Improved Clock-Gating through Transparent Pipelining

Size: px
Start display at page:

Download "Improved Clock-Gating through Transparent Pipelining"

Transcription

1 2. Improved Clock-Gting through Trnsprent Pipelining Hns M. Jcobson IM T.J. Wtson Reserch Center, Yorktown, NY. STRCT This pper re-exmines the well estblished clocking principles of pipelines. It is observed tht clock gting techniques tht hve long been ssumed optiml in relity produce significnt mount of redundnt clock pulses. The pper presents new theory for optiml clocking of synchronous pipelines, presents prcticl implementtions nd evlutes the clock power benefits on multiply/dd-ccumulte unit design. Trnsistor level simultions show tht dynmic clock power dissiption cn be reduced by 4-6% t pipeline utiliztion fctors between 2-6%, on top of trditionl stge-level clock gting, without ffecting pipeline ltency or throughput. Ctegories nd Subject Descriptors C..3 [Processor rchitectures]: Other rchitecture Styles - Pipeline processors. Generl Terms Design, Performnce. Keywords Optiml pipeline clocking, Trnsprent pipeline, Pipeline stge unifiction, dptive pipeline depth, Dynmic pipeline scling, Clock gting, Low power, High performnce, Microrchitecture, Circuits.. INTRODUCTION Clock power is key design constrint in modern VLSI design. Despite increses in lekge power, clock power remins significnt prt of the totl power dissiption in modern microprocessors []. Clock gting hs shown to be n efficient technique to significntly reduce dynmic power dissiption [3, 5, 8, 2]. However, despite fine grined clock gting, power consumed by the clock remins mjor contributor to overll chip power dissiption. The work presented in this pper re-exmines the fundmentl clocking principles of pipelines. Our work shows tht trditionl clock gting techniques tht hve long been thought to gte the clock optimlly produce significnt mount of clock pulses tht re redundnt to the correct opertion of the pipeline. Our observtions of requirements for correct pipeline opertion rrive t novel nd prcticl clocking solution tht cn significntly reduce clock power by relxing the clocking requirements of pipelines. Permission to mke digitl or hrd copies of ll or prt of this work for personl or clssroom use is grnted without fee provided tht copies re not mde or distributed for profit or commercil dvntge nd tht copies ber this notice nd the full cittion on the first pge. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission nd/or fee. ISLPED 4, ugust 9, 24, Newport ech, Cliforni, US. Copyright 24 CM /4/8...$5.. Relted work The gol of the clocking technique presented in this pper is to reduce the clock power in trditionl synchronous pipelines. This is chieved by reducing the number of clock pulses required to propgte dt item through the pipeline. In relted work, severl techniques hve trgeted clock power reduction by reducing the number of clock pulses generted to the ltch stges in the pipeline. Clock gting [3, 5, 8, 2] hs been used to reduce dynmic clock power through ll levels of the design hierrchy from the top-most chip level down to the individul ltch level. While these techniques reduce the number of clock pulses generted in the pipeline, they ll hve one thing in common: To void dt rces through the ltches of djcent pipeline stges, ech ltch stge is clocked t lest once for ech dt item propgting through the pipeline. Collpsible pipelining techniques hve been presented to reduce clock power in pipelines [6, 7, 4]. During runtime, dt ltches in collpsed pipeline stges re mde permnently trnsprent for durtion of time. The collpsing technique subsequently reduces the number of ltches tht hve to be clocked. These novel techniques cn significntly reduce clock power, but re lso limited in two wys. First, when collpsing pipeline stges, the logic depth of ech collpsed pipeline stge is doubled (or more depending on how mny stges in sequence re collpsed). This cuses the pipeline opertion frequency to be cut in hlf (or more). Pipeline collpsing techniques subsequently trde frequency for power which my cuse significnt drop in performnce. Second, while the pipeline cn be collpsed for durtion of time during runtime, this collpsing is sttic in nture nd ffects throughput for the whole pipeline. ecuse of fundmentl limittions of these collpsing techniques, they cnnot be pplied dynmiclly on cycle-by-cycle bsis to sve clock power. F E D C ) Collpsible pipeline operting in deep mode frequency = F, throughput = T, clock power = C C ) Collpsible pipeline operting in shllow mode frequency = F/2, throughput = T/2, clock power = C/2 Figure : Prior work: sttic collpsible pipeline. Figure illustrtes the concept of sttic collpsible pipeline. In deep mode ll ltches operte in opque mode nd the pipeline cn run t full frequency. In shllow mode, every other ltch stge is mde trnsprent, collpsing two pipeline stges into one. In shllow mode, the pipeline runs t hlf frequency nd only every other ltch stge is clocked. 26

2 Contributions To summrize, the common limittion of previous work in the re of clock gting is tht, for given opertion frequency (tht my chnge during runtime), ll ltch stges required to be ctive in order to run t tht frequency hve to be clocked t lest once for ech dt item pssing through the pipeline. In contrst, the work presented in this pper cn void clocking ctive ltch stges by dynmiclly dpting to the current stte of the pipeline, on cycle-by-cycle bsis, without reducing the opertion frequency or throughput of the pipeline. Our technique is bsed on the observtion tht ltch stge only needs to go opque in order to seprte closely spced dt items in pipeline. y keeping ltches trnsprent by defult, our technique llows dt items tht re sufficiently seprted in time (clock cycles) to propgte through the pipeline without generting ny clock pulses. Such seprtion occur frequently in, for exmple, microprocessors due to pipeline stlls cused by dt dependencies. The reminder of this pper first discusses the theoreticl clocking requirements of pipelines in Section 2. Section 3 presents trnsprent pipelines t conceptul level. Prcticl reliztions for control logic nd clock blocks re presented in Section 4. Results re presented in Section 5 nd conclusions re given in Section 6. ssumptions Throughout this pper we ssume timing constrints tht re stndrd for sttic trnsprent ltch bsed pipelines. The long pth nd short pth delys through pipeline re llowed to be rbitrry nd stndrd ltch setup nd hold times pply. No new timing constrints re introduced by our technique nd the technique works with ny type of trnsprent ltches. The correct opertion of trnsprent pipeline, s presented in this pper, plces two behviorl constrints on its input nd output environment. First, trnsprent pipeline requires tht the vlue of dt item is held stble t the environment input until next subsequent vlid dt item rrives. Second, trnsprent pipeline requires tht the output environment only ltches dt indicted vlid by the trnsprent pipeline. Simply put, the environment input nd output stges need to be clock gted t the stge level in trditionl opque fshion. 2. PIPELINE CLOCKING RE-EXMINED In trditionl pipeline implementtions, fundmentl ssumption hs been tht ltch stges of pipeline must be held opque by defult in order to void dt rces between the ltch stges. Such dt rces my cuse dt residing in n upstrem stge to ccidentlly overwrite the dt in downstrem stge due to differences in the depth of logic pths between ltches. y keeping ltches opque by defult, dt rces through ltches re voided. However, this invribly results in pessimistic clocking model tht produces clock pulses tht re redundnt to the functionlity of the pipeline. The following sections propose new technique for clocking synchronous pipelines. To eliminte redundnt clocking of pipeline, we first introduce the notion of trnsprent pipeline in which ll ltches re trnsprent by defult. Second, we develop new clocking model in which ltches re mde opque (clocked) only when true dt rce is present in the pipeline. The proposed technique cn significntly reduce the number of clock pulses required to propgte dt item through pipeline. Pipeline correctness criteri true dt rce exists only if two dt items propgte through pipeline without ny opque ltch stge to seprte them. ssuming rbitrry min/mx dely of logic pths through combintionl logic nd ltches, the following criteri must be met to ensure correct opertion of trnsprent pipeline. Requirements to void dt rces: For ech pir of distinct djcent dt items (,) propgting through pipeline, where is downstrem of, t lest one opque ltch stge must seprte from (nd from in cse of circulr pipeline). For ech dt item propgting through circulr pipeline, t lest one opque ltch stge must seprte from the til of. Given these requirements to void dt rces, the criteri required to implement optimum clocking of trnsprent pipeline cn be derived. In this context the concept of stte holder is introduced. stte holder for dt item is the opque ltch stge holding the most recent vlue of stble. This is typiclly the opque ltch stge closest upstrem of the current position of. Criteri for optimum clocking of pipeline: For ech pir of djcent dt items (,) propgting through pipeline, where is downstrem of, the ltch stge for is clocked only when overwrites the current stte holder for, nd For ech dt item propgting through circulr pipeline, only one stge for is clocked, nd only once ech itertion. In non-liner pipelines, such s forks nd joins, cn hve multiple stte holders ech holding prt of the vlue of. If ny of these stte holders re overwritten, new stte holder must be provided for t its current loction in the pipeline. The consequence of the stted clocking criteri for trnsprent pipeline is tht ltch stge only needs to be clocked in order to seprte pir of dt items moving concurrently through the pipeline. This is in contrst to the trditionl clocking criteri of n opque pipeline tht sttes tht ltch stge needs to be clocked in order to propgte dt item moving through the pipeline. The first criteri is the significntly more relxed of the two nd llows for tngible reduction in required clock pulses. 3. TRNSPRENT PIPELINE: CONCEPT trnsprent pipeline keeps its ltch stges trnsprent by defult. This defult stte represents the trnsprent clock gted mode of the ltch stge (trnsprent mode). Dt rces between ltches re voided by physiclly seprting ech pir of dt items concurrently propgting through the trnsprent pipeline. pir of dt items re seprted by forcing ltch stge residing between the pir to enter n opque stte. This opque stte cn be either the opquely clock gted mode of the ltch stge (opque mode), or the norml clocking of the ltch stge (clocked mode). ltch stge in generlized trnsprent pipeline cn thus operte in three different modes. Figure 2 illustrte the behvior of five stge liner pipeline. The three middle ltch stges, 2, 3, nd 4 form trnsprent pipeline segment. These ltch stges operte in trnsprent mode by defult. Ltch stges nd 5 form the input nd output environment of the trnsprent pipeline. These ltch stges operte in trditionl opque mode by defult. vlid ltch is ssocited with ech stge to keep trck of the loction of vlid dt in the pipeline. The vlid ltches re clocked ech clock cycle. In the figure, dotted lines indicte tht the ltch is clock gted in trnsprent mode. Solid lines 27

