Maximum Weight Matching Dispatching Scheme in Buffered Clos-Network Packet Switches

Size: px
Start display at page:

Download "Maximum Weight Matching Dispatching Scheme in Buffered Clos-Network Packet Switches"

Transcription

1 Maxmum Weght Matchng Dspatchng Scheme n Buffered Clos-Network Packet Swtches Roberto Roas-Cessa, Member, IEEE, E Ok, Member, IEEE, and H. Jonathan Chao, Fellow, IEEE Abstract The scalablty of Clos-network swtches make them an alternatve to sngle-stages swtches for mplementng largesze packet swtches. Ths paper ntroduces a cell dspatchng scheme, called maxmum weght matchng dspatchng (MWMD) scheme, for buffered Clos-network swtches. The MWMD scheme s based on a maxmum weght matchng algorthm for nputbuffered swtches. Ths paper shows that, wth request queues n the buffered Clos-network archtecture, the MWMD scheme s able to acheve a 00% throughput for ndependent admssble traffc, wthout allocatng any buffers n the second stage and wthout expandng the nternal bandwdth. As a practcal scheme, a maxmal oldest-cell-frst matchng dspatchng (MOMD) scheme s also ntroduced. MOMD shows that usng a fnte number of teratons n the dspatchng scheme, the throughout under unbalanced traffc pattern can be hgh. Index Terms Packet swtch, clos-network, dspatchng, throughput, maxmum-weght matchng I. INTRODUCTION It s well known that sngle-stage swtches have lmted scalablty, n terms of the number of ports. Ths lmted scalablty s the result of the lmted number of connecton pns that a swtch chp can allocate. The three-stage Clos-network [], [2] swtch can provde a hgher degree of scalablty than snglestage swtches, because of the smaller number of swtch chps needed for a large sze swtch. We can categorze the Clos-network swtch archtecture nto two types: bufferless and buffered. A bufferless Clos-network swtch has no memory n any stage. To avod contenton n any stage, schedulng needs to be performed at the nput ports pror to sendng the packets through the swtch. Ths approach has the advantage of smplfyng the desgn of the swtch modules. However, the matchng process may be complex and requre a long resoluton tme. Schedulng of bufferless Clos-network swtches can also be used n optcal swtches where swtch reconfguraton may not have to change as fast as per tme slot. There are several studes on schedulng for bufferless Clos-network swtches [7], [8], [9]. They, however, are out of the scope of ths paper. Here, we assume that varable-length packets are segmented nto several fxed-szed packets, or cells, when they arrve. Cells are swtched through the swtch fabrc, and reassembled nto packets before they depart. The tme to transmt a cell through the swtch s called tme slot. Roberto Roas-Cessa s wth the Department of Electrcal and Computer Engneerng, New Jersey Insttute of Technology, Unversty Heghts, Newark NJ, Emal: rroas@nt.edu E Ok s wth NTT Network Innovaton Laboratores, NTT Corporaton, 3-9- Mdor-cho Musashno-sh, Tokyo Japan. E-mal: ok.e@lab.ntt.co.p H. Jonathan Chao s wth the Department of Electrcal and Computer Engneerng, Polytechnc Unversty, Brooklyn, NY, 20. Emal: chao@poly.edu One way to ease the complexty of schedulng n Closnetwork swtches s by allocatng memory n the frst and thrd stages. In ths way, f contenton for an nternal lnk occurs, loser cells are stored n the buffers n the frst stage modules. These swtches can be referred to as buffered Clos-network swtches. As the memory technology evolves, the memory amount that can be embedded nto a chp s no longer a strct lmtaton. Wthn buffered Clos-network swtches, we can consder two groups: wth and wthout buffers n the second-stage modules. A ggabt ATM (asynchronous transfer mode) swtch usng buffers n the second-stage was presented n [3]. In ths archtecture, every cell s randomly dstrbuted from the frststage to the second-stage modules to balance the traffc load n the second-stage. Implementng buffers n the second-stage modules resolves contenton among cells from dfferent frststage modules [5]. However, t requres a re-sequencng functon at the thrd-stage modules, because the buffers n the second-stage modules cause an out-of-sequence problem. A three-stage swtch wth buffers n the frst and thrd stages and bufferless second stage s called a buffered Closnetwork swtch. In [4], an ATM swtch was developed. Ths approach does not suffer from the out-of-sequence problem. Snce there are no buffers n the second-stage modules to resolve contenton, dspatchng cells from the frst stage to the second stage becomes an mportant ssue. A random dspatchng (RD) scheme s used for cell dspatchng from the frst stage to the second stage [4], as adopted n the case of the buffered second-stage modules n [3]. However, RD s not able to acheve a hgh throughput unless the nternal bandwdth s expanded, because the contenton at the second stage cannot be avoded. To acheve 00% throughput for unform traffc by usng RD, the nternal expanson rato s set to about.6 when the swtch sze s large [2], [4]. Ths expanson makes a hgh-speed swtch dffcult to mplement n a cost-effectve manner. It has been shown that t s possble to provde 00% throughput under unform traffc wthout expandng the nternal bandwdth on a buffered Clos-network swtch wth a round-robn-based dspatchng scheme [2]. Moreover, dverse dspatchng schemes have been proposed to reduce the average cell delay [5], [] under unform traffc. However, real traffc patterns are not only unform, but a wde varety of admssble traffc patterns, ncludng those wth nonunform dstrbutons. One queston arses: Is t possble to acheve a 00% throughput under ndependent admssble traffc, wthout allocatng any buffers n the second stage to avod the out-ofsequence problem and wthout expandng the nternal bandwdth? Ths paper proposes a cell dspatchng scheme, called IEEE Communcatons Socety 075

