Power-Performance Trade-offs for Energy-Efficient Architectures: A Quantitative Study

Size: px
Start display at page:

Download "Power-Performance Trade-offs for Energy-Efficient Architectures: A Quantitative Study"

Transcription

1 In the Proc. of the Intl. Conf. on Computer Design (ICCD-02), Frieburg, March 2002 Power-Performance Trae-offs for Energy-Efficient Architectures: A Quantitative Stuy Hongbo Yang R. Govinarajan Guang R. Gao Kevin B. Theobal Electrical an Computer Engineering Supercomputer Eucation & Research Centre University of Delaware Dept. of Computer Science & Automation Newark, DE Inian Institute of Science Delaware, U.S.A. Bangalore, , INDIA govin@csa.iisc.ernet.in Abstract The rastic increase in power consumption by moern processors emphasizes the nee for power-performance trae-offs in architecture esign space exploration an compiler optimizations. This paper reports a quantitative stuy on the power-performance trae-offs in software pipeline scheules for an Itanium-like EPIC architecture with ual-spee pipelines, in which functional units are partitione into fast ones an slow ones. We have evelope an integer linear programming formulation to capture the power-performance traeoffs for software pipeline loops. The propose integer linear programming formulation an its solution metho have been implemente an teste on a set of SPEC2000 benchmarks. The results are compare with an Itanium-like architecture(baseline) in which there are four functional units (FUs) an all of them are fast units. Our quantitative stuy reveals that by introucing a few slow FUs in place of fast FUs in the baseline architecture, the total energy consume by FUs can be consierably reuce. When 2 out of 4 FUs are set as slow, the total energy consume by FUs is reuce by up to 31.1% (with an average reuction of 25.2%) compare with the baseline configuration, while the performance egraation cause by using slow FUs is small. If performance eman is less critical, then energy reuction of up to 40.3% compare with the baseline configuration can be achieve. 1. Introuction The past ecae has seen a tremenous increase in the performance of general purpose microprocessors. However, This work is supporte by DARPA through contract number AFRL F this higher performance is often accompanie by an unesirable, sometime excessive, power consumption. This is preicte soon to become a limiting factor in highperformance processor esign [2]. Consequently, architects must consier power-performance traeoffs when exploring the processor esign space. This paper eals with low power architectures, compiler techniques, an their interplay. Recent research in low power esign proposes architectures with components operating at ifferent pipeline spees [12]. In such processors, some functional units (FUs) operate at full CPU spee ( fast FUs) while others run at a slower spee ( slow FUs), typically half of the full spee. Slow FUs can run at lower voltages, which lea to a better-than-linear energy reuction per operation. The rationale behin such a esign is that scheule slack is present for some instructions, i.e., they are not on the critical path, so prolonging their execution time shoul not hurt performance. A limit stuy [3] emonstrates that more than 75% of execution cycles have at least one instruction that has slack. By issuing instructions on the critical path to fast pipelines an those which are off the critical path to slower functional units, significant power an energy reuction can be achieve [10, 21]. This paper reports a quantitative stuy of powerperformance traeoffs in the esign space by exploring the interplay between low-power architecture features an compiler optimizations. Our stuy concentrates on software pipelining [11, 15], a compile-time instruction scheuling technique for loops. In our prior work, performanceoriente software pipelining uner given timing an resource constraints has been formulate as an integer linear programming problem [1, 5, 6] an was successfully use in evaluating the software pipelining algorithm implemente in a prouction-quality compiler [17]. However in the context of low power, effective software pipelining algorithm