3 stge: clock cycle: glb.clk lcl.clk lcl.clk 2 clock cycle lcl.clk 3 clock cycle lcl.clk 4 lcl.clk 5 () Trnsprent clock gting time clock cycle 2 glb.clk lcl.clk lcl.clk 2 lcl.clk 3 b lcl.clk 4 lcl.clk 5 clock cycle 3 (b) Trditionl opque clock gting Figure 3: Clock wveforms corresponding to Figure 2. clock cycle 4 clock cycle 5 clock cycle 6 clock cycle 7 b Figure 2: liner three-stge Trnsprent Pipeline indicte tht ltch is clock gted in opque mode (or is clocked the current cycle). Thick lines indicte tht the ltch is the current stte holder for the dt item mrked inside it, i.e., the ltch is responsible for holding the vlue for tht dt item stble. drk shdow indictes tht the ltch in question cptures dt item, nd lso represents the completion of one cycle of its locl clock. Consider strem of dt items entering the pipeline, where represents the bsence of vlid dt ( bubble). ssume the pipeline is initilly empty. Stges nd 5 re then clock gted in opque mode, nd stges 2, 3, nd 4 re clock gted in trnsprent mode. Clock cycle : s dt item enters stge, t clock cycle, it is cptured nd held stble in the dt ltches. Stge is now the stte holder for. Since the dt ltches for stges 2, 3, nd 4 re clock gted in trnsprent mode cn now propgte freely through this trnsprent segment of the pipeline. In the figure, the short pth propgtion of is indicted in lower cse. The long pth propgtion of is indicted in upper cse. ssuming rbitrry dely on short pths through the logic the dt inputs to the ltches of stge 5 cn chnge t ny time. Since stge 5 is clock gted in opque mode, these unstble vlues re not ltched. Subsequently, there is no risk for metstbility to occur. t the end of clock cycle, the longest pth through the logic in stge hs completed nd the output of the stge is now vlid. Clock cycle 2: t the strt of clock cycle 2, the ssocited vlid bit is cptured by the vlid ltch in stge 2 to indicte the new position of (note tht the vlid ltch is lwys clocked). Note, however, tht since no vlid dt item immeditely follows in the pipeline, stge continues to hold stble. There is subsequently no need to clock the dt ltches of stge 2. The dt ltches of stge 2 therefore remin clock gted in trnsprent mode. Clock cycle 3: In clock cycle 3, stge ltches dt item. Since stge no longer holds dt item stble, stge 3, where currently resides, must cpture nd hold stble. The dt ltches for stge 3 re therefore clocked this cycle nd re therefter held in opque gted mode. Stge is now the stte holder for, while cn stge 3 is the stte holder for. s stge 2 is trnsprent, propgte freely through stges nd 2 during this clock cycle. The short pth propgtion of is indicted with lower cse. Since the dt ltches for stge 3 re opque, cnnot propgte further thn stge 2, nd there is subsequently no risk for dt rces between dt items nd. The vlid ltches re updted to indicte tht resides in stge nd resides in stge 3. Clock cycle 4: During clock cycle 4, no dt ltches hve to be clocked s stge continues to hold stble nd stge 3 continues to hold stble. The vlid ltches re updted to indicte tht now resides in stge 2 nd resides in stge 4. Clock cycle 5: In clock cycle 5, the dt ltches for stge 3 re forced trnsprent to let propgte through nd therefter remin 28