2 maxmum weght matchng dspatchng (MWMD) scheme, for buffered Clos-network swtches. The MWMD scheme s based on the maxmum weght matchng algorthm for nput-buffered swtches [6]. It s known that for ndependent admssble traffc, a maxmum throughput of 00% n an nput-buffered swtch s achevable by usng a maxmum weght matchng algorthm [6]. We show that by the consderng request queues n a buffered Clos-network swtch and the MWMD scheme t s possble to acheve a 00% throughput for ndependent admssble traffc, wthout allocatng any buffers n the swtch modules at the second stage and wthout expandng the nternal bandwdth. Furthermore, we ntroduce an teratve dspatchng scheme, maxmal oldest cell frst matchng dspatchng (MOMD), based on the oldest-cell-frst (OCF) scheme [3] for sngle-stage nput-buffered swtches, and show that t can provde 00% throughput under our nonunform traffc model, called unbalanced. Ths paper s organzed as follows. Secton II descrbes our swtch model. Secton III ntroduces the MWMD scheme. Secton IV ntroduces the MOMD scheme. Secton V shows the performance study of MOMD. Secton VI presents the conclusons. II. BUFFERED CLOS-NETWORK SWITCH MODEL Fgure shows a buffered Clos-network swtch. The frst stage conssts of k nput modules (IMs), each of whch has an n m dmenson. The second stage conssts of m buffer-less central modules (CMs) each of whch has a k k dmenson. The thrd stage conssts of k output modules (OMs), each of whch has an m n dmenson. The termnology used n ths paper s as follows: : ( +)th nput module, where 0 k. CM(r): (r +)th central module, where 0 r m. OM(): ( +)th output module, where 0 k. n: number of nput/output ports n each IM/OM, respectvely. k: number of IMs/OMs. m: number of CMs. IP(, h): (h +)th nput port (IP) at, where 0 h n. OP(, l): (l +)th output port (OP) at OM(), where 0 l n. VOMQ(, ): vrtual output-module queue at that stores cells destned for OM(). RQ(,, r): Request queue (RQ) at that stores requests of cells destned for OM() through CM(r). RQ(,, r) also keeps the watng tme W (,, r), whch s the number of slots a head-of-lne (HOL) request has been watng. L I (, r): output lnk of that s connected to CM(r). L C (r, ): output lnk at CM(r) that s connected to OM(). λ(, h,, l): arrval rate at IP(, h) for OP(, l). λ(, ): arrval rate at for OM() = h l λ(, h,, l). An has k vrtual output-module queues (VOMQs) to elmnate HOL blockng. A VOMQ s smlar to a vrtual output queue (VOQ), whch s used n nput-buffered swtches [3], [6]. When a cell enters VOMQ(, ), the cell request s randomly dstrbuted and stored n RQ(,, r) among m request queues. A request n RQ(,, r) s related to VOMQ(, ), but s not related to a specfc cell n VOMQ(, ). Every tme a request n RQ(,, r) s granted for transmsson, one cell s dequeued from VOMQ(, ) n a FIFO manner. A VOMQ can receve at most n cells from n nput ports and can send at most m cells to m CMs n one tme slot. Each has m output lnks. Each output lnk L I (, r) s connected to each CM(r). IP(, h) IP(0, 0) IP(0, n-) VOMQ(0, k-) IP(, 0) IP(, n-) IP(k-, 0) VOMQ(0, 0) VOMQ(, 0) VOMQ(, k-) VOMQ(k-, 0) IP(k-, n-) VOMQ(k-, k-) RQ(0, k-, m-) RQ(, 0, 0) RQ(, k-, 0) RQ(, 0, m-) RQ(, k-, m-) RQ(k-, 0, 0) RQ(k-, k-, 0) RQ(k-, k-, m-) CM(0) CM(r) CM(m-) OM() OP(0, 0) OP(, l) OP(0, n-) OP(, 0) OP(, n-) OP(k-, 0) OP(k-, n-) Fg.. Clos-network swtch wth vrtual output-module queues (VOMQs) n the nput modules A CM(r) has k output lnks, L C (r, ), connected to OM(). AnOM() has n output ports, each of whch s denoted as OP(, l) and has an output buffer. Each output buffer receves at most m cells n one cell tme slot, and each output port at the OM forwards one cell n a frst-n-frst-out (FIFO) manner to the output lne. Scalablty, n terms of the port speeds, has also been of research nterest. A sngle-stage parallel packet swtch resembles the swtch fabrc of a Clos-network swtch (.e., the set of IMs, CMs, and OMs) because of the way the nternal lnks are connected n the latter. Parallel swtches consder usng swtch planes runnng at lower speed than the external connecton lnks to provde hgh-speed ports [4]. We consder that Clos-network swtches use nternal lnks transferrng data at the same speed as the external lnks. Moreover, followng the prncple of the parallel swtch, Clos-networks swtches can also use parallel swtchng planes to provde hgh port speeds. However, the dscusson about ncreasng the port speed n a Clos-network swtch s out of the scope of ths paper. III. MAXIMUM WEIGHT MATCHING DISPATCHING (MWMD) SCHEME The MWMD scheduler conssts of m subschedulers, each of whch s denoted as S(r), as shown n Fgure 2. Subscheduler S(r) selects up to k requests from k 2 RQs, where correspondng cells to the selected RQs are transmtted through CM(r) at the next tme slot. In S(r), k RQs: RQ(, 0,r),..., RQ(, k,r), are the requests from to all OMs An L-bt cell must be wrtten to and read from a VOMQ memory n a tme less than L Cn and L, respectvely, where C s a lne speed. For example, Cm L when L =64 8 bts, C=0 Gbt/s, and n = m =8, Cn = L s 6.4 Cm ns. Ths s feasble when we consder 2-port memores wth current avalable CMOS technologes. IEEE Communcatons Socety 076

3 Fg. 2. Subscheduler S(0) RQ(, 0, 0) RQ(, k-, 0) RQ(k-, 0, 0) RQ(k-, k-, 0) Subcheduler S(m-) RQ(0, k-, m-) RQ(, 0, m-) RQ(, k-, m-) RQ(k-, k-, m-) OM() OM() Maxmum Weght Matchng Dspatchng (MWMD) Scheduler through CM(r), and k RQs: RQ(0,,r),..., RQ(k,,r), are the requests from all IMs to OM() through CM(r). S(r) selects one request from each and one request to each OM(). S(r) fnds an approprate match accordng to ts schedulng algorthm as a scheduler of an nput-buffered swtch does. In the request selecton n S(r) at each tme slot, a maxmum weght matchng algorthm, the OCF algorthm [6], s employed. RQ(,, r) keeps the watng tme W (,, r), whch s the number of slots a HOL request has been watng. S(r) fnds a match M(r) at each tme slot, so that (,) M(r) W (,, r) s maxmzed. Note that each S(r) behaves ndependently and concurrently. S(r) uses only k 2 W (,, r) to fnd M(r). S(r) and S(r ), where r < r, do not exchange any nformaton to fnd M(r) and M(r ), respectvely. When RQ(,, r) s granted by S(r), the HOL request n RQ(,, r) and a cell n VOMQ(, ) are dequeued. The dequeued cell s one of the HOL cells n VOMQ(, ). Here, the number of dequeued HOL cells s equal to the number of granted requests by all S(r)s. In order to send cells n sequence to an output buffer at OP(, l), we adopt a smple rule as follows. Consder that Y RQs for the same (, ) par, whch are RQ(,, r ), RQ(,, r 2 ),..., RQ(,, r Y ), where r < r 2 < r Y, are granted by more than one subscheduler. Y cells from the HOL n VOMQ(, ) are transmtted to OM() through CM(r ), CM(r 2 ),..., CM(r Y ) at the same tme slot. The yth cell from the HOL n VOMQ(, ) goes through CM(r y ).Note that each request n RQ(,, r) s related to VOMQ(, ), but s not related to a specfc cell n VOMQ(, ), as descrbed n Secton II. When more than one cell goes to an output buffer at OP(, l), a cell from CM(r) enters the output buffer earler than the cell from CM(r ), where r<r. As a result, cells are sent n sequence. A. 00% throughput by MWMD We prove that MWMD acheves 00% throughput for all admssble ndependent arrval processes wthout nternal bandwdth expanson,.e., wth n = m. Theorem : MWMD acheves 00% throughput for all admssble ndependent arrval processes wthout expanson of the nternal bandwdth. Proof: Wth ths archtecture, we use the rate matrx of the arrval process as: Λ [, h,, l], where the assocate rate vector s λ (λ 0,0,0,0,...,λ 0,0,0,n,...,λ 0,0,k,0,..., λ 0,n,0,0,...,λ k,0,0,0,...,λ k,n,k,n ). We assume that our admssble nput traffc condtons, λ(, h,, l) < and λ(, h,, l) <, () h are satsfed. The arrval process A (t), whch s the aggregated cell arrvals of n nput ports at that are destned to VOMQ(, ), s statonary and ergodc. The arrval matrx representng the sequence of arrvals A(t) [A, (t)], where, at tme t: g, g n f arrval(s) occurs at VOMQ, A, (t) = (2) and the assocated rate vector s A(t) (A 0,0 (t),...,a 0,k (t),...,a k,k (t)) T. The servce matrx, Γ(t) [Γ, (t)] ndcates whch VOMQs are servced at tme t: p, p n f VOMQ, gets servce at t Γ, (t) = (3) and the assocate vector Γ(t) s smlarly defned as A(t). However, Γ(t) may not be used as a permutaton matrx. To overcome ths, consder usng request queues, RQ(,, r), as descrbed n Secton III. As the MWMD scheme, performed by the rth subscheduler consders the RQs related the rth CM for dspatchng cells through CM(r), the set of RQs can be consdered ndependent of the others. They can be separated as a local set of RQs, denoted as LRQ(, ), for each CM(r). Each LRQ has a correspondng arrval request matrx (t) [, (t)]. The request arrval matrx s such that f an arrval to LRQ, occurs at tme t, (t) = (4) as RQs accept n average one request from nput to output each tme slot, and the RQ dstrbuton s ndependently and nformngly dstrbuted. The assocated vector, (t), s smlarly defned. The servce receved by LRQs s determned by the servce matrx Θ(t) Θ, (t), where: f LRQ, s served at tme t Θ, (t) = (5) and the assocated vector, Θ(t), s smlarly defned. Therefore, the LQRs and CMs are equvalent to the VOQs of an nputbuffered swtch, consderng (t) and Θ(t) as the arrval l IEEE Communcatons Socety 077