2 s2 s4 + + s0 s1 s3 + + * Figure 1. Motivating Example which strives for both optimal performance an minimal energy is still unexplore. The problem that we are aressing in this paper is: Given a loop an a machine configuration with slow an fast FUs, fin a rate-optimal software pipeline scheule which consumes minimal energy. We use the term rate-optimal scheule to imply that no other scheule for the loop can have a lower initiation interval(ii) [15, 11, 14] for the machine configuration. The term minimal energy scheule means that no other scheule for the loop can have a lower energy consumption with given II for the given architecture. This paper formulates the above problem as an ILP problem an explores both the architecture an compiler sie of esign space using the ILP formulation an its solution metho. In particular, this stuy explores: [Architecture sie:] In orer to achieve significant power reuction without incurring performance egraation, how many of the FUs in an architecture can be slow FUs? By increasing the number of slow FUs, how oes the application performance egrae? [Compiler sie:] To what extent can the compiler exploit this knowlege of slow an fast FUs in the architecture? Further, in such architectures, to what extent can compilers trae-off performance for energy reuction? 1.1. Example We illustrate the benefits of slow FUs with a simple example. Figure 1 shows a ata epenence graph (DDG) for a loop boy, with a loop-carrie epenence (from one iteration of the loop to the next) from s5 to s4. Suppose we want to run this loop on an architecture a1 with 3 Integer A units an 2 Integer Multiply units. All FUs are fast an fully pipeline, an their latencies are 1 cycle for A an 3 for Mult. A rate-optimal scheule for this DDG with an II of 2 is shown in Figure 2. The repetitive kernel in cycles 4 an 5 is shae. We can see that there is some slack in the scheule, an not all FUs are fully utilize. If one A FU s5 + Iteration Time Steps s0 s1,s2 s3 s4 s5 1 s0 s1,s2 s3 s4 s5 2 s0 s1,s2 s3 s4 s5 Figure 2. A Scheule for Architecture a1 Iteration Time Steps s0,s2 s1 s3 s4 s5 1 s0,s2 s1 s3 s4 2 s0,s2 s1 s3 3 s0,s2 s1 s3 4 s0,s2 s1 Figure 3. A Scheule for Architecture a2 an one Mult FU in a1 were replace with slow FUs operating at half the frequency, we coul still achieve the same II with this moifie architecture a2, as shown in Figure 3. In this scheule, instructions s2 an s3 are scheule in the slow A an Mult FUs; these are shown in italics. Although the kernel appears later, i.e., the prologue is longer, the scheule is still optimal with. Architecture a2 may save power relative to a1, without slowing own the loop. On the other han, if two of the A units were slow (architecture a3), the scheule in Figure 3 woul not work. Each loop iteration woul require 5 aitions, but each of the two slow units can perform only one aition every 2 cycles, so that all three A units together can perform only 4 aitions every 2 cycles. The best II that we can achieve for this DDG woul be 3. This example shows that by varying the number of fast an slow FUs, the architecture can be mae more power efficient, but at some point this efficiency will lea to a performance penalty Synopsis Given an architecture with a fixe number of FUs in each resource class, it is interesting to fin out how many of these FUs in each class coul be operate at a slower clock rate, an hence at a reuce power consumption, an what is the egraation in performance, if any. In this paper, we propose an integer linear programming formulation for the above problem base on our earlier work. We apply our approach on SPEC2000 integer benchmarks to evaluate ifferent configurations (with ifferent number of slow an fast FUs) of Itanium processor from both the performance an power angles. Also we stuy the aitional power savings that can be obtaine when the compiler traes performance for energy. Our results show that by introucing two slow FUs in the architecture, in the place of fast FUs, an

3 b average energy savings of 25.2% can be obtaine while the performance egraation is within 6.2% on the average. The rest of the paper is organize as follows. The next section presents an elegant ILP formulation for poweraware software pipelining base on our previous work. In Section 3 we elaborate our experimental framework. Section 4 eals with the experimental results. A iscussion on relate work an concluing remarks are provie respectively in Sections 5 an Integer Linear Programming Formulation Let the number of noes in the DDG be an let be the initiation interval. In this paper we consier only repetitive scheules or moulo scheuling. The software pipelining methos attempts to fin a scheule for each value of initiation interval, starting from the minimum initiation interval (MII) [15, 11, 14]. The methos stops as soon as a scheule is foun for the first greater than or equal to MII. In moulo scheuling, the scheule time of operation in the th iteration is given by where is the scheule time of operation in the first iteration. We use an vector! ""#$&%'%&%"(*)#,+.- to represent the scheule time of the operations in the first iteration. We use an /* matrix 0 to represent the repetitive pattern of instructions in the moulo scheule. The matrix where if an only if noe is scheule at time step in the moulo scheule. It is easy to see that 7&8 9 ;:=< > 5? if mo We 6AB!,AC#D'%&%&%&,A!(E)F#,+ -, where G 9ÏH CJ? Further an 0 are relate by the following equation. K LNMOPRQSUTWV XY<:=<"MMM<ZN[U:]\^S (1) Since each instruction is scheule in the repetitive pattern exactly once, the following equation must be satisfie for a legal scheule: a`cb 8.e9f 7&8 9 ;: for all gch VXY<Zij[U:]\ (2) For resources that operate on a single spee, that is those which are not partitione to fast an slow ones, e.g., the issue unit, the resource constraint can be specifie using the following simple inequality..k$lamon"p 7&8 rqs n for all h VXY<t&N[U:]\ (3) where uwv is the number of resources of type x an y{z x} represent the set of instructions in the loop that use resource type x. Next we formulate the resource constraints for resources types which have slow an fast FUs. For an instructions which goes to these resource types x, we also nee to know whether the instruction is scheule on slow or fast FU. We use two sets of variables ~ 354 an 354 for this purpose: ~ 354 if instruction is scheule on fast FU an 2354 ; likewise D354 if instruction is scheule on slow FU an Since the resource assignment for instruction can be either fast FU or slow FU but not both, the following equation hols: 7&8 9 ƒ 8 w 8 for all h VXY<5 [:]\.< an for all g h VXY<Zi [:]\ (4) If there are u v4 $ˆ ]3 fast FUs an u v4 ]ŠŒ slow FUs in resource type x, then the resource constraint for software pipeline scheule can be enforce using the inequality ƒ 8 qs n] a = 8 for all k$l,mžnp h VXY<t [U:]\ an for all (5) 8 rqs n] 5 for all.kdl,mžnp h VXY<t [U:]\ an for all (6) Since the time steps in our scheule is base on the fast clock, slow FUs (operating at half the clock frequency) have a structural hazar, an can have instructions scheule on them only once every two cycles [19]. To reflect this, we a the conition that if z 354 Y an z Y, then z C 3 š# 54 shoul also be 1, where œ represents z Y ŸžLD. This can be expresse in the form of a linear constraint as: 8! 7'8 {[ m 8 p U? (7) Note that if both z 354 $ an z Y then C 3 šw# 54 shoul be 1 in orer to make the left han sie (strictly) less than 2. Next we will express the preceence constraint. For this, we efine a variable for each instruction to inicate whether instruction is scheule on fast FU (in which case D 1 ) or on slow FU (in which case B ª ). The variable D is efine as: c a`cb 8.e9f ƒ 8 for all g h VXY<]i[ :]\ (8) Now, for each epenence arc z a5 in the DDG, the preceence constraint between an is formulate as [16]: ^ P[ F R & P[ T* w where o± is the latency from instruction to an ž Ž± is the epenence istance [15, 11, 14]. In the ual-spee architecture, the latency Ž± will be o± 4 Yˆ& 53 an o± 4 5ŠŒ " for fast an slow FUs. The preceence constraints can be expresse as: ^ [/ r 'M. " a = 8 ² :[ ³'M. " 5 [ T w for all ² g <.'³ (9) A power/energy aware software pipelining metho shoul try to construct a scheule that oes not incur any performance egraation an also consumes the minimum energy. A simple objective function to minimize the energy consumption for the same initiation interval, is the one which minimizes the number of instructions issue to fast FUs. Assuming µ v"4 Yˆ& 53 an µ v4 ]ŠŒ are the weights for fast