4 E T T2 E3 go C gt C gt C go C vlid[e] vlid[e] vlid[t] vlid[t2] vlid[e3] Figure 4: Control logic implementtion for two-stge trnsprent pipeline. in trnsprent mode. t the sme time, the vlid bit ssocited with indictes to stge 5 of the output environment tht the output of the trnsprent pipeline is now vlid. Stge 5 is subsequently clocked nd cptures. The vlid ltches re updted to indicte tht resides in stge 3 nd resides in stge 5. Clock cycle 6: During clock cycle 6, no dt ltches need to be clocked s stge continues to hold stble. The vlid ltches re updted to indicte tht now resides in stge 4. Clock cycle 7: t the strt of clock cycle 7, the vlid bit ssocited with indictes to stge 5 of the output environment tht the output of the trnsprent pipeline is gin vlid. Stge 5 is subsequently clocked nd cptures. The vlid ltches re updted to indicte tht now resides in stge 5. Figure 3 illustrtes the clock wveforms generted for the exmple illustrted in Figure 2. Figure 3() illustrtes the clock wveforms when stges 2, 3, nd 4 mke use of trnsprent mode clock gting. Figure 3(b) illustrtes the clock wveforms when stges 2, 3, nd 4 mke use of trditionl opque mode clock gting. s cn be redily observed from Figure 3(), the three-stge trnsprent segment of the pipeline genertes the equivlent of only one clock pulse in order to let dt items nd propgte through it. In comprison, the trditionlly opque clock gted pipeline in Figure 3(b) hs to generte totl of six clock pulses to propgte nd through it. The exmple clerly illustrtes the potentil benefit of trnsprent mode clock gting in pipelines with moderte utiliztion fctors. 4. TRNSPRENT PIPELINE: IMPLEMEN- TTION There re mny wys to implement trnsprent pipeline. To better focus the presenttion on the min concepts of trnsprent pipelining, simple nd efficient specil cse implementtion for two-stge non-stllble trnsprent pipeline is given in this pper. The presented technique cn be redily extended to cover generl N-stge stllble pipelines s well. The specil cse implementtion covered in this section is possible when the trnsprent pipeline segment hs two stges or less. In this cse, the behvior of the opque clock gted mode equls the behvior of the clocked mode. The number of opertion modes tht ltch stge hs to support cn thus be reduced from three down to two, providing solution very similr to trditionlly clock gted pipelines. The two opertion modes tht need to be supported re () normlly clocked mode, nd (2) trnsprent mode clock gting. 4. Control logic pipeline stge tht is clock gted in trnsprent mode hs to be ble to detect whether it should switch to clocked mode or sty trnsprent. Clocked mode only hs to be entered in order to seprte two dt items propgting concurrently through the trnsprent segment of the pipeline. For given clock cycle, ny trnsprent pipeline stge should enter clocked opertion mode only under the following conditions: trnsprent pipeline stge should be clocked only if:. vlid dt is present t the input of, nd 2. vlid dt is present t the input of ny trnsprent stge upstrem of, or vlid dt is present t the input of the environment input stge. Whenever the stted conditions do not hold, the trnsprent pipeline stge should operte in trnsprent clock gted mode. Given this simple condition, it is possible to provide strightforwrd implementtion. The first condition cn be implemented by observing the vlid bit feeding into stge. From the second condition, it is cler tht look-behind function is needed to detect whether there is nother dt item upstrem of stge. This lookbehind function cn be implemented by observing the vlid bits of the upstrem pipeline stges. The clock gting conditions for the two trnsprent stges nd, with the input environment nd output environment, then become:! #"%$'&( )+*-,/ (6 3 "7$'&8)+*9,:.; 243=<4>@? *-,/.! 5 6( "7$'&8)+*9,:.; 6 3<!>C? )+*-,/.-243ED!FG*-,/.-!55!H "%$'&( )+*-,/.I6(J5 Figure 4 illustrtes the necessry control logic for n implementtion of synchronous pipeline with two trnsprent stges. 4.2 Clock block clock block (C) supporting trnsprent mode clock gting in two stge trnsprent pipeline is strightforwrd to implement. Figure 5 illustrtes trnsprent nd n opque mode clock block for two-phse clocked mster/slve pipeline. The trnsprent mode clock block contins one mster nd one slve ltch internlly. These ltches re used to ltch the clock gting signl to prevent glitches on the clock. Since both the mster nd slve dt ltches (not shown) re gted in trnsprent mode, the mster nd slve clock signls both need to be gted in their high stte (logic ). In trditionl opque clock gting the gting signl for the mster nd slve clocks cn both be tken from the internl mster ltch. However, when clock gting in trnsprent mode, the clock polrity t the input to the clock gting point for the slve clock, gte 9K in Figure 5, mkes gte 9K sensitive to glitches on the clock gting input while the internl mster ltch is trnsprent. The gting 29

5 gm mster_clk XY US CTL US glb_clk gs slve_clk OOTH gte_trns glb_clk M S ) Clock block supporting proposed trnsprent mode clock gting gm gs mster_clk slve_clk Q DDER YPSS PRTIL PRODUCTS 3/2 COMPRESS x2 3/2 COMPRESS x2 3/2 COMPRESS x2 RESULT FORWRDING Q muxsel gte_opq M y s c x z b) Clock block supporting trditionl opque mode clock gting DDER Figure 5: Two-phse clock block implementtions. DDER or signl for the slve clock must therefore be tken from the internl slve ltch insted. 5. RESULTS 5. Power contributors The power svings chievble in trnsprent pipeline depend on two fctors. First, power is sved by reducing the number of clock pulses tht re generted in the pipeline. This is due to keeping ltches trnsprent so tht dt items tht re sufficiently seprted in time cn propgte through the pipeline without requiring the ltch stges to be clocked. Second, however, dditionl power is consumed s n effect of incresed glitching on dt signls. Opque ltches ct s brriers nd prevent glitches on dt signls to propgte down the pipeline. In trnsprent segments of the pipeline, ltches re held trnsprent. Glitches cn therefore propgte down the pipeline nd cuse extr switching of wires nd trnsistors. The power svings chievble in trnsprent pipelines is subsequently trdeoff between how much power is sved by reducing the number of clock pulses tht hve to be generted versus how much dditionl power is consumed by dditionl glitching on dt signls. 5.2 Evlution results For technique tht dpts dynmiclly to the current utiliztion of the pipeline, it is importnt to evlute the mount of sved clock power over rnge of pipeline utiliztion fctors. Introduced glitch power depends on both pipeline utiliztion nd dt switching fctors. Glitch power therefore hs to be evluted over rnge of both these fctors. It is importnt to estimte worst cse bounds for the introduced glitch power in order to determine the prcticl pplicbility of trnsprent pipelines. The design chosen for evlution purposes is therefore bsed on logic with high glitch tendency. The trnsprent pipeline techniques were evluted on high frequency Multiply/dd-ccumulte (MC) unit illustrted in Figure 6. This type of design ws chosen since dd nd multiply functions re bsed hevily on XOR-type logic tht hs high glitch tendency. The unit implements 32x32 fix-point ooth encoded multiplier with finl dder. The unit fetures bypss pth to llow RESULT US Figure 6: Trnsprently pipelined MC unit. dd instructions to enter the finl dder directly without hving to pss through the multiply stges. forwrding pth is provided to the multiplier to llow multiply-ccumulte instructions. The unit ws implemented s seven stge pipeline in.3 micron technology with trget frequency of 3 GHz. Two implementtions of the unit re evluted. The first design implements trditionl opque stge level clock gting of ll pipeline stges. The second design implements stges 2, 3, 5, nd 6 in trnsprent clock gted mode (shdowed ltch stges in Figure 6). The unit ws simulted t the trnsistor level under rnge of pseudo rndomly generted input vectors for the dt inputs nd vlid inputs. The vlid signls indicte the utiliztion of the pipeline. The simultion includes ll dtpth logic, control logic, ltches, clock gting logic, clock blocks, nd clock buffers. The grph in Figure 7 illustrtes the clock power of the trnsprently clock gted pipeline stges s compred to the sme pipeline stges clock gted in trditionl opque mode over pipeline utiliztion (vlid switch fctor) rnge of -5%. The verge clock power sving over the given utiliztion rnge is 52%. The reltive clock power sving peks t 6% t pipeline utiliztion of 2%. The grph in Figure 8 illustrtes the bsolute clock power svings of the trnsprently clock gted pipeline stges over pipeline utiliztion rnge of -%. The two curves in Figure 8 illustrte the trnsprent clock power svings when the dt input switching fctor is t % nd t % respectively. The difference between the curves illustrtes the rnge between best nd worst cse glitching power introduced s result of keeping ltches trnsprent. From this grph it cn be observed tht the introduced glitch power is never more thn % of the clock power svings. This is n importnt result s it shows tht trnsprent pipelining techniques cn be pplied dvntgeously even in situtions where the dt logic hs high glitch tendency. Figure 8 lso clerly illustrtes the rnge of pipeline utiliztion for which trnsprent pipelining is most effective. The trnsprent pipeline opertes most efficiently in the utiliztion rnge between 2-6% where the clock power dissip- 3