4 and servce processes. The servce receved by LRQs s the result of MWMD, where only the HOL requests at LRQs are matched. From (), we obtan that λ(, ) = λ(, h,, l) <n, and h l λ(, ) = λ(, h,, l) <n. (6) h l Here, = =n s used. By dvdng (6) by m, λ(, ) < n m m and λ(, ) < n m m. (7) By usng n = m, we obtan λ(, ) < and m m λ(, ) <. (8) Therefore, the total servce receved by VOMQ(, ) s r Θ,(t). If the receved servce by Θ, (t) for each CM and ts LRQ set s such that the RQs are stable, then MWMD s stable. Note that mλ(, ) s the request arrval rate for RQ(,, r). In addton, snce the cell arrval process at VOMQ(, ) s ndependent, a request arrval process to RQ(,, r) s also ndependent. Ths s because, when a cell enters VOMQ(, ), ts request s randomly dstrbuted and stored among m request queues. As MWMD s based on the maxmum-weght matchng scheme, Lemma 7 n [6] can be appled. Therefore, the subscheduler model n MWMD s equvalent to the scheduler model n the maxmum weght matchng (MWM) of the nput-buffered swtch model defned n [6]. Usng Lemmas 8-0 and Theorem 4 n [6], where the OCF-based MWM algorthm has the queue occupancy stable, achevng 00% throughput n an nput-buffered swtch for all admssble and ndependent arrval processes, a subscheduler n MWMD provdes 00% throughput. As all subschedulers have equvalent performance, MWMD provdes 00% throughput for all admssble and ndependent arrval process wth n = m. IV. MAXIMAL OLDEST CELL FIRST MATCHING DISPATCHING (MOMD) In ths secton, we ntroduce the maxmal oldest-cell frst matchng dspatchng (MOMD) scheme wth multple memory access (.e., queues can dspatch up to m cells) as a maxmal dspatchng scheme wth lower complexty than MWMD. MOMD uses the swtch model presented n Secton II. There are k VOMQs at each IM, each denoted as VOMQ(, ), and m request queues, RQs, each assocated wth a VOMQ, and denoted as RQ(,, r). MOMD has dstrbuted arbters n IM and CM. In, there are m output-lnk arbters. In CM(r), there are k arbters, each of whch corresponds to OM(). The VOMQs and dstrbuted lnk arbters are shown n Fgure 3. When a cell enters a VOMQ(, ), the cell gets a tme stamp assgned, whch s entered along wth a request to RQ(,, r), where r s randomly selected. We consder lnk arbters for the output lnks at IMs and CMs. To determne the matchng between VOMQ(, ) and the output lnk L I (, r), IP(, h) IP(0, 0) IP(0, n-) VOMQ(0, k-) IP(, 0) IP(, n-) IP(k-, 0) IP(k-, n-) VOMQ(k-, k-) Fg. 3. VOMQ(0, 0) VOMQ(, 0) VOMQ(, k-) VOMQ(k-, 0) RQ(0, k-,m-) RQ(, 0, 0) RQ(, k-, 0) RQ(, 0, m-) RQ(, k-, m-) RQ(k-, 0, 0) RQ(k-, k-, 0) RQ(k-, k-, m-) IM lnk arbter CM(0) CM(r) CM(m-) CM lnk arbter OM() Dstrbuted arbters n a buffered Clos-network wth MOMD OP(0, 0) OP(, l) OP(0, n-) OP(, 0) OP(, n-) OP(k-, 0) OP(k-, n-) the output lnks arbters select a VOMQ (through ther RQs) based on tme stamps. The scheme s descrbed as follows: Phase : Matchng wthn IM Step : A non-empty RQ(,, r) sends a request to the unmatched output lnk arbter assocated to L I (, r). The request ncludes a weght, whch s the tme stamp of the HOL request. Step 2: Each output-lnk arbter L(, ) chooses one RQ request by selectng the oldest tme stamp. If a te occurs, the output lnk arbter selects the RQ(,, r) by selectng the VOMQ wth the largest ndex. The output lnk arbter sends the grant to the selected RQ and VOMQ. Phase 2: Matchng between IM and CM Step : After phase s completed, L I (, r) sends the request to CM(r) belongng to the selected VOMQ. At CM(r), each arbter assocated wth OM() chooses one request by selectng the oldest tme stamp. Tes are broken by selectng, among the older requests, the largest ndex. The arbter at CM(r) sends the grant to L I (, r) of. Step 2: If an IM receves a grant from a CM, the IM sends a (HOL) cell from that VOMQ at the next tme slot. Note that up to m cells, at the head of lne, may be dspatched from a VOMQ. MOMD dspatches cells as n MWMD to get cells arrve the output buffer at OP(, l) n sequence. The descrbed scheme refers to a sngle teraton between IM and CM. To perform more teratons, the unmatched VOMQs, lnks, and CM arbters are consdered, and the reected requests are nhbted n the remanng teratons, performed between IM and CM, n the tme slot. V. PERFORMANCE EVALUATION OF MOMD We show the delay and throughput performance of swtches, where n = m = k =8, usng the RD [4], CRRD [2], and MOMD schemes. Note that MOMD does not use teratons n IM, denoted as IM, as n the CRRD scheme. MOMD uses request queues that smplfy the process, such that one lnk arbter selects one out of m request queues. However, we consder that a number of teratons between IM and CM n all schemes, denoted as IM-CM, can be IEEE Communcatons Socety 078