4 n an slow functional units of type x, the objective function is given by: BkPl,mŽnp ² n = a 8? B n] 5 N² :F[ ³t³ (10) The full ILP formulation has the objective function as Equation 10 uner the constraints given by Equations Experimental Framework Using the ILP formulation propose in the previous section, we explore how the esign space of an Itanium-like architecture [8, 9] can be mae power-efficient by introucing slow FUs. Itanium is an Explicitly Parallel Instruction Computing (EPIC) processor that can issue 6 instructions in a single cycle. It has four types of functional units: M-unit, I-unit, F-unit, an B-unit, which work primarily for memory access, integer operations, floating-point operations an branch instructions respectively [9, 8]. In our work, we concentrate on M-unit an I-unit for power/energy reuction. We use a simple power moel in our experiments. Empirically, it was foun that FUs working on half frequency consume about 40% (represente by µ in following equation) of the energy consume by fast FUs [21]. That is, 5 U a = 8? Also, since we consier only slow M-unit an I-unit FUs in our experiments, we assume that the energy consume by the two types of FUs are same. Thus, Further, we assume the total energy consume by FUs incurre for the software pipeline scheule is the sum of the energy incurre for each instruction in the scheule. Ignoring the energy consume by inactive functional units an components such as issue logic an caches, the energy consume by a single iteration of the scheule of a loop as: = Li a = 8 a = 8 i 5 5 wherew Yˆ& 5354 anw 5ŠŒ " 4 represent the number of instructions scheule on fast an slow FUs. Using the trip counts of iniviual loops, the total energy consume by all software pipelineable loops as well as their total execution time were calculate. For etails on the energy moel, we reaers are referre to [19]. In our implementation, we use the Open64 compiler [18], with the highest optimization level enable, to extract the DDGs for software pipelineable loops from the SPEC2000 integer benchmark suite. For the extracte DDGs, our implementation generates the ILP formulation which is solve by a commercial integer linear programming solver, CPLEX [7]. The CPLEX solver gave us the scheule, the initiation interval of the loop, an instruction partitioning (to fast an slow FUs) of each DDG. By taking trip count of each DDG (which we obtaine by profiling using the train ata sets) into account, we get an overall estimate of the program performance an energy consumption. Table 1 summarizes the benchmarks use in our experiments, the number of software pipelineable loops in them, an the characteristics of the DDGs associate with these loops. Benchmark No. of # of noes # of eges Minimum II Loops (avg) (max) (avg) (max) (avg) (max) 164.gzip vpr mcf crafty parser bzip twolf Table 1. Statistics on DDGs Use 4. Experimental Results 4.1 Impact of Slow FUs In all our experiments, we consiere fast an slow FUs for the I-unit FUs an M-unit FUs only, together referre to as integer FUs. Further, in our experiments we kept the total number of integer FUs as 4, while the number of slow/fast FUs is varie among the configurations. We consier the following four configurations: C0 2 fast I-unit an 2 fast M-unit FUs C1 1 fast I-unit, 1 slow I-unit, an 2 fast M-unit FUs C2 1 fast an 1 slow in each integer FU type C3 2 slow I-unit, 1 slow M-unit, an 1 fast M-unit FUs Figure 4(a) an (b) show the energy gain an performance loss of configurations C1 C3 on all benchmarks, compare against the energy an performance of the baseline configuration C0. The last set of bars inicate the average energy reuction an performance egraation taken across all benchmarks. For configuration C1, we observe an energy gain of 12% or more in all the applications. With configurations C2, the energy savings increase to 18% an more. Lastly, for configuration C3, the energy gain is higher, ranging from 20% to 41%. It is important to note that the above energy savings are obtaine only by compromising a little or no performance in many cases. For configuration C1, except for benchmarks gzip an crafty which show relatively large performance egraation, 19.6% an 10% respectively, all other benchmarks incur very little performance egraation, often less than 1%. With configurations C2, the performance egraes, by 5% 20%, in 4 of the benchmarks, while for other three (mcf, parser an twolf) the performance egraation is negligible. Lastly, for configuration C3, both the performance egraation an the energy gain are higher. It shoul be