6 2 5 Clock power (mw) (*) Totl power sving (mw) opque trnsprent (*) lso includes introduced glitch power Vlid switch fctor (%) Figure 7: Clock power dissiption for MC unit over pipeline utiliztion rnge of -5% (opque = trditionl opque stge-level clock gting, trnsprent = proposed trnsprent stge-level clock gting). Vlid switch fctor (%) Dt switch fctor % Dt switch fctor % Figure 8: bsolute clock power svings for trnsprently clock gted MC unit, on top of trditionl stge-level clock gting, over pipeline utiliztion rnge of -%. The two curves illustrte best/worst cse bounds for introduced dt glitch power. tion is reduced by 4-6%. The bsolute power sving peks t pipeline utiliztion of 5%. nother importnt thing to note in Figure 8 is tht s the pipeline utiliztion fctor increses, the introduced dt glitch power decreses s more ltch stges re opque nd therefore ct s glitch brriers. s cn be observed in the figure, the trnsprent pipeline therefore lwys performs s good s, or better, thn the opque clock gted pipeline in terms of power. Lstly, it is importnt to keep in mind tht since the presented trnsprent pipelining technique dpts dynmiclly to the current utiliztion stte of the pipeline, on cycle-by-cycle bsis, ltency, throughput, nd IPC of the pipeline re not ffected. The power sving benefits of trnsprent pipelining re chieved without ny performnce degrdtion. The only limittion in the ppliction of trnsprent pipelining techniques is in keeping the dely on the clock gting signls within the trget cycle time. This limits the prcticl length of trnsprent pipeline segment. In high frequency pipelines this limit is typiclly between three to five pipeline stges nd depends gretly on signl rrivl times, ltch count, nd wire distribution. 6. CONCLUSIONS This pper hs re-exmined the clocking principles of trditionl synchronous pipelines. It hs been shown tht clock ctivity cn be significntly reduced, on top of trditionl stge level clock gting, through use of trnsprent pipelining techniques. The pper presents new theory for optiml clocking of synchronous pipelines nd outlines the conceptul opertion of such optimlly clocked pipelines. Prcticl circuit implementtions for control logic nd clock blocks re presented nd the clock power benefits re evluted on high frequency multiply/dd-ccumulte unit design. Trnsistor level simultions of dtpth, control logic, nd clock blocks show tht dynmic clock power cn be reduced by 4-6% t pipeline utiliztion fctors between 2-6% on top of trditionl stge level clock gting. This power benefit is chieved without negtively ffecting pipeline ltency or throughput. It is lso shown tht dt glitch power introduced s result of keep- ing ltch stges trnsprent is limited to below % of the chieved clock power svings, mking trnsprent pipelining techniques pplicble even for logic with high glitch tendencies. cknowledgements. The uthors wish to thnk Peter Cook for providing the combintionl dtflow for the MC unit. 7. REFERENCES [] ROOKS, D. M., OSE, P., SCHUSTER, S. E., JCOSON, H., KUDV, P. N., UYUKTOSUNOGLU,., WELLMN, J., ZYUN, V., GUPT, M., ND COOK, P. W. Power-wre Microrchitecture: Design nd Modeling Chllenges for Next-Genertion Microprocessors. IEEE Micro 2, 6 (November/December 2), [2] CHNDRKSN,., ND RODERSEN, R. Low Power CMOS Design. IEEE Press., 998. [3] CORRELE JR.,. Overview of the power minimiztion techniques employed in the IM PowerPC 4xx embedded controllers. In Proc. Interntionl Symposium on Low Power Electronics nd Design (ISLPED) (995), pp [4] EFTHYMIOU,., ND GRSIDE, J. dptive Pipeline Depth Control for Processor Power-Mngement. In ICCD (22). [5] GOWN, M., IRO, L., ND JCKSON, D. Power Considertions in the Design of the lph 2264 Microprocessor. In DC (June 998), pp [6] KOPPNLIL, J., RMRKHYNI, P., DESI, S., VIDYNTHN,., ND ROTENERG, E. Cse for Dynmic Pipeline Scling. In CSES (October 22). [7] SHIMD, H., NDO, H., ND SHIMD, T. Pipeline Stge Unifiction: Low-Energy Consumption Technique for Future Mobile Processors. In ISLPED (ugust 23), pp [8] TIWRI, V., SINGH, D., RJGOPL, S., MEHT, G., PTEL, R., ND EZ, F. Reducing Power in High-Performnce Microprocessors. In DC (June 998), pp

Parallel Square and Cube Computations

Parallel Square and Cube Computations Prllel Squre nd Cube Computtions Albert A. Liddicot nd Michel J. Flynn Computer Systems Lbortory, Deprtment of Electricl Engineering Stnford University Gtes Building 5 Serr Mll, Stnford, CA 945, USA liddicot@stnford.edu

More information

EECS150 - Digital Design Lecture 23 - High-level Design and Optimization 3, Parallelism and Pipelining

EECS150 - Digital Design Lecture 23 - High-level Design and Optimization 3, Parallelism and Pipelining EECS150 - Digitl Design Lecture 23 - High-level Design nd Optimiztion 3, Prllelism nd Pipelining Nov 12, 2002 John Wwrzynek Fll 2002 EECS150 - Lec23-HL3 Pge 1 Prllelism Prllelism is the ct of doing more

More information

On Computation and Resource Management in Networked Embedded Systems

On Computation and Resource Management in Networked Embedded Systems On Computtion nd Resource Mngement in Networed Embedded Systems Soheil Ghisi Krlene Nguyen Elheh Bozorgzdeh Mjid Srrfzdeh Computer Science Deprtment University of Cliforni, Los Angeles, CA 90095 soheil,

More information

Accelerating 3D convolution using streaming architectures on FPGAs

Accelerating 3D convolution using streaming architectures on FPGAs Accelerting 3D convolution using streming rchitectures on FPGAs Hohun Fu, Robert G. Clpp, Oskr Mencer, nd Oliver Pell ABSTRACT We investigte FPGA rchitectures for ccelerting pplictions whose dominnt cost

More information

Engineer-to-Engineer Note

Engineer-to-Engineer Note Engineer-to-Engineer Note EE-245 Technicl notes on using Anlog Devices DSPs, processors nd development tools Contct our technicl support t dsp.support@nlog.com nd t dsptools.support@nlog.com Or visit our

More information

Overview. Network characteristics. Network architecture. Data dissemination. Network characteristics (cont d) Mobile computing and databases

Overview. Network characteristics. Network architecture. Data dissemination. Network characteristics (cont d) Mobile computing and databases Overview Mobile computing nd dtbses Generl issues in mobile dt mngement Dt dissemintion Dt consistency Loction dependent queries Interfces Detils of brodcst disks thlis klfigopoulos Network rchitecture

More information

2 Computing all Intersections of a Set of Segments Line Segment Intersection

2 Computing all Intersections of a Set of Segments Line Segment Intersection 15-451/651: Design & Anlysis of Algorithms Novemer 14, 2016 Lecture #21 Sweep-Line nd Segment Intersection lst chnged: Novemer 8, 2017 1 Preliminries The sweep-line prdigm is very powerful lgorithmic design

More information

ECEN 468 Advanced Logic Design Lecture 36: RTL Optimization

ECEN 468 Advanced Logic Design Lecture 36: RTL Optimization ECEN 468 Advnced Logic Design Lecture 36: RTL Optimiztion ECEN 468 Lecture 36 RTL Design Optimiztions nd Trdeoffs 6.5 While creting dtpth during RTL design, there re severl optimiztions nd trdeoffs, involving

More information

Before We Begin. Introduction to Spatial Domain Filtering. Introduction to Digital Image Processing. Overview (1): Administrative Details (1):

Before We Begin. Introduction to Spatial Domain Filtering. Introduction to Digital Image Processing. Overview (1): Administrative Details (1): Overview (): Before We Begin Administrtive detils Review some questions to consider Winter 2006 Imge Enhncement in the Sptil Domin: Bsics of Sptil Filtering, Smoothing Sptil Filters, Order Sttistics Filters

More information

Digital Design. Chapter 6: Optimizations and Tradeoffs

Digital Design. Chapter 6: Optimizations and Tradeoffs Digitl Design Chpter 6: Optimiztions nd Trdeoffs Slides to ccompny the tetbook Digitl Design, with RTL Design, VHDL, nd Verilog, 2nd Edition, by Frnk Vhid, John Wiley nd Sons Publishers, 2. http://www.ddvhid.com

More information

Systems I. Logic Design I. Topics Digital logic Logic gates Simple combinational logic circuits

Systems I. Logic Design I. Topics Digital logic Logic gates Simple combinational logic circuits Systems I Logic Design I Topics Digitl logic Logic gtes Simple comintionl logic circuits Simple C sttement.. C = + ; Wht pieces of hrdwre do you think you might need? Storge - for vlues,, C Computtion

More information

An Efficient Divide and Conquer Algorithm for Exact Hazard Free Logic Minimization