5 Average cell delay (cell slot) CRRD, IM-CM= RD MOMD, IM-CM= MOMD, IM-CM= Input load Fg. 4. MOMD under bernoull unform traffc n a (n = m = k =8) swtch used. 2 We note that MOMD cannot acheve a 00% throughput under unform traffc wth a sngle IM-CM teraton. However, MOMD gets hgher throughput as the number of IM- CM teratons ncreases, as the fgure shows. In the swtch under smulaton, the number of teratons to provde 00% throughput s four (IM-CM = 4). The smulaton shows that the round-robn based scheme, CRRD, s more effectve under unform traffc than MOMD, as CRRD acheves hgh throughput wth one teraton, IM-CM =. Throughput Unbalanced probablty, w RD MOMD, IM-CM= MOMD, IM-CM=2 MOMD, IM-CM=4 MOMD, IM-CM=8 CRRD, IM-CM = CRRD, IM-CM =8 Fg. 5. MOMD under bernoull unbalanced traffc and n = m = k =8 We also consdered a nonunform traffc pattern, called unbalanced [5]. The unbalanced traffc pattern has one fracton of traffc wth unform dstrbuton and the other fracton w of traffc destned to the output wth the same ndex number as the nput. When w =0, the traffc s unform; when w = the traffc s totally drectonal. We evaluate RD, CRRD, and MOMD under ths traffc pattern. To obtan the best performance, CRRD adopts IM = m (.e., the number of teratons performed n IM to match the VOQs used n CRRD and output lnks [5]) for any number IM-CM. Fgure 5 shows that RD, wth one teraton, and CRRD, wth multple IM-CM teratons, cannot offer 00% throughput under unbalanced traffc. Although CRRD wth IM-CM = 8 ncreases the throughput, ths s not close to 00%. The reason for that s that CRRD, based on round-robn selecton, tres to allocate the same servce for all queues, ndependently of the HOL cell age or queue load n each tme slot. MOMD, as t uses weghtbased selecton, provdes hgher throughput as the number of IM-CM teratons ncreases. MOMD wth IM-CM = offers low throughput. However, ts throughput approaches to 00% when IM-CM =8. Ths presents the challenge of performng a large number of teratons, therefore, we would need to consder the adopton of a ppelnng technque [0] for mplementaton of MOMD. These results also show that wth a fnte number of IM-CM teratons n ths swtch can provde hgh throughput under ths traffc pattern. VI. CONCLUSIONS Ths paper proposed the MWMD scheme for buffered Closnetwork swtches, whch s based on a maxmum weght matchng algorthm for nput-buffered swtches. Ths paper showed that the combnaton of MWMD and request queues n a buffered Clos-network swtch provdes 00% throughput under ndependent admssble traffc, wthout allocatng any buffers n the second stage and wthout expandng the nternal bandwdth. In addton, we ntroduced a maxmalweght matchng dspatchng scheme, MOMD, based on the OCF scheme for sngle stage swtches, whch can provde hgh throughput under a nonunform traffc pattern, called unbalanced. REFERENCES [] C. Clos, A Study of Non-Blockng Swtchng Networks, Bell Sys. Tech. Jour., pp , March 953. [2] E. Ok, Z. Jng, R. Roas-Cessa, and H. J. Chao, Concurrent Round- Robn Dspatchng Scheme n a Clos-Network Swtch, IEEE ICC 200, pp. 07-, June 200. [3] T. Chaney, J. A. Fngerhut, M. Flucke, and J. S. Turner, Desgn of a Ggabt ATM swtch, Proc. IEEE INFOCOM 97, pp. 2-, Apr [4] F. M. Chuss, J. G. Kneuer, and V. P. Kumar, Low-Cost Scalable Swtchng Solutons for Broadband Networkng: The ATLANTA Archtecture and Chpset, IEEE Commun. Mag., pp , Dec [5] E. Ok, Z. Jng, R. Roas-Cessa, and H. J. Chao, Concurrent Round-Robn-Based Dspatchng Schemes for Clos-Network Swtches, IEEE/ACM Trans. Networkng, vol. 0, no. 6, pp , Dec [6] N. McKeown, A. Mekkttkul, V. Anantharam, J. Walrand, Achevng 00% Throughput n an Input-queued Swtch, IEEE Trans. Commun., vol. 47, no. 8, pp , Aug [7] C. Y. Lee and A. Y. Qruc, A Fast Parallel Algorthm for Routng Uncast Assgnments n Benes Networks, IEEE Trans. Parallel Dstrbuted Sys., vol. 6, no. 3, pp , Mar [8] T. T. Lee and S-Y Lew, Parallel Routng Algorthm n Benes-Clos Networks, Proc. IEEE INFOCOM 96, pp , 996. [9] K. Pun, M. Hamd, Dstro: A Dstrbuted Statc Round-Robn Schedulng Algorthm for Bufferless Clos-Network Swtches, IEEE Globecom 2002, vol. 3, pp , [0] E. Ok, R. Roas-Cessa, H. Jonathan Chao, PCRRD: A Ppelne- Based Concurrent Round-Robn Dspatchng Scheme for Clos-Network Swtches, IEEE ICC 2002, vol.4, pp:22-225, [] P. Konghong, M. Hamd, Statc Round-robn Dspatchng Schemes for Clos-network Swtches, IEEE HPSR 2002, pp , May [2] T. Anderson, S. Owck, J. Saxe, and C. Thacker, Hgh speed swtch schedulng for local area networks, ACM Trans. Comput. Syst., vol., no. 4, pp , Nov [3] N. McKeown, Ph. D. Thess, Unversty of Calforna at Berkeley, May 995. [4] S. Iyer, N. McKeown, Analyss of the Parallel Packet Swtch Archtecture, IEEE/ACM Trans. Networkng, vol., No. 2, pp , Apr [5] J. Turner and N. Yamanaka, Archtectural Choces n Large Scale ATM Swtches, IEICE Trans. Commun. vol., E8-B, no. 2, pp , Feb Ths s an extended verson of CRRD, presented n [5]. IEEE Communcatons Socety 079

Maximum Weight Matching Dispatching Scheme in Buffered Clos-network Packet Switches

Maximum Weight Matching Dispatching Scheme in Buffered Clos-network Packet Switches Maximum Weight Matching Dispatching Scheme in Buffered Clos-network Packet Switches Roberto Rojas-Cessa, Member, IEEE, Eiji Oki, Member, IEEE, and H. Jonathan Chao, Fellow, IEEE Abstract The scalability

More information

PCRRD: A Pipeline-Based Concurrent Round-Robin Dispatching Scheme for Clos-Network Switches

PCRRD: A Pipeline-Based Concurrent Round-Robin Dispatching Scheme for Clos-Network Switches : A Pipeline-Based Concurrent Round-Robin Dispatching Scheme for Clos-Network Switches Eiji Oki, Roberto Rojas-Cessa, and H. Jonathan Chao Abstract This paper proposes a pipeline-based concurrent round-robin

More information

Concurrent Round-Robin Dispatching Scheme in a Clos-Network Switch

Concurrent Round-Robin Dispatching Scheme in a Clos-Network Switch Concurrent Round-Robin Dispatching Scheme in a Clos-Network Switch Eiji Oki * Zhigang Jing Roberto Rojas-Cessa H. Jonathan Chao NTT Network Service Systems Laboratories Department of Electrical Engineering