5 gzip vpr mcf crafty parser bzip2 twolf avg C1 C2 C C1 C2 C rate optimal PDT = 10% PDT = 20% PDT = 30% Energy saving Performance loss Energy saving with performance slack Benchmarks (a) Energy Gain uner Different Configurations 0 gzip vpr mcf crafty parser bzip2 twolf avg Benchmarks (b) Performance Penalty uner Different Configurations 0 gzip vpr mcf crafty parser bzip2 twolf avg Benchmarks (c) Energy Gain Traing Performance Figure 4. Energy-Performance Traeoff note here that even with configurations C2 an C3 (with, respectively, 2 an 3 slow FUs), the egraation in performance is within 15% for all the benchmarks, except gzip. Lastly, we observe that gzip suffers relatively large performance loss when slow functional units are introuce. This observation agrees with that of relate work by Seng et al [10]. Interestingly, two benchmarks, namely mcf an parser, incur no performance egraation even when 3 of the 4 integer units act as slow FUs. To summarize, for the benchmarks we stuie, configurations C1, C2 an C3 consume, respectively, 17.5%, 25.3%, an 30.2% less energy, on the average, compare to the baseline architecture. 4.2 Traing Performance for Energy Next we aress the question that whether further energy savings can be obtaine in ual spee architectures if the compiler is allowe to trae performance. To answer this question, we conuct a set of experiments on configuration C2 (with 2 fast an 2 slow FUs). Here, for each software pipelineable loop, we set the optimal II obtaine for configuration C2 as the base II value. We allow our compiler to choose an II value which is within a performance egraation threshol (PDT) from the base II value, an try to obtain energy-efficient scheules. We chose PDTs as 10%, 20%, 30% for each application. In Figure 4(c) we plot the energy savings compare to the baseline configuration (C0) for various PDT values. As shown in the figure, energy gains in three benchmarks, namely vpr, mcf, an twolf, increase with increase PDT value. However, in the other benchmarks, the energy savings appear only when the PDT is increase from 10% to 20%. The insignificant benefit in performance when the PDT is increase from 0% to 10% can be explaine by the fact that these benchmarks have a lower average Minimum II(MII) than the three benchmarks, as shown in Table 1. For those DDGs with small minii, 10% PDT leaves no space for them to increase its II. When a PDT value of 30%, is use, the energy gain is more apparent for all benchmarks. The average energy savings for PDTs of 20% an 30% are 31.8% an 34% respectively. Lastly, we note that a PDT value refers to an increase in the iniviual II. Hence a PDT of, say 20%, oes not necessarily mean a 20% egraation on the execution time. The average performance egraation for the benchmarks is only 15.4% when the PDT is 20%. 5. Relate Work A limit stuy on available scheule slack was performe by Casmira et al [3] which report that in more than 75% of the execution cycles there exists at least one instruction that has slack. Pyrey et al [12] exten this iea by partitioning functional units to fast ones an slow ones, operating the slow ones at half the rate of fast ones, an using runtime profiling to guie instruction issue to these two sets of functional units. In contrast, Seng et al propose to analyze instruction criticality ynamically an issue critical instructions to fast functional units while non-critical instructions to slow ones [10]. The mechanism they aopt to preict instruction criticality is the critical path preictor buffer [4]. Although interesting, the buffer consumes aitional power which may offset the power gain in functional units. The work by Zhang et al [21] uses compilation techniques to etermine the fast or slow functional units where each instruction shoul be issue to. They esigne a energy-oriente heuristic algorithm to re-orer a scheule mae by some performance-oriente scheuling algorithm to achieve energy saving with minimal performance egraation. A relevant work by Yun et al [20] targets poweraware software pipelining. They propose a heuristic algorithm which extene iterative moulo scheuling [13], tries to minimize step power for a software pipeline loop on a cycle-by-cycle basis.