An Efficient Divide and Conquer Algorithm for Exact Hazard Free Logic Minimization An Efficient Divide nd Conquer Algorithm for Exct Hzrd Free Logic Minimiztion J.W.J.M. Rutten, M.R.C.M. Berkelr, C.A.J. vn Eijk, M.A.J. Kolsteren Eindhoven University of Technology Informtion nd Communiction

More information

Unit #9 : Definite Integral Properties, Fundamental Theorem of Calculus

Unit #9 : Definite Integral Properties, Fundamental Theorem of Calculus Unit #9 : Definite Integrl Properties, Fundmentl Theorem of Clculus Gols: Identify properties of definite integrls Define odd nd even functions, nd reltionship to integrl vlues Introduce the Fundmentl

More information

Enginner To Engineer Note

Enginner To Engineer Note Technicl Notes on using Anlog Devices DSP components nd development tools from the DSP Division Phone: (800) ANALOG-D, FAX: (781) 461-3010, EMAIL: dsp_pplictions@nlog.com, FTP: ftp.nlog.com Using n ADSP-2181

More information

What do all those bits mean now? Number Systems and Arithmetic. Introduction to Binary Numbers. Questions About Numbers

What do all those bits mean now? Number Systems and Arithmetic. Introduction to Binary Numbers. Questions About Numbers Wht do ll those bits men now? bits (...) Number Systems nd Arithmetic or Computers go to elementry school instruction R-formt I-formt... integer dt number text chrs... floting point signed unsigned single

More information

In the last lecture, we discussed how valid tokens may be specified by regular expressions.

In the last lecture, we discussed how valid tokens may be specified by regular expressions. LECTURE 5 Scnning SYNTAX ANALYSIS We know from our previous lectures tht the process of verifying the syntx of the progrm is performed in two stges: Scnning: Identifying nd verifying tokens in progrm.

More information

Questions About Numbers. Number Systems and Arithmetic. Introduction to Binary Numbers. Negative Numbers?

Questions About Numbers. Number Systems and Arithmetic. Introduction to Binary Numbers. Negative Numbers? Questions About Numbers Number Systems nd Arithmetic or Computers go to elementry school How do you represent negtive numbers? frctions? relly lrge numbers? relly smll numbers? How do you do rithmetic?

More information

II. THE ALGORITHM. A. Depth Map Processing

II. THE ALGORITHM. A. Depth Map Processing Lerning Plnr Geometric Scene Context Using Stereo Vision Pul G. Bumstrck, Bryn D. Brudevold, nd Pul D. Reynolds {pbumstrck,brynb,pulr2}@stnford.edu CS229 Finl Project Report December 15, 2006 Abstrct A

More information

Engineer To Engineer Note

Engineer To Engineer Note Engineer To Engineer Note EE-186 Technicl Notes on using Anlog Devices' DSP components nd development tools Contct our technicl support by phone: (800) ANALOG-D or e-mil: dsp.support@nlog.com Or visit

More information

UT1553B BCRT True Dual-port Memory Interface

UT1553B BCRT True Dual-port Memory Interface UTMC APPICATION NOTE UT553B BCRT True Dul-port Memory Interfce INTRODUCTION The UTMC UT553B BCRT is monolithic CMOS integrted circuit tht provides comprehensive MI-STD- 553B Bus Controller nd Remote Terminl

More information

Memory-Optimized Software Synthesis from Dataflow Program Graphs withlargesizedatasamples

Memory-Optimized Software Synthesis from Dataflow Program Graphs withlargesizedatasamples EURSIP Journl on pplied Signl Processing 2003:6, 54 529 c 2003 Hindwi Publishing orportion Memory-Optimized Softwre Synthesis from tflow Progrm Grphs withlrgesizetsmples Hyunok Oh The School of Electricl

More information

CSEP 573 Artificial Intelligence Winter 2016

CSEP 573 Artificial Intelligence Winter 2016 CSEP 573 Artificil Intelligence Winter 2016 Luke Zettlemoyer Problem Spces nd Serch slides from Dn Klein, Sturt Russell, Andrew Moore, Dn Weld, Pieter Abbeel, Ali Frhdi Outline Agents tht Pln Ahed Serch

More information

From Dependencies to Evaluation Strategies

From Dependencies to Evaluation Strategies From Dependencies to Evlution Strtegies Possile strtegies: 1 let the user define the evlution order 2 utomtic strtegy sed on the dependencies: use locl dependencies to determine which ttriutes to compute

More information

Engineer-to-Engineer Note

Engineer-to-Engineer Note Engineer-to-Engineer Note EE-204 Technicl notes on using Anlog Devices DSPs, processors nd development tools Visit our Web resources http://www.nlog.com/ee-notes nd http://www.nlog.com/processors or e-mil

More information

File Manager Quick Reference Guide. June Prepared for the Mayo Clinic Enterprise Kahua Deployment

File Manager Quick Reference Guide. June Prepared for the Mayo Clinic Enterprise Kahua Deployment File Mnger Quick Reference Guide June 2018 Prepred for the Myo Clinic Enterprise Khu Deployment NVIGTION IN FILE MNGER To nvigte in File Mnger, users will mke use of the left pne to nvigte nd further pnes

More information

Fall 2018 Midterm 1 October 11, ˆ You may not ask questions about the exam except for language clarifications.

Fall 2018 Midterm 1 October 11, ˆ You may not ask questions about the exam except for language clarifications. 15-112 Fll 2018 Midterm 1 October 11, 2018 Nme: Andrew ID: Recittion Section: ˆ You my not use ny books, notes, extr pper, or electronic devices during this exm. There should be nothing on your desk or

More information

Complete Coverage Path Planning of Mobile Robot Based on Dynamic Programming Algorithm Peng Zhou, Zhong-min Wang, Zhen-nan Li, Yang Li

Complete Coverage Path Planning of Mobile Robot Based on Dynamic Programming Algorithm Peng Zhou, Zhong-min Wang, Zhen-nan Li, Yang Li 2nd Interntionl Conference on Electronic & Mechnicl Engineering nd Informtion Technology (EMEIT-212) Complete Coverge Pth Plnning of Mobile Robot Bsed on Dynmic Progrmming Algorithm Peng Zhou, Zhong-min

More information

Engineer To Engineer Note

Engineer To Engineer Note Engineer To Engineer Note EE-169 Technicl Notes on using Anlog Devices' DSP components nd development tools Contct our technicl support by phone: (800) ANALOG-D or e-mil: dsp.support@nlog.com Or visit

More information

12-B FRACTIONS AND DECIMALS

12-B FRACTIONS AND DECIMALS -B Frctions nd Decimls. () If ll four integers were negtive, their product would be positive, nd so could not equl one of them. If ll four integers were positive, their product would be much greter thn

More information

MA1008. Calculus and Linear Algebra for Engineers. Course Notes for Section B. Stephen Wills. Department of Mathematics. University College Cork

MA1008. Calculus and Linear Algebra for Engineers. Course Notes for Section B. Stephen Wills. Department of Mathematics. University College Cork MA1008 Clculus nd Liner Algebr for Engineers Course Notes for Section B Stephen Wills Deprtment of Mthemtics University College Cork s.wills@ucc.ie http://euclid.ucc.ie/pges/stff/wills/teching/m1008/ma1008.html

More information

Computer Arithmetic Logical, Integer Addition & Subtraction Chapter

Computer Arithmetic Logical, Integer Addition & Subtraction Chapter Computer Arithmetic Logicl, Integer Addition & Sutrction Chpter 3.-3.3 3.3 EEC7 FQ 25 MIPS Integer Representtion -it signed integers,, e.g., for numeric opertions 2 s s complement: one representtion for

More information

SoC Architecture Design Approaches

SoC Architecture Design Approaches Wireless Informtion Trnsmission System Lb. SoC Architecture Design Approches Hung-Chih Ching Institute of Communictions Engineering Ntionl Sun Yt-sen University IP-Bsed System IC Design Block-Bsed Architecture