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

Real-Time Guarantees. Traffic Characteristics. Flow Control

Real-Time Guarantees. Traffic Characteristics. Flow Control Real-Tme Guarantees Requrements on RT communcaton protocols: delay (response s) small jtter small throughput hgh error detecton at recever (and sender) small error detecton latency no thrashng under peak

More information

CS 268: Lecture 8 Router Support for Congestion Control

CS 268: Lecture 8 Router Support for Congestion Control CS 268: Lecture 8 Router Support for Congeston Control Ion Stoca Computer Scence Dvson Department of Electrcal Engneerng and Computer Scences Unversty of Calforna, Berkeley Berkeley, CA 9472-1776 Router

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

A fair buffer allocation scheme

A fair buffer allocation scheme A far buffer allocaton scheme Juha Henanen and Kalev Klkk Telecom Fnland P.O. Box 228, SF-330 Tampere, Fnland E-mal: juha.henanen@tele.f Abstract An approprate servce for data traffc n ATM networks requres

More information

Using Memory and Random Sampling for Load Balancing in High-radix Switches

Using Memory and Random Sampling for Load Balancing in High-radix Switches Unversty of Toronto SNL Techncal Report TR-SN-UT-7--2. c Soudeh Ghorban Khaled et al. Usng Memory and Random Samplng for Load Balancng n Hgh-radx Swtches Soudeh Ghorban Khaled, Yashar Ganjal Department

More information

Buffering High-Speed Packets with Tri-Stage Memory Array and Its Performance Analysis

Buffering High-Speed Packets with Tri-Stage Memory Array and Its Performance Analysis 000-9825/2005/6(2)28 2005 Journal of Software Vol6, No2,2+, 2,, (,00084) 2 (, 450002) Bufferng Hgh-Speed Packets wth Tr-Stage Memory Array and Its Performance Analyss WANG Peng,2+, YI Peng 2, JIN De-Peng,

More information

Load-Balanced Anycast Routing

Load-Balanced Anycast Routing Load-Balanced Anycast Routng Chng-Yu Ln, Jung-Hua Lo, and Sy-Yen Kuo Department of Electrcal Engneerng atonal Tawan Unversty, Tape, Tawan sykuo@cc.ee.ntu.edu.tw Abstract For fault-tolerance and load-balance

More information

Routing in Degree-constrained FSO Mesh Networks

Routing in Degree-constrained FSO Mesh Networks Internatonal Journal of Hybrd Informaton Technology Vol., No., Aprl, 009 Routng n Degree-constraned FSO Mesh Networks Zpng Hu, Pramode Verma, and James Sluss Jr. School of Electrcal & Computer Engneerng

More information

K-Selector-Based Dispatching Algorithm for Clos-Network Switches

K-Selector-Based Dispatching Algorithm for Clos-Network Switches K-Selector-Based Dispatching Algorithm for Clos-Network Switches Mei Yang, Mayauna McCullough, Yingtao Jiang, and Jun Zheng Department of Electrical and Computer Engineering, University of Nevada Las Vegas,

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

Closed Loop Credit-Based Flow Control with Internal Backpressure in Input and Output Queued Switches

Closed Loop Credit-Based Flow Control with Internal Backpressure in Input and Output Queued Switches Closed Loop Credt-Based Flow Control wth Internal Backpressure n Input and Output Queued Swtches Raner Schoenen and Achm Dahlhoff Insttute for Integrated Sgnal Processng Systems Aachen Unversty of Technology,

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

Constructing Minimum Connected Dominating Set: Algorithmic approach

Constructing Minimum Connected Dominating Set: Algorithmic approach Constructng Mnmum Connected Domnatng Set: Algorthmc approach G.N. Puroht and Usha Sharma Centre for Mathematcal Scences, Banasthal Unversty, Rajasthan 304022 usha.sharma94@yahoo.com Abstract: Connected

More information

A New Token Allocation Algorithm for TCP Traffic in Diffserv Network

A New Token Allocation Algorithm for TCP Traffic in Diffserv Network A New Token Allocaton Algorthm for TCP Traffc n Dffserv Network A New Token Allocaton Algorthm for TCP Traffc n Dffserv Network S. Sudha and N. Ammasagounden Natonal Insttute of Technology, Truchrappall,

More information

DESIGNING TRANSMISSION SCHEDULES FOR WIRELESS AD HOC NETWORKS TO MAXIMIZE NETWORK THROUGHPUT

DESIGNING TRANSMISSION SCHEDULES FOR WIRELESS AD HOC NETWORKS TO MAXIMIZE NETWORK THROUGHPUT DESIGNING TRANSMISSION SCHEDULES FOR WIRELESS AD HOC NETWORKS TO MAXIMIZE NETWORK THROUGHPUT Bran J. Wolf, Joseph L. Hammond, and Harlan B. Russell Dept. of Electrcal and Computer Engneerng, Clemson Unversty,

More information

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

Gateway Algorithm for Fair Bandwidth Sharing

Gateway Algorithm for Fair Bandwidth Sharing Algorm for Far Bandwd Sharng We Y, Rupnder Makkar, Ioanns Lambadars Department of System and Computer Engneerng Carleton Unversty 5 Colonel By Dr., Ottawa, ON KS 5B6, Canada {wy, rup, oanns}@sce.carleton.ca

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

with `ook-ahead for Broadcast WDM Networks TR May 14, 1996 Abstract

with `ook-ahead for Broadcast WDM Networks TR May 14, 1996 Abstract HPeR-`: A Hgh Performance Reservaton Protocol wth `ook-ahead for Broadcast WDM Networks Vjay Svaraman George N. Rouskas TR-96-06 May 14, 1996 Abstract We consder the problem of coordnatng access to the

More information

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

A Saturation Binary Neural Network for Crossbar Switching Problem

A Saturation Binary Neural Network for Crossbar Switching Problem A Saturaton Bnary Neural Network for Crossbar Swtchng Problem Cu Zhang 1, L-Qng Zhao 2, and Rong-Long Wang 2 1 Department of Autocontrol, Laonng Insttute of Scence and Technology, Benx, Chna bxlkyzhangcu@163.com

More information

Integrated Congestion-Control Mechanism in Optical Burst Switching Networks

Integrated Congestion-Control Mechanism in Optical Burst Switching Networks Ths full text paper was peer revewed at the drecton of IEEE Communcatons Socety subect matter experts for publcaton n the IEEE GLOBECOM 2005 proceedngs Integrated Congeston-Control Mechansm n Optcal Burst

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

WIRELESS communication technology has gained widespread

WIRELESS communication technology has gained widespread 616 IEEE TRANSACTIONS ON MOBILE COMPUTING, VOL. 4, NO. 6, NOVEMBER/DECEMBER 2005 Dstrbuted Far Schedulng n a Wreless LAN Ntn Vadya, Senor Member, IEEE, Anurag Dugar, Seema Gupta, and Paramvr Bahl, Senor

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

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

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

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

Specifications in 2001

Specifications in 2001 Specfcatons n 200 MISTY (updated : May 3, 2002) September 27, 200 Mtsubsh Electrc Corporaton Block Cpher Algorthm MISTY Ths document shows a complete descrpton of encrypton algorthm MISTY, whch are secret-key