6 6. Conclusions This paper reports a quantitative stuy of powerperformance trae-offs in the esign space of energy efficient architectures. It stuies the interplay between lowpower architecture features an compiler optimization techniques, specifically software pipelining. It explores both the architecture an the compiler sie of esign space. An Itanium-like architecture moel involving ual spee pipelines (slow an fast FUs) is use in this stuy. We have propose an elegant integer linear programming formulation for rate-optimal software pipelining on architectures involving ual-spee pipelines. We explore the esign space of Itanium-like architectures for energyperformance traeoffs. Our experimental results show that consierable energy saving can be achieve by introucing a few small functional units, with very small performance egraation. We also emonstrate that further energy gains can be obtaine by traing performance. Acknowlegment The authors woul like to acknowlege George Cai from Intel for his support on builing our power-aware computing testbe. We wish to thank Roy Ju from Intel, Chenyong Wu from Open Research Compiler (ORC) team of Chinese Acaemy of Sciences, an Peng Zhao from University of Alberta, for their help an comments on the initial rafts of this paper. Last, we acknowlege the research funings from DOE an NSF for builing our compiler infrastructure. References [1] E. R. Altman, R. Govinarajan, an G. R. Gao. Scheuling an mapping: Software pipelining in the presence of structural hazars. In Proc. of SIGPLAN PLDI 95, pages , La Jolla, Calif., Jun [2] D. Brooks, P. Bose, S. Schuster, H.Jacobson, P. K. Kuva, A. Buyuktosunoglu, J.-D. Wellman, V. Zyuban, M. Gupta, an P. Cook. Power-aware microarchtiecture: Design an moelling challenges for next-generation microprocessors. IEEE Micro, 20(6), Nov [3] J. Casmira an D. Grunwal. Dynamic instruction scheuling slack. In Proceeings of the 2000 KoolChips workshop, Monterey, California, Dec 10th [4] E.Tune, D.Liang, D.Tullsen, an B.Caler. Dynamic preiction of critical path instructions. In Proceeings of the 7th International Symposium on High Performance Computer Architecture, Feb [5] R. Govinarajan, E. R. Altman, an G. R. Gao. Minimizing register requirements uner resource-constraine rateoptimal software pipelining. In Proc. of MICRO-27, pages 85 94, San Jose, Calif., Nov. Dec [6] R. Govinarajan, E. R. Altman, an G. R. Gao. A framework for resource-constraine rate-optimal software pipelining. IEEE Trans. on Parallel an Distrib. Systems, 7(11): , Nov [7] ILOG. CPLEX mixe integer solver. [8] Intel. Intel IA-64 Architecture Software Developer s Manual, Aug [9] Intel. Itanium Processor Microarchitecture Reference for Software Optimization, Aug [10] J.Seng, E.Tune, an D.Tullsen. Reucing power with ynamic critical path information. In Proc. of MICRO-34, pages , Austin, TX, Dec [11] M. Lam. Software pipelining: An effective scheuling technique for VLIW machines. In Proc. of SIGPLAN PLDI 88, pages , Atlanta, Geor., Jun [12] R. Pyrey an G. Tyson. Evaluating esign traeoffs in ual spee pipelines. In Workshop on Complexity-Effective Design, Goteborg, Sween, Jun [13] B. R. Rau. Iterative moulo scheuling: An algorithm for software pipelining loops. In Proc. of MICRO-27, pages 63 74, San Jose, Calif., Nov. Dec [14] B. R. Rau an J. A. Fisher. Instruction-level parallel processing: History, overview an perspective. J. of Supercomputing, 7:9 50, May [15] B. R. Rau an C. D. Glaeser. Some scheuling techniques an an easily scheulable horizontal architecture for high performance scientific computing. In Proc. of MICRO-14, pages , Chatham, Mass., Oct [16] R. Reiter. Scheuling parallel computations. J. of the ACM, 15(4): , Oct [17] J. Ruttenberg, G. R. Gao, A. Stouchinin, an W. Lichtenstein. Software pipelining showown: Optimal vs. heuristic methos in a prouction compiler. In Proc. of SIGPLAN PLDI 96, pages 1 11, Philaelphia, May [18] SGI. Open64 open source compiler. [19] H. Yang, R. Govinarajan, G. R. Gao, an K. B. Theobal. A quantitative stuy on performance-power impact of ualspee pipeli ne architectures. CAPSL Tech. Memo 42, Dept. of Elec. an Computer Eng., U. of Delaware, Newark, Del., Jun In ftp://ftp.capsl.uel.eu/pub/oc/memos. [20] H.-S. Yun an J. Kim. Power-aware moulo scheuling for high-performance vliw processors. In Proceeings of the International Symposium on Low Power Electronics an Design 2001, Aug [21] W. Zhang, N.Vijaykrishnan, M.Kanemir, M.Irwin, D.Duarte, an Y. Tsai. Exploiting VLIW scheule slacks for ynamic an leakage energy reuction. In Proc. of MICRO-34, pages , Austin, TX, Dec 2001.

