assembler (Ch. 6) Where we are at: Human Thought Abstract desgn Chapters 9, 12 abstract nterface H.L. Language

More information

Comparison of Heuristics for Scheduling Independent Tasks on Heterogeneous Distributed Environments

Comparison of Heuristics for Scheduling Independent Tasks on Heterogeneous Distributed Environments Comparson of Heurstcs for Schedulng Independent Tasks on Heterogeneous Dstrbuted Envronments Hesam Izakan¹, Ath Abraham², Senor Member, IEEE, Václav Snášel³ ¹ Islamc Azad Unversty, Ramsar Branch, Ramsar,

More information

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

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

More information

Array transposition in CUDA shared memory

Array transposition in CUDA shared memory Array transposton n CUDA shared memory Mke Gles February 19, 2014 Abstract Ths short note s nspred by some code wrtten by Jeremy Appleyard for the transposton of data through shared memory. I had some

More information

5 The Primal-Dual Method

5 The Primal-Dual Method 5 The Prmal-Dual Method Orgnally desgned as a method for solvng lnear programs, where t reduces weghted optmzaton problems to smpler combnatoral ones, the prmal-dual method (PDM) has receved much attenton

More information

A NOTE ON FUZZY CLOSURE OF A FUZZY SET

A NOTE ON FUZZY CLOSURE OF A FUZZY SET (JPMNT) Journal of Process Management New Technologes, Internatonal A NOTE ON FUZZY CLOSURE OF A FUZZY SET Bhmraj Basumatary Department of Mathematcal Scences, Bodoland Unversty, Kokrajhar, Assam, Inda,

More information

Intro. Iterators. 1. Access

Intro. Iterators. 1. Access Intro Ths mornng I d lke to talk a lttle bt about s and s. We wll start out wth smlartes and dfferences, then we wll see how to draw them n envronment dagrams, and we wll fnsh wth some examples. Happy

More information

Reducing Frame Rate for Object Tracking

Reducing Frame Rate for Object Tracking Reducng Frame Rate for Object Trackng Pavel Korshunov 1 and We Tsang Oo 2 1 Natonal Unversty of Sngapore, Sngapore 11977, pavelkor@comp.nus.edu.sg 2 Natonal Unversty of Sngapore, Sngapore 11977, oowt@comp.nus.edu.sg

More information