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

Scheduling and queue management. DigiComm II

Scheduling and queue management. DigiComm II Schedulng and queue management Tradtonal queung behavour n routers Data transfer: datagrams: ndvdual packets no recognton of flows connectonless: no sgnallng Forwardng: based on per-datagram forwardng

More information

Quantifying Responsiveness of TCP Aggregates by Using Direct Sequence Spread Spectrum CDMA and Its Application in Congestion Control

Quantifying Responsiveness of TCP Aggregates by Using Direct Sequence Spread Spectrum CDMA and Its Application in Congestion Control Quantfyng Responsveness of TCP Aggregates by Usng Drect Sequence Spread Spectrum CDMA and Its Applcaton n Congeston Control Mehd Kalantar Department of Electrcal and Computer Engneerng Unversty of Maryland,

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

On Achieving Fairness in the Joint Allocation of Buffer and Bandwidth Resources: Principles and Algorithms

On Achieving Fairness in the Joint Allocation of Buffer and Bandwidth Resources: Principles and Algorithms On Achevng Farness n the Jont Allocaton of Buffer and Bandwdth Resources: Prncples and Algorthms Yunka Zhou and Harsh Sethu (correspondng author) Abstract Farness n network traffc management can mprove

More information

TECHNICAL REPORT AN OPTIMAL DISTRIBUTED PROTOCOL FOR FAST CONVERGENCE TO MAXMIN RATE ALLOCATION. Jordi Ros and Wei K Tsai

TECHNICAL REPORT AN OPTIMAL DISTRIBUTED PROTOCOL FOR FAST CONVERGENCE TO MAXMIN RATE ALLOCATION. Jordi Ros and Wei K Tsai TECHNICAL REPORT AN OPTIMAL DISTRIUTED PROTOCOL FOR FAST CONVERGENCE TO MAXMIN RATE ALLOCATION Jord Ros and We K Tsa Department of Electrcal and Computer Engneerng Unversty of Calforna, Irvne 1 AN OPTIMAL

More information

Sample Solution. Advanced Computer Networks P 1 P 2 P 3 P 4 P 5. Module: IN2097 Date: Examiner: Prof. Dr.-Ing. Georg Carle Exam: Final exam

Sample Solution. Advanced Computer Networks P 1 P 2 P 3 P 4 P 5. Module: IN2097 Date: Examiner: Prof. Dr.-Ing. Georg Carle Exam: Final exam Char of Network Archtectures and Servces Department of Informatcs Techncal Unversty of Munch Note: Durng the attendance check a stcker contanng a unque QR code wll be put on ths exam. Ths QR code contans

More information

Goals and Approach Type of Resources Allocation Models Shared Non-shared Not in this Lecture In this Lecture

Goals and Approach Type of Resources Allocation Models Shared Non-shared Not in this Lecture In this Lecture Goals and Approach CS 194: Dstrbuted Systems Resource Allocaton Goal: acheve predcable performances Three steps: 1) Estmate applcaton s resource needs (not n ths lecture) 2) Admsson control 3) Resource

More information

Advanced Computer Networks

Advanced Computer Networks Char of Network Archtectures and Servces Department of Informatcs Techncal Unversty of Munch Note: Durng the attendance check a stcker contanng a unque QR code wll be put on ths exam. Ths QR code contans

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

Dynamic Bandwidth Allocation Schemes in Hybrid TDM/WDM Passive Optical Networks

Dynamic Bandwidth Allocation Schemes in Hybrid TDM/WDM Passive Optical Networks Dynamc Bandwdth Allocaton Schemes n Hybrd TDM/WDM Passve Optcal Networks Ahmad R. Dhan, Chad M. Ass, and Abdallah Sham Concorda Insttue for Informaton Systems Engneerng Concorda Unversty, Montreal, Quebec,

More information

Reliability and Performance Models for Grid Computing

Reliability and Performance Models for Grid Computing Relablty and Performance Models for Grd Computng Yuan-Shun Da,2, Jack Dongarra,3,4 Department of Electrcal Engneerng and Computer Scence, Unversty of Tennessee, Knoxvlle 2 Department of Industral and Informaton

More information

Term Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task

Term Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task Proceedngs of NTCIR-6 Workshop Meetng, May 15-18, 2007, Tokyo, Japan Term Weghtng Classfcaton System Usng the Ch-square Statstc for the Classfcaton Subtask at NTCIR-6 Patent Retreval Task Kotaro Hashmoto

More information

Classifier Selection Based on Data Complexity Measures *

Classifier Selection Based on Data Complexity Measures * Classfer Selecton Based on Data Complexty Measures * Edth Hernández-Reyes, J.A. Carrasco-Ochoa, and J.Fco. Martínez-Trndad Natonal Insttute for Astrophyscs, Optcs and Electroncs, Lus Enrque Erro No.1 Sta.

More information

ARTICLE IN PRESS. Computer Networks xxx (2008) xxx xxx. Contents lists available at ScienceDirect. Computer Networks

ARTICLE IN PRESS. Computer Networks xxx (2008) xxx xxx. Contents lists available at ScienceDirect. Computer Networks Computer Networks xxx (28) xxx xxx Contents lsts avalable at ScenceDrect Computer Networks journal homepage: www.elsever.com/locate/comnet On applyng far queung dscplne to schedule requests at access gateway

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

Efficient Load-Balanced IP Routing Scheme Based on Shortest Paths in Hose Model. Eiji Oki May 28, 2009 The University of Electro-Communications

Efficient Load-Balanced IP Routing Scheme Based on Shortest Paths in Hose Model. Eiji Oki May 28, 2009 The University of Electro-Communications Effcent Loa-Balance IP Routng Scheme Base on Shortest Paths n Hose Moel E Ok May 28, 2009 The Unversty of Electro-Communcatons Ok Lab. Semnar, May 28, 2009 1 Outlne Backgroun on IP routng IP routng strategy

More information

Towards Efficient Optical Burst-Switched Networks without All-Optical Wavelength Converters

Towards Efficient Optical Burst-Switched Networks without All-Optical Wavelength Converters Towards Effcent Optcal Burst-Swtched Networks wthout All-Optcal Wavelength Converters João Pedro 1,2, Paulo Montero 1,3, and João Pres 2 1 D1 Research, Semens Networks S.A., R. Irmãos Semens 1, 2720-093

More information

Efficient Content Distribution in Wireless P2P Networks

Efficient Content Distribution in Wireless P2P Networks Effcent Content Dstrbuton n Wreless P2P Networs Qong Sun, Vctor O. K. L, and Ka-Cheong Leung Department of Electrcal and Electronc Engneerng The Unversty of Hong Kong Pofulam Road, Hong Kong, Chna {oansun,

More information

The Codesign Challenge

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

More information

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

3. CR parameters and Multi-Objective Fitness Function

3. CR parameters and Multi-Objective Fitness Function 3 CR parameters and Mult-objectve Ftness Functon 41 3. CR parameters and Mult-Objectve Ftness Functon 3.1. Introducton Cogntve rados dynamcally confgure the wreless communcaton system, whch takes beneft

More information

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

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

More information

Computer Communications

Computer Communications Computer Communcatons 3 (22) 3 48 Contents lsts avalable at ScVerse ScenceDrect Computer Communcatons journal homepage: www.elsever.com/locate/comcom On the queueng behavor of nter-flow asynchronous network