More information

UNIT 11. Query Optimization

UNIT 11. Query Optimization UNIT Query Optimiztion Contents Introduction to Query Optimiztion 2 The Optimiztion Process: An Overview 3 Optimiztion in System R 4 Optimiztion in INGRES 5 Implementing the Join Opertors Wei-Png Yng,

More information

Small Business Networking

Small Business Networking Why network is n essentil productivity tool for ny smll business Effective technology is essentil for smll businesses looking to increse the productivity of their people nd processes. Introducing technology

More information

Lecture 10 Evolutionary Computation: Evolution strategies and genetic programming

Lecture 10 Evolutionary Computation: Evolution strategies and genetic programming Lecture 10 Evolutionry Computtion: Evolution strtegies nd genetic progrmming Evolution strtegies Genetic progrmming Summry Negnevitsky, Person Eduction, 2011 1 Evolution Strtegies Another pproch to simulting

More information

On the Detection of Step Edges in Algorithms Based on Gradient Vector Analysis

On the Detection of Step Edges in Algorithms Based on Gradient Vector Analysis On the Detection of Step Edges in Algorithms Bsed on Grdient Vector Anlysis A. Lrr6, E. Montseny Computer Engineering Dept. Universitt Rovir i Virgili Crreter de Slou sin 43006 Trrgon, Spin Emil: lrre@etse.urv.es

More information

Small Business Networking

Small Business Networking Why network is n essentil productivity tool for ny smll business Effective technology is essentil for smll businesses looking to increse the productivity of their people nd business. Introducing technology

More information

What do all those bits mean now? Number Systems and Arithmetic. Introduction to Binary Numbers. Questions About Numbers

What do all those bits mean now? Number Systems and Arithmetic. Introduction to Binary Numbers. Questions About Numbers Wht do ll those bits men now? bits (...) Number Systems nd Arithmetic or Computers go to elementry school instruction R-formt I-formt... integer dt number text chrs... floting point signed unsigned single

More information

Small Business Networking

Small Business Networking Why network is n essentil productivity tool for ny smll business Effective technology is essentil for smll businesses looking to increse the productivity of their people nd business. Introducing technology

More information

Section 10.4 Hyperbolas

Section 10.4 Hyperbolas 66 Section 10.4 Hyperbols Objective : Definition of hyperbol & hyperbols centered t (0, 0). The third type of conic we will study is the hyperbol. It is defined in the sme mnner tht we defined the prbol

More information

Midterm 2 Sample solution

Midterm 2 Sample solution Nme: Instructions Midterm 2 Smple solution CMSC 430 Introduction to Compilers Fll 2012 November 28, 2012 This exm contins 9 pges, including this one. Mke sure you hve ll the pges. Write your nme on the

More information

Presentation Martin Randers

Presentation Martin Randers Presenttion Mrtin Rnders Outline Introduction Algorithms Implementtion nd experiments Memory consumption Summry Introduction Introduction Evolution of species cn e modelled in trees Trees consist of nodes

More information

Engineer To Engineer Note

Engineer To Engineer Note Engineer To Engineer Note EE-208 Technicl Notes on using Anlog Devices' DSP components nd development tools Contct our technicl support by phone: (800) ANALOG-D or e-mil: dsp.support@nlog.com Or visit

More information

Data Flow on a Queue Machine. Bruno R. Preiss. Copyright (c) 1987 by Bruno R. Preiss, P.Eng. All rights reserved.

Data Flow on a Queue Machine. Bruno R. Preiss. Copyright (c) 1987 by Bruno R. Preiss, P.Eng. All rights reserved. Dt Flow on Queue Mchine Bruno R. Preiss 2 Outline Genesis of dt-flow rchitectures Sttic vs. dynmic dt-flow rchitectures Pseudo-sttic dt-flow execution model Some dt-flow mchines Simple queue mchine Prioritized

More information

Small Business Networking

Small Business Networking Why network is n essentil productivity tool for ny smll business Effective technology is essentil for smll businesses looking to increse the productivity of their people nd business. Introducing technology

More information

Transparent neutral-element elimination in MPI reduction operations

Transparent neutral-element elimination in MPI reduction operations Trnsprent neutrl-element elimintion in MPI reduction opertions Jesper Lrsson Träff Deprtment of Scientific Computing University of Vienn Disclimer Exploiting repetition nd sprsity in input for reducing

More information

10.5 Graphing Quadratic Functions

10.5 Graphing Quadratic Functions 0.5 Grphing Qudrtic Functions Now tht we cn solve qudrtic equtions, we wnt to lern how to grph the function ssocited with the qudrtic eqution. We cll this the qudrtic function. Grphs of Qudrtic Functions

More information

A New Learning Algorithm for the MAXQ Hierarchical Reinforcement Learning Method

A New Learning Algorithm for the MAXQ Hierarchical Reinforcement Learning Method A New Lerning Algorithm for the MAXQ Hierrchicl Reinforcement Lerning Method Frzneh Mirzzdeh 1, Bbk Behsz 2, nd Hmid Beigy 1 1 Deprtment of Computer Engineering, Shrif University of Technology, Tehrn,

More information

1 Introduction

1 Introduction Published in IET Computers & Digitl Techniques Received on 6th July 2006 Revised on 21st September 2007 ISSN 1751-8601 Hrdwre rchitecture for high-speed rel-time dynmic progrmming pplictions B. Mtthews

More information

Tree Structured Symmetrical Systems of Linear Equations and their Graphical Solution

Tree Structured Symmetrical Systems of Linear Equations and their Graphical Solution Proceedings of the World Congress on Engineering nd Computer Science 4 Vol I WCECS 4, -4 October, 4, Sn Frncisco, USA Tree Structured Symmetricl Systems of Liner Equtions nd their Grphicl Solution Jime

More information

Digital Design using HDLs EE 4755 Final Examination

Digital Design using HDLs EE 4755 Final Examination Nme Solution Digitl Design using HDLs EE 4755 Finl Exmintion Thursdy, 8 Decemer 6 :3-4:3 CST Alis The Hottest Plce in Hell Prolem Prolem Prolem 3 Prolem 4 Prolem 5 Prolem 6 Exm Totl (3 pts) ( pts) (5 pts)

More information

A Heuristic Approach for Discovering Reference Models by Mining Process Model Variants

A Heuristic Approach for Discovering Reference Models by Mining Process Model Variants A Heuristic Approch for Discovering Reference Models by Mining Process Model Vrints Chen Li 1, Mnfred Reichert 2, nd Andres Wombcher 3 1 Informtion System Group, University of Twente, The Netherlnds lic@cs.utwente.nl

More information

c360 Add-On Solutions

c360 Add-On Solutions c360 Add-On Solutions Functionlity Dynmics CRM 2011 c360 Record Editor Reltionship Explorer Multi-Field Serch Alerts Console c360 Core Productivity Pck "Does your tem resist using CRM becuse updting dt

More information

Small Business Networking

Small Business Networking Why network is n essentil productivity tool for ny smll business Effective technology is essentil for smll businesses looking to increse the productivity of their people nd processes. Introducing technology

More information

Allocator Basics. Dynamic Memory Allocation in the Heap (malloc and free) Allocator Goals: malloc/free. Internal Fragmentation

Allocator Basics. Dynamic Memory Allocation in the Heap (malloc and free) Allocator Goals: malloc/free. Internal Fragmentation Alloctor Bsics Dynmic Memory Alloction in the Hep (mlloc nd free) Pges too corse-grined for llocting individul objects. Insted: flexible-sized, word-ligned blocks. Allocted block (4 words) Free block (3

More information

1. SEQUENCES INVOLVING EXPONENTIAL GROWTH (GEOMETRIC SEQUENCES)

1. SEQUENCES INVOLVING EXPONENTIAL GROWTH (GEOMETRIC SEQUENCES) Numbers nd Opertions, Algebr, nd Functions 45. SEQUENCES INVOLVING EXPONENTIAL GROWTH (GEOMETRIC SEQUENCES) In sequence of terms involving eponentil growth, which the testing service lso clls geometric

More information

Small Business Networking

Small Business Networking Why network is n essentil productivity tool for ny smll business Effective technology is essentil for smll businesses looking to increse the productivity of their people nd processes. Introducing technology

More information

Slides for Data Mining by I. H. Witten and E. Frank

Slides for Data Mining by I. H. Witten and E. Frank Slides for Dt Mining y I. H. Witten nd E. Frnk Simplicity first Simple lgorithms often work very well! There re mny kinds of simple structure, eg: One ttriute does ll the work All ttriutes contriute eqully

More information

Pipeline Example: Cycle 1. Pipeline Example: Cycle 2. Pipeline Example: Cycle 4. Pipeline Example: Cycle 3. 3 instructions. 3 instructions.