Just-In-Time Software Pipelining

Just-In-Time Software Pipelining Just-In-Time Software Pipelining Hongbo Rong Hyunchul Park Youfeng Wu Cheng Wang Programming Systems Lab Intel Labs, Santa Clara What is software pipelining? A loop optimization exposing instruction-level

More information

SURVIVABLE IP OVER WDM: GUARANTEEEING MINIMUM NETWORK BANDWIDTH

SURVIVABLE IP OVER WDM: GUARANTEEEING MINIMUM NETWORK BANDWIDTH SURVIVABLE IP OVER WDM: GUARANTEEEING MINIMUM NETWORK BANDWIDTH Galen H Sasaki Dept Elec Engg, U Hawaii 2540 Dole Street Honolul HI 96822 USA Ching-Fong Su Fuitsu Laboratories of America 595 Lawrence Expressway

More information

Online Appendix to: Generalizing Database Forensics

Online Appendix to: Generalizing Database Forensics Online Appenix to: Generalizing Database Forensics KYRIACOS E. PAVLOU an RICHARD T. SNODGRASS, University of Arizona This appenix presents a step-by-step iscussion of the forensic analysis protocol that

More information

Yet Another Parallel Hypothesis Search for Inverse Entailment Hiroyuki Nishiyama and Hayato Ohwada Faculty of Sci. and Tech. Tokyo University of Scien

Yet Another Parallel Hypothesis Search for Inverse Entailment Hiroyuki Nishiyama and Hayato Ohwada Faculty of Sci. and Tech. Tokyo University of Scien Yet Another Parallel Hypothesis Search for Inverse Entailment Hiroyuki Nishiyama an Hayato Ohwaa Faculty of Sci. an Tech. Tokyo University of Science, 2641 Yamazaki, Noa-shi, CHIBA, 278-8510, Japan hiroyuki@rs.noa.tus.ac.jp,

More information

filtering LETTER An Improved Neighbor Selection Algorithm in Collaborative Taek-Hun KIM a), Student Member and Sung-Bong YANG b), Nonmember

filtering LETTER An Improved Neighbor Selection Algorithm in Collaborative Taek-Hun KIM a), Student Member and Sung-Bong YANG b), Nonmember 107 IEICE TRANS INF & SYST, VOLE88 D, NO5 MAY 005 LETTER An Improve Neighbor Selection Algorithm in Collaborative Filtering Taek-Hun KIM a), Stuent Member an Sung-Bong YANG b), Nonmember SUMMARY Nowaays,

More information

Computer Organization

Computer Organization Computer Organization Douglas Comer Computer Science Department Purue University 250 N. University Street West Lafayette, IN 47907-2066 http://www.cs.purue.eu/people/comer Copyright 2006. All rights reserve.

More information

EFFICIENT ON-LINE TESTING METHOD FOR A FLOATING-POINT ADDER

EFFICIENT ON-LINE TESTING METHOD FOR A FLOATING-POINT ADDER FFICINT ON-LIN TSTING MTHOD FOR A FLOATING-POINT ADDR A. Droz, M. Lobachev Department of Computer Systems, Oessa State Polytechnic University, Oessa, Ukraine Droz@ukr.net, Lobachev@ukr.net Abstract In

More information

Transient analysis of wave propagation in 3D soil by using the scaled boundary finite element method

Transient analysis of wave propagation in 3D soil by using the scaled boundary finite element method Southern Cross University epublications@scu 23r Australasian Conference on the Mechanics of Structures an Materials 214 Transient analysis of wave propagation in 3D soil by using the scale bounary finite

More information

Solution Representation for Job Shop Scheduling Problems in Ant Colony Optimisation

Solution Representation for Job Shop Scheduling Problems in Ant Colony Optimisation Solution Representation for Job Shop Scheuling Problems in Ant Colony Optimisation James Montgomery, Carole Faya 2, an Sana Petrovic 2 Faculty of Information & Communication Technologies, Swinburne University

More information

Software Reliability Modeling and Cost Estimation Incorporating Testing-Effort and Efficiency

Software Reliability Modeling and Cost Estimation Incorporating Testing-Effort and Efficiency Software Reliability Moeling an Cost Estimation Incorporating esting-effort an Efficiency Chin-Yu Huang, Jung-Hua Lo, Sy-Yen Kuo, an Michael R. Lyu -+ Department of Electrical Engineering Computer Science

More information

Study of Network Optimization Method Based on ACL

Study of Network Optimization Method Based on ACL Available online at www.scienceirect.com Proceia Engineering 5 (20) 3959 3963 Avance in Control Engineering an Information Science Stuy of Network Optimization Metho Base on ACL Liu Zhian * Department