More information

Analysis of Collaborative Distributed Admission Control in x Networks

Analysis of Collaborative Distributed Admission Control in x Networks 1 Analyss of Collaboratve Dstrbuted Admsson Control n 82.11x Networks Thnh Nguyen, Member, IEEE, Ken Nguyen, Member, IEEE, Lnha He, Member, IEEE, Abstract Wth the recent surge of wreless home networks,

More information

Polyhedral Compilation Foundations

Polyhedral Compilation Foundations Polyhedral Complaton Foundatons Lous-Noël Pouchet pouchet@cse.oho-state.edu Dept. of Computer Scence and Engneerng, the Oho State Unversty Feb 8, 200 888., Class # Introducton: Polyhedral Complaton Foundatons

More information

Pricing Network Resources for Adaptive Applications in a Differentiated Services Network

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

More information

A Sub-Critical Deficit Round-Robin Scheduler

A Sub-Critical Deficit Round-Robin Scheduler A Sub-Crtcal Defct ound-obn Scheduler Anton Kos, Sašo Tomažč Unversty of Ljubljana, Faculty of Electrcal Engneerng, Ljubljana, Slovena E-mal: anton.kos@fe.un-lj.s Abstract - A scheduler s an essental element

More information

A RECONFIGURABLE ARCHITECTURE FOR MULTI-GIGABIT SPEED CONTENT-BASED ROUTING. James Moscola, Young H. Cho, John W. Lockwood

A RECONFIGURABLE ARCHITECTURE FOR MULTI-GIGABIT SPEED CONTENT-BASED ROUTING. James Moscola, Young H. Cho, John W. Lockwood A RECONFIGURABLE ARCHITECTURE FOR MULTI-GIGABIT SPEED CONTENT-BASED ROUTING James Moscola, Young H. Cho, John W. Lockwood Dept. of Computer Scence and Engneerng Washngton Unversty, St. Lous, MO {jmm5,

More information

International Journal of Scientific & Engineering Research, Volume 4, Issue 12, December-2013 ISSN

International Journal of Scientific & Engineering Research, Volume 4, Issue 12, December-2013 ISSN 833 Polynomal Tme Algorthm (PTA) to generate a survvable rng vrtual topology for Optcal WDM based on Hamltonan Crcut Detecton Heurstcs(HCDH) *Sajal Kanta Das, Department of Computer Scence & Technology,

More information

THere are increasing interests and use of mobile ad hoc

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

More information

RAP. Speed/RAP/CODA. Real-time Systems. Modeling the sensor networks. Real-time Systems. Modeling the sensor networks. Real-time systems:

RAP. Speed/RAP/CODA. Real-time Systems. Modeling the sensor networks. Real-time Systems. Modeling the sensor networks. Real-time systems: Speed/RAP/CODA Presented by Octav Chpara Real-tme Systems Many wreless sensor network applcatons requre real-tme support Survellance and trackng Border patrol Fre fghtng Real-tme systems: Hard real-tme:

More information

Response-Time Guarantees in ATM Networks

Response-Time Guarantees in ATM Networks Response-Tme Guarantees n ATM Networks Andreas Ermedahl Hans Hansson Mkael Sjödn Department of Computer Systems Uppsala Unversty Sweden E-mal: febbe,hansh,mcg@docs.uu.se Abstract We present a method for

More information

ARTICLE IN PRESS. Signal Processing: Image Communication

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

More information

VISUAL SELECTION OF SURFACE FEATURES DURING THEIR GEOMETRIC SIMULATION WITH THE HELP OF COMPUTER TECHNOLOGIES

VISUAL SELECTION OF SURFACE FEATURES DURING THEIR GEOMETRIC SIMULATION WITH THE HELP OF COMPUTER TECHNOLOGIES UbCC 2011, Volume 6, 5002981-x manuscrpts OPEN ACCES UbCC Journal ISSN 1992-8424 www.ubcc.org VISUAL SELECTION OF SURFACE FEATURES DURING THEIR GEOMETRIC SIMULATION WITH THE HELP OF COMPUTER TECHNOLOGIES

More information

2x x l. Module 3: Element Properties Lecture 4: Lagrange and Serendipity Elements

2x x l. Module 3: Element Properties Lecture 4: Lagrange and Serendipity Elements Module 3: Element Propertes Lecture : Lagrange and Serendpty Elements 5 In last lecture note, the nterpolaton functons are derved on the bass of assumed polynomal from Pascal s trangle for the fled varable.

More information

Avoiding congestion through dynamic load control

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

More information

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

A Semi-Distributed Load Balancing Architecture and Algorithm for Heterogeneous Wireless Networks

A Semi-Distributed Load Balancing Architecture and Algorithm for Heterogeneous Wireless Networks A Sem-Dstrbuted oad Balancng Archtecture and Algorthm for Heterogeneous reless Networks Md. Golam Rabul Ala Choong Seon Hong * Kyung Hee Unversty, Korea rob@networkng.khu.ac.kr, cshong@khu.ac.kr Abstract

More information

NOVEL CONSTRUCTION OF SHORT LENGTH LDPC CODES FOR SIMPLE DECODING

NOVEL CONSTRUCTION OF SHORT LENGTH LDPC CODES FOR SIMPLE DECODING Journal of Theoretcal and Appled Informaton Technology 27 JATIT. All rghts reserved. www.jatt.org NOVEL CONSTRUCTION OF SHORT LENGTH LDPC CODES FOR SIMPLE DECODING Fatma A. Newagy, Yasmne A. Fahmy, and

More information

Fibre-Optic AWG-based Real-Time Networks

Fibre-Optic AWG-based Real-Time Networks Fbre-Optc AWG-based Real-Tme Networks Krstna Kunert, Annette Böhm, Magnus Jonsson, School of Informaton Scence, Computer and Electrcal Engneerng, Halmstad Unversty {Magnus.Jonsson, Krstna.Kunert}@de.hh.se

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

A Fair MAC Algorithm with Dynamic Priority for e WLANs

A Fair MAC Algorithm with Dynamic Priority for e WLANs 29 Internatonal Conference on Communcaton Software and Networks A Far MAC Algorthm wth Dynamc Prorty for 82.e WLANs Rong He, Xumng Fang Provncal Key Lab of Informaton Codng & Transmsson, Southwest Jaotong

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

Design of Structure Optimization with APDL

Design of Structure Optimization with APDL Desgn of Structure Optmzaton wth APDL Yanyun School of Cvl Engneerng and Archtecture, East Chna Jaotong Unversty Nanchang 330013 Chna Abstract In ths paper, the desgn process of structure optmzaton wth

More information

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

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

More information

APPLICATION OF MULTIVARIATE LOSS FUNCTION FOR ASSESSMENT OF THE QUALITY OF TECHNOLOGICAL PROCESS MANAGEMENT

APPLICATION OF MULTIVARIATE LOSS FUNCTION FOR ASSESSMENT OF THE QUALITY OF TECHNOLOGICAL PROCESS MANAGEMENT 3. - 5. 5., Brno, Czech Republc, EU APPLICATION OF MULTIVARIATE LOSS FUNCTION FOR ASSESSMENT OF THE QUALITY OF TECHNOLOGICAL PROCESS MANAGEMENT Abstract Josef TOŠENOVSKÝ ) Lenka MONSPORTOVÁ ) Flp TOŠENOVSKÝ