Pipeline Example: Cycle 1. Pipeline Example: Cycle 2. Pipeline Example: Cycle 4. Pipeline Example: Cycle 3. 3 instructions. 3 instructions. ipeline Exmple: Cycle 1 ipeline Exmple: Cycle X X/ /W X X/ /W $3,$,$1 lw $,0($5) $3,$,$1 3 instructions 8 9 ipeline Exmple: Cycle 3 ipeline Exmple: Cycle X X/ /W X X/ /W sw $6,($7) lw $,0($5) $3,$,$1 sw

More information

ECE 468/573 Midterm 1 September 28, 2012

ECE 468/573 Midterm 1 September 28, 2012 ECE 468/573 Midterm 1 September 28, 2012 Nme:! Purdue emil:! Plese sign the following: I ffirm tht the nswers given on this test re mine nd mine lone. I did not receive help from ny person or mteril (other

More information

Small Business Networking

Small Business Networking Why network is n essentil productivity tool for ny smll business Effective technology is essentil for smll businesses looking to increse the productivity of their people nd business. Introducing technology

More information

Distributed Systems Principles and Paradigms

Distributed Systems Principles and Paradigms Distriuted Systems Principles nd Prdigms Chpter 11 (version April 7, 2008) Mrten vn Steen Vrije Universiteit Amsterdm, Fculty of Science Dept. Mthemtics nd Computer Science Room R4.20. Tel: (020) 598 7784

More information

Engineer-to-Engineer Note

Engineer-to-Engineer Note Engineer-to-Engineer Note EE-232 Technicl notes on using Anlog Devices DSPs, processors nd development tools Contct our technicl support t dsp.support@nlog.com nd t dsptools.support@nlog.com Or visit our

More information

Geometric transformations

Geometric transformations Geometric trnsformtions Computer Grphics Some slides re bsed on Shy Shlom slides from TAU mn n n m m T A,,,,,, 2 1 2 22 12 1 21 11 Rows become columns nd columns become rows nm n n m m A,,,,,, 1 1 2 22

More information

Unit 5 Vocabulary. A function is a special relationship where each input has a single output.

Unit 5 Vocabulary. A function is a special relationship where each input has a single output. MODULE 3 Terms Definition Picture/Exmple/Nottion 1 Function Nottion Function nottion is n efficient nd effective wy to write functions of ll types. This nottion llows you to identify the input vlue with

More information

Mobile IP route optimization method for a carrier-scale IP network

Mobile IP route optimization method for a carrier-scale IP network Moile IP route optimiztion method for crrier-scle IP network Tkeshi Ihr, Hiroyuki Ohnishi, nd Ysushi Tkgi NTT Network Service Systems Lortories 3-9-11 Midori-cho, Musshino-shi, Tokyo 180-8585, Jpn Phone:

More information

Digital Signal Processing: A Hardware-Based Approach

Digital Signal Processing: A Hardware-Based Approach Digitl Signl Processing: A Hrdwre-Bsed Approch Roert Esposito Electricl nd Computer Engineering Temple University troduction Teching Digitl Signl Processing (DSP) hs included the utilition of simultion

More information

Functor (1A) Young Won Lim 8/2/17

Functor (1A) Young Won Lim 8/2/17 Copyright (c) 2016-2017 Young W. Lim. Permission is grnted to copy, distribute nd/or modify this document under the terms of the GNU Free Documenttion License, Version 1.2 or ny lter version published

More information

Misrepresentation of Preferences

Misrepresentation of Preferences Misrepresenttion of Preferences Gicomo Bonnno Deprtment of Economics, University of Cliforni, Dvis, USA gfbonnno@ucdvis.edu Socil choice functions Arrow s theorem sys tht it is not possible to extrct from

More information

Small Business Networking

Small Business Networking Why network is n essentil productivity tool for ny smll business Effective technology is essentil for smll businesses looking to increse the productivity of their people nd processes. Introducing technology

More information

Graphs with at most two trees in a forest building process

Graphs with at most two trees in a forest building process Grphs with t most two trees in forest uilding process rxiv:802.0533v [mth.co] 4 Fe 208 Steve Butler Mis Hmnk Mrie Hrdt Astrct Given grph, we cn form spnning forest y first sorting the edges in some order,

More information

A Priority-based Distributed Call Admission Protocol for Multi-hop Wireless Ad hoc Networks

A Priority-based Distributed Call Admission Protocol for Multi-hop Wireless Ad hoc Networks A Priority-bsed Distributed Cll Admission Protocol for Multi-hop Wireless Ad hoc Networks un Sun Elizbeth M. Belding-Royer Deprtment of Computer Science University of Cliforni, Snt Brbr suny, ebelding

More information

CSCI 446: Artificial Intelligence

CSCI 446: Artificial Intelligence CSCI 446: Artificil Intelligence Serch Instructor: Michele Vn Dyne [These slides were creted by Dn Klein nd Pieter Abbeel for CS188 Intro to AI t UC Berkeley. All CS188 mterils re vilble t http://i.berkeley.edu.]

More information

Solving Problems by Searching. CS 486/686: Introduction to Artificial Intelligence Winter 2016

Solving Problems by Searching. CS 486/686: Introduction to Artificial Intelligence Winter 2016 Solving Prolems y Serching CS 486/686: Introduction to Artificil Intelligence Winter 2016 1 Introduction Serch ws one of the first topics studied in AI - Newell nd Simon (1961) Generl Prolem Solver Centrl

More information

Agilent Mass Hunter Software

Agilent Mass Hunter Software Agilent Mss Hunter Softwre Quick Strt Guide Use this guide to get strted with the Mss Hunter softwre. Wht is Mss Hunter Softwre? Mss Hunter is n integrl prt of Agilent TOF softwre (version A.02.00). Mss

More information

Address Register Assignment for Reducing Code Size

Address Register Assignment for Reducing Code Size Address Register Assignment for Reducing Code Size M. Kndemir 1, M.J. Irwin 1, G. Chen 1, nd J. Rmnujm 2 1 CSE Deprtment Pennsylvni Stte University University Prk, PA 16802 {kndemir,mji,guilchen}@cse.psu.edu

More information

Functor (1A) Young Won Lim 10/5/17

Functor (1A) Young Won Lim 10/5/17 Copyright (c) 2016-2017 Young W. Lim. Permission is grnted to copy, distribute nd/or modify this document under the terms of the GNU Free Documenttion License, Version 1.2 or ny lter version published

More information

Today. Search Problems. Uninformed Search Methods. Depth-First Search Breadth-First Search Uniform-Cost Search

Today. Search Problems. Uninformed Search Methods. Depth-First Search Breadth-First Search Uniform-Cost Search Uninformed Serch [These slides were creted by Dn Klein nd Pieter Abbeel for CS188 Intro to AI t UC Berkeley. All CS188 mterils re vilble t http://i.berkeley.edu.] Tody Serch Problems Uninformed Serch Methods

More information

5 Regular 4-Sided Composition

5 Regular 4-Sided Composition Xilinx-Lv User Guide 5 Regulr 4-Sided Composition This tutoril shows how regulr circuits with 4-sided elements cn be described in Lv. The type of regulr circuits tht re discussed in this tutoril re those

More information

6.2 Volumes of Revolution: The Disk Method

6.2 Volumes of Revolution: The Disk Method mth ppliction: volumes by disks: volume prt ii 6 6 Volumes of Revolution: The Disk Method One of the simplest pplictions of integrtion (Theorem 6) nd the ccumultion process is to determine so-clled volumes

More information

P(r)dr = probability of generating a random number in the interval dr near r. For this probability idea to make sense we must have

P(r)dr = probability of generating a random number in the interval dr near r. For this probability idea to make sense we must have Rndom Numers nd Monte Crlo Methods Rndom Numer Methods The integrtion methods discussed so fr ll re sed upon mking polynomil pproximtions to the integrnd. Another clss of numericl methods relies upon using

More information

CS 268: IP Multicast Routing

CS 268: IP Multicast Routing Motivtion CS 268: IP Multicst Routing Ion Stoic April 5, 2004 Mny pplictions requires one-to-mny communiction - E.g., video/udio conferencing, news dissemintion, file updtes, etc. Using unicst to replicte

More information

Today. CS 188: Artificial Intelligence Fall Recap: Search. Example: Pancake Problem. Example: Pancake Problem. General Tree Search.