More information

Comparison of Methods for Increasing the Performance of a DUA Computation

Comparison of Methods for Increasing the Performance of a DUA Computation Comparison of Methos for Increasing the Performance of a DUA Computation Michael Behrisch, Daniel Krajzewicz, Peter Wagner an Yun-Pang Wang Institute of Transportation Systems, German Aerospace Center,

More information

Improving Performance of Sparse Matrix-Vector Multiplication

Improving Performance of Sparse Matrix-Vector Multiplication Improving Performance of Sparse Matrix-Vector Multiplication Ali Pınar Michael T. Heath Department of Computer Science an Center of Simulation of Avance Rockets University of Illinois at Urbana-Champaign

More information

Table-based division by small integer constants

Table-based division by small integer constants Table-base ivision by small integer constants Florent e Dinechin, Laurent-Stéphane Diier LIP, Université e Lyon (ENS-Lyon/CNRS/INRIA/UCBL) 46, allée Italie, 69364 Lyon Ceex 07 Florent.e.Dinechin@ens-lyon.fr

More information

Adaptive Load Balancing based on IP Fast Reroute to Avoid Congestion Hot-spots

Adaptive Load Balancing based on IP Fast Reroute to Avoid Congestion Hot-spots Aaptive Loa Balancing base on IP Fast Reroute to Avoi Congestion Hot-spots Masaki Hara an Takuya Yoshihiro Faculty of Systems Engineering, Wakayama University 930 Sakaeani, Wakayama, 640-8510, Japan Email:

More information

d 3 d 4 d d d d d d d d d d d 1 d d d d d d

d 3 d 4 d d d d d d d d d d d 1 d d d d d d Proceeings of the IASTED International Conference Software Engineering an Applications (SEA') October 6-, 1, Scottsale, Arizona, USA AN OBJECT-ORIENTED APPROACH FOR MANAGING A NETWORK OF DATABASES Shu-Ching

More information

Loop Scheduling and Partitions for Hiding Memory Latencies

Loop Scheduling and Partitions for Hiding Memory Latencies Loop Scheuling an Partitions for Hiing Memory Latencies Fei Chen Ewin Hsing-Mean Sha Dept. of Computer Science an Engineering University of Notre Dame Notre Dame, IN 46556 Email: fchen,esha @cse.n.eu Tel:

More information

Backpressure-based Packet-by-Packet Adaptive Routing in Communication Networks

Backpressure-based Packet-by-Packet Adaptive Routing in Communication Networks 1 Backpressure-base Packet-by-Packet Aaptive Routing in Communication Networks Eleftheria Athanasopoulou, Loc Bui, Tianxiong Ji, R. Srikant, an Alexaner Stolyar Abstract Backpressure-base aaptive routing

More information

Non-homogeneous Generalization in Privacy Preserving Data Publishing

Non-homogeneous Generalization in Privacy Preserving Data Publishing Non-homogeneous Generalization in Privacy Preserving Data Publishing W. K. Wong, Nios Mamoulis an Davi W. Cheung Department of Computer Science, The University of Hong Kong Pofulam Roa, Hong Kong {wwong2,nios,cheung}@cs.hu.h

More information

Generalized Edge Coloring for Channel Assignment in Wireless Networks

Generalized Edge Coloring for Channel Assignment in Wireless Networks Generalize Ege Coloring for Channel Assignment in Wireless Networks Chun-Chen Hsu Institute of Information Science Acaemia Sinica Taipei, Taiwan Da-wei Wang Jan-Jan Wu Institute of Information Science

More information

Almost Disjunct Codes in Large Scale Multihop Wireless Network Media Access Control

Almost Disjunct Codes in Large Scale Multihop Wireless Network Media Access Control Almost Disjunct Coes in Large Scale Multihop Wireless Network Meia Access Control D. Charles Engelhart Anan Sivasubramaniam Penn. State University University Park PA 682 engelhar,anan @cse.psu.eu Abstract

More information

Optimal Link Capacity Dimensioning in Proportionally Fair Networks

Optimal Link Capacity Dimensioning in Proportionally Fair Networks Optimal Link Capacity Dimensioning in Proportionally Fair Networks Micha l Pióro 1,Gábor Malicskó 2, an Gábor Foor 3 1 Department of Communication Systems, Lun Institute of Technology, Sween, Michal.Pioro@telecom.lth.se

More information

Optimizing the quality of scalable video streams on P2P Networks

Optimizing the quality of scalable video streams on P2P Networks Optimizing the quality of scalable vieo streams on PP Networks Paper #7 ASTRACT The volume of multimeia ata, incluing vieo, serve through Peer-to-Peer (PP) networks is growing rapily Unfortunately, high

More information

Register-Sensitive Software Pipelining

Register-Sensitive Software Pipelining Regier-Sensitive Software Pipelining Amod K. Dani V. Janaki Ramanan R. Govindarajan Veritas Software India Pvt. Ltd. Supercomputer Edn. & Res. Centre Supercomputer Edn. & Res. Centre 1179/3, Shivajinagar

More information

Improving Spatial Reuse of IEEE Based Ad Hoc Networks

Improving Spatial Reuse of IEEE Based Ad Hoc Networks mproving Spatial Reuse of EEE 82.11 Base A Hoc Networks Fengji Ye, Su Yi an Biplab Sikar ECSE Department, Rensselaer Polytechnic nstitute Troy, NY 1218 Abstract n this paper, we evaluate an suggest methos

More information

Classical Mechanics Examples (Lagrange Multipliers)

Classical Mechanics Examples (Lagrange Multipliers) Classical Mechanics Examples (Lagrange Multipliers) Dipan Kumar Ghosh Physics Department, Inian Institute of Technology Bombay Powai, Mumbai 400076 September 3, 015 1 Introuction We have seen that the

More information

Queueing Model and Optimization of Packet Dropping in Real-Time Wireless Sensor Networks

Queueing Model and Optimization of Packet Dropping in Real-Time Wireless Sensor Networks Queueing Moel an Optimization of Packet Dropping in Real-Time Wireless Sensor Networks Marc Aoun, Antonios Argyriou, Philips Research, Einhoven, 66AE, The Netherlans Department of Computer an Communication

More information

Coordinating Distributed Algorithms for Feature Extraction Offloading in Multi-Camera Visual Sensor Networks

Coordinating Distributed Algorithms for Feature Extraction Offloading in Multi-Camera Visual Sensor Networks Coorinating Distribute Algorithms for Feature Extraction Offloaing in Multi-Camera Visual Sensor Networks Emil Eriksson, György Dán, Viktoria Foor School of Electrical Engineering, KTH Royal Institute

More information

On the Placement of Internet Taps in Wireless Neighborhood Networks

On the Placement of Internet Taps in Wireless Neighborhood Networks 1 On the Placement of Internet Taps in Wireless Neighborhoo Networks Lili Qiu, Ranveer Chanra, Kamal Jain, Mohamma Mahian Abstract Recently there has emerge a novel application of wireless technology that

More information

Backpressure-based Packet-by-Packet Adaptive Routing in Communication Networks

Backpressure-based Packet-by-Packet Adaptive Routing in Communication Networks 1 Backpressure-base Packet-by-Packet Aaptive Routing in Communication Networks Eleftheria Athanasopoulou, Loc Bui, Tianxiong Ji, R. Srikant, an Alexaner Stoylar arxiv:15.4984v1 [cs.ni] 27 May 21 Abstract

More information

Cluster Center Initialization Method for K-means Algorithm Over Data Sets with Two Clusters

Cluster Center Initialization Method for K-means Algorithm Over Data Sets with Two Clusters Available online at www.scienceirect.com Proceia Engineering 4 (011 ) 34 38 011 International Conference on Avances in Engineering Cluster Center Initialization Metho for K-means Algorithm Over Data Sets

More information

Particle Swarm Optimization Based on Smoothing Approach for Solving a Class of Bi-Level Multiobjective Programming Problem

Particle Swarm Optimization Based on Smoothing Approach for Solving a Class of Bi-Level Multiobjective Programming Problem BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 17, No 3 Sofia 017 Print ISSN: 1311-970; Online ISSN: 1314-4081 DOI: 10.1515/cait-017-0030 Particle Swarm Optimization Base

More information

Provisioning Virtualized Cloud Services in IP/MPLS-over-EON Networks

Provisioning Virtualized Cloud Services in IP/MPLS-over-EON Networks Provisioning Virtualize Clou Services in IP/MPLS-over-EON Networks Pan Yi an Byrav Ramamurthy Department of Computer Science an Engineering, University of Nebraska-Lincoln Lincoln, Nebraska 68588 USA Email:

More information

Skyline Community Search in Multi-valued Networks

Skyline Community Search in Multi-valued Networks Syline Community Search in Multi-value Networs Rong-Hua Li Beijing Institute of Technology Beijing, China lironghuascut@gmail.com Jeffrey Xu Yu Chinese University of Hong Kong Hong Kong, China yu@se.cuh.eu.h

More information

Wish Branch: A New Control Flow Instruction Combining Conditional Branching and Predicated Execution

Wish Branch: A New Control Flow Instruction Combining Conditional Branching and Predicated Execution Wish Branch: A New Control Flow Instruction Combining Conditional Branching and Predicated Execution Hyesoon Kim Onur Mutlu Jared Stark David N. Armstrong Yale N. Patt High Performance Systems Group Department

More information

An Algorithm for Building an Enterprise Network Topology Using Widespread Data Sources

An Algorithm for Building an Enterprise Network Topology Using Widespread Data Sources An Algorithm for Builing an Enterprise Network Topology Using Wiesprea Data Sources Anton Anreev, Iurii Bogoiavlenskii Petrozavosk State University Petrozavosk, Russia {anreev, ybgv}