More information

Comparisons of Packet Scheduling Algorithms for Fair Service among Connections on the Internet

Comparisons of Packet Scheduling Algorithms for Fair Service among Connections on the Internet Comparsons of Packet Schedulng Algorthms for Far Servce among Connectons on the Internet Go Hasegawa, Takahro Matsuo, Masayuk Murata and Hdeo Myahara Department of Infomatcs and Mathematcal Scence Graduate

More information

38050 Povo Trento (Italy), Via Sommarive 14 Efficient Scheduling for Heterogeneous Fractional Lambda Switching (FλS) Networks

38050 Povo Trento (Italy), Via Sommarive 14  Efficient Scheduling for Heterogeneous Fractional Lambda Switching (FλS) Networks UNIVERSITY OF TRENTO DEPARTMENT OF INFORMATION AND COMMUNICATION TECHNOLOGY 38050 Povo Trento (Italy), Va Sommarve 14 http://www.dt.untn.t Effcent Schedulng for Heterogeneous Fractonal Lambda Swtchng (FλS)

More information

ELEC 377 Operating Systems. Week 6 Class 3

ELEC 377 Operating Systems. Week 6 Class 3 ELEC 377 Operatng Systems Week 6 Class 3 Last Class Memory Management Memory Pagng Pagng Structure ELEC 377 Operatng Systems Today Pagng Szes Vrtual Memory Concept Demand Pagng ELEC 377 Operatng Systems

More information

On the Exact Analysis of Bluetooth Scheduling Algorithms

On the Exact Analysis of Bluetooth Scheduling Algorithms On the Exact Analyss of Bluetooth Schedulng Algorth Gl Zussman Dept. of Electrcal Engneerng Technon IIT Hafa 3000, Israel glz@tx.technon.ac.l Ur Yechal Dept. of Statstcs and Operatons Research School of

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

VRT012 User s guide V0.1. Address: Žirmūnų g. 27, Vilnius LT-09105, Phone: (370-5) , Fax: (370-5) ,

VRT012 User s guide V0.1. Address: Žirmūnų g. 27, Vilnius LT-09105, Phone: (370-5) , Fax: (370-5) , VRT012 User s gude V0.1 Thank you for purchasng our product. We hope ths user-frendly devce wll be helpful n realsng your deas and brngng comfort to your lfe. Please take few mnutes to read ths manual

More information

Quantifying Performance Models

Quantifying Performance Models Quantfyng Performance Models Prof. Danel A. Menascé Department of Computer Scence George Mason Unversty www.cs.gmu.edu/faculty/menasce.html 1 Copyrght Notce Most of the fgures n ths set of sldes come from

More information

Technical Report. i-game: An Implicit GTS Allocation Mechanism in IEEE for Time- Sensitive Wireless Sensor Networks

Technical Report. i-game: An Implicit GTS Allocation Mechanism in IEEE for Time- Sensitive Wireless Sensor Networks www.hurray.sep.pp.pt Techncal Report -GAME: An Implct GTS Allocaton Mechansm n IEEE 802.15.4 for Tme- Senstve Wreless Sensor etworks Ans Koubaa Máro Alves Eduardo Tovar TR-060706 Verson: 1.0 Date: Jul

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

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

Solutions for Real-Time Communication over Best-Effort Networks

Solutions for Real-Time Communication over Best-Effort Networks Solutons for Real-Tme Communcaton over Best-Effort Networks Anca Hangan, Ramona Marfevc, Gheorghe Sebestyen Techncal Unversty of Cluj-Napoca, Computer Scence Department {Anca.Hangan, Ramona.Marfevc, Gheorghe.Sebestyen}@cs.utcluj.ro

More information

Algorithms for Routing and Centralized Scheduling to Provide QoS in IEEE Mesh Networks

Algorithms for Routing and Centralized Scheduling to Provide QoS in IEEE Mesh Networks Algorthms for Routng and Centralzed Schedulng to Provde QoS n IEEE 802.16 Mesh Networks Harsh Shetya Dept of Electrcal Communcaton Engneerng Indan Insttute of Scence Bangalore, Inda. 560012 harsh@pal.ece.sc.ernet.n

More information

MULTIHOP wireless networks are a paradigm in wireless

MULTIHOP wireless networks are a paradigm in wireless 400 IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, VOL. 17, NO. 1, JANUARY 2018 Toward Optmal Dstrbuted Node Schedulng n a Multhop Wreless Network Through Local Votng Dmtros J. Vergados, Member, IEEE, Natala

More information

Queueing Network-based Optimisation Techniques for Workload Allocation in Clusters of Computers *

Queueing Network-based Optimisation Techniques for Workload Allocation in Clusters of Computers * Queueng Network-based Optmsaton Technques for Workload Allocaton n Clusters of Computers * Lgang He, Stephen A. Jarvs, Davd Bacgalupo, Danel P. Spooner, Xnuo Chen and Graham R. Nudd Department of Computer

More information

Connection-information-based connection rerouting for connection-oriented mobile communication networks

Connection-information-based connection rerouting for connection-oriented mobile communication networks Dstrb. Syst. Engng 5 (1998) 47 65. Prnted n the UK PII: S0967-1846(98)90513-7 Connecton-nformaton-based connecton reroutng for connecton-orented moble communcaton networks Mnho Song, Yanghee Cho and Chongsang

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

Tapping into the Router s Unutilized Processing Power

Tapping into the Router s Unutilized Processing Power Tappng nto the Router s Unutlzed Processng Power Marat Radan, Isaac Keslassy Technon {radan@tx, saac@ee}.technon.ac.l Abstract The growng demand for network programmablty has led to the ntroducton of complex

More information

IR-HARQ vs. Joint Channel-Network coding for Cooperative Wireless Communication

IR-HARQ vs. Joint Channel-Network coding for Cooperative Wireless Communication Cyber Journals: ultdscplnary Journals n Scence and Technology, Journal of Selected Areas n Telecommuncatons (JSAT), August Edton, 2 IR-HARQ vs. Jont Channel-Network codng for Cooperatve Wreless Communcaton

More information

Repeater Insertion for Two-Terminal Nets in Three-Dimensional Integrated Circuits

Repeater Insertion for Two-Terminal Nets in Three-Dimensional Integrated Circuits Repeater Inserton for Two-Termnal Nets n Three-Dmensonal Integrated Crcuts Hu Xu, Vasls F. Pavlds, and Govann De Mchel LSI - EPFL, CH-5, Swtzerland, {hu.xu,vasleos.pavlds,govann.demchel}@epfl.ch Abstract.

More information

CACHE MEMORY DESIGN FOR INTERNET PROCESSORS

CACHE MEMORY DESIGN FOR INTERNET PROCESSORS CACHE MEMORY DESIGN FOR INTERNET PROCESSORS WE EVALUATE A SERIES OF THREE PROGRESSIVELY MORE AGGRESSIVE ROUTING-TABLE CACHE DESIGNS AND DEMONSTRATE THAT THE INCORPORATION OF HARDWARE CACHES INTO INTERNET

More information