Today. CS 188: Artificial Intelligence Fall Recap: Search. Example: Pancake Problem. Example: Pancake Problem. General Tree Search. CS 88: Artificil Intelligence Fll 00 Lecture : A* Serch 9//00 A* Serch rph Serch Tody Heuristic Design Dn Klein UC Berkeley Multiple slides from Sturt Russell or Andrew Moore Recp: Serch Exmple: Pncke

More information

Dynamic Tone Mapping with Head-Mounted Displays

Dynamic Tone Mapping with Head-Mounted Displays Dynmic Tone Mpping with Hed-Mounted Displys Mtt Yu Deprtment of Electricl Engineering, Stnford University Abstrct The rel world consists of mny scenes which contin high dynmic rnge. While modern cmers

More information

Performance Evaluation of Dynamic Reconfiguration in High-Speed Local Area Networks

Performance Evaluation of Dynamic Reconfiguration in High-Speed Local Area Networks Performnce Evlution of Dynmic Reconfigurtion in High-Speed Locl Are Networks Rfel Csdo, Aurelio Bermúdez, Frncisco J. Quiles, JoséL.Sánchez Depto. de Informátic Universidd de Cstill-L Mnch 271- Albcete,

More information

Fault injection attacks on cryptographic devices and countermeasures Part 2

Fault injection attacks on cryptographic devices and countermeasures Part 2 Fult injection ttcks on cryptogrphic devices nd countermesures Prt Isrel Koren Deprtment of Electricl nd Computer Engineering University of Msschusetts Amherst, MA Countermesures - Exmples Must first detect

More information

OSPF WEIGHT SETTING OPTIMIZATION FOR SINGLE LINK FAILURES

OSPF WEIGHT SETTING OPTIMIZATION FOR SINGLE LINK FAILURES OSP WEIGH SEIG OPIMIZAIO OR SIGLE LIK AILURES Mohmmed H. Sqlli, Sdiq M. Sit, nd Syed Asdullh Computer Engineering Deprtment King hd University of Petroleum & Minerls Dhhrn 31261, Sudi Arbi {sqlli,sdiq,ssd}@kfupm.edu.s

More information

Improving the Connectivity of Heterogeneous Multi-Hop Wireless Networks

Improving the Connectivity of Heterogeneous Multi-Hop Wireless Networks 1569363897 1 Improving the Connectivity of Heterogeneous Multi-Hop Wireless Networks Te-Hoon Kim, Dvid Tipper, nd Prshnt Krishnmurthy Grdute Networking nd Telecommunictions Progrm University of Pittsburgh,

More information

Fig.1. Let a source of monochromatic light be incident on a slit of finite width a, as shown in Fig. 1.

Fig.1. Let a source of monochromatic light be incident on a slit of finite width a, as shown in Fig. 1. Answer on Question #5692, Physics, Optics Stte slient fetures of single slit Frunhofer diffrction pttern. The slit is verticl nd illuminted by point source. Also, obtin n expression for intensity distribution

More information

Replicating Web Applications On-Demand

Replicating Web Applications On-Demand Replicting Web Applictions On-Demnd Swminthn Sivsubrmnin Guillume Pierre Mrten vn Steen Dept. of Computer Science, Vrije Universiteit, Amsterdm {swmi,gpierre,steen}@cs.vu.nl Abstrct Mny Web-bsed commercil

More information

Stained Glass Design. Teaching Goals:

Stained Glass Design. Teaching Goals: Stined Glss Design Time required 45-90 minutes Teching Gols: 1. Students pply grphic methods to design vrious shpes on the plne.. Students pply geometric trnsformtions of grphs of functions in order to

More information

PNC NC code PROGRAMMER'S MANUAL

PNC NC code PROGRAMMER'S MANUAL PNC-3200 NC code PROGRAMMER'S MANUAL Thnk you very much for purchsing the PNC-3200. To ensure correct nd sfe usge with full understnding of this product's performnce, plese be sure to red through this

More information

9 Graph Cutting Procedures

9 Graph Cutting Procedures 9 Grph Cutting Procedures Lst clss we begn looking t how to embed rbitrry metrics into distributions of trees, nd proved the following theorem due to Brtl (1996): Theorem 9.1 (Brtl (1996)) Given metric

More information

Preserving Constraints for Aggregation Relationship Type Update in XML Document

Preserving Constraints for Aggregation Relationship Type Update in XML Document Preserving Constrints for Aggregtion Reltionship Type Updte in XML Document Eric Prdede 1, J. Wenny Rhyu 1, nd Dvid Tnir 2 1 Deprtment of Computer Science nd Computer Engineering, L Trobe University, Bundoor

More information

Epson Projector Content Manager Operation Guide

Epson Projector Content Manager Operation Guide Epson Projector Content Mnger Opertion Guide Contents 2 Introduction to the Epson Projector Content Mnger Softwre 3 Epson Projector Content Mnger Fetures... 4 Setting Up the Softwre for the First Time

More information

MATH 2530: WORKSHEET 7. x 2 y dz dy dx =

MATH 2530: WORKSHEET 7. x 2 y dz dy dx = MATH 253: WORKSHT 7 () Wrm-up: () Review: polr coordintes, integrls involving polr coordintes, triple Riemnn sums, triple integrls, the pplictions of triple integrls (especilly to volume), nd cylindricl

More information

Small Business Networking

Small Business Networking Why network is n essentil productivity tool for ny smll business Effective technology is essentil for smll businesses looking to increse the productivity of their people nd processes. Introducing technology

More information

If you are at the university, either physically or via the VPN, you can download the chapters of this book as PDFs.

If you are at the university, either physically or via the VPN, you can download the chapters of this book as PDFs. Lecture 5 Wlks, Trils, Pths nd Connectedness Reding: Some of the mteril in this lecture comes from Section 1.2 of Dieter Jungnickel (2008), Grphs, Networks nd Algorithms, 3rd edition, which is ville online

More information

OUTPUT DELIVERY SYSTEM

OUTPUT DELIVERY SYSTEM Differences in ODS formtting for HTML with Proc Print nd Proc Report Lur L. M. Thornton, USDA-ARS, Animl Improvement Progrms Lortory, Beltsville, MD ABSTRACT While Proc Print is terrific tool for dt checking

More information

CS311H: Discrete Mathematics. Graph Theory IV. A Non-planar Graph. Regions of a Planar Graph. Euler s Formula. Instructor: Işıl Dillig

CS311H: Discrete Mathematics. Graph Theory IV. A Non-planar Graph. Regions of a Planar Graph. Euler s Formula. Instructor: Işıl Dillig CS311H: Discrete Mthemtics Grph Theory IV Instructor: Işıl Dillig Instructor: Işıl Dillig, CS311H: Discrete Mthemtics Grph Theory IV 1/25 A Non-plnr Grph Regions of Plnr Grph The plnr representtion of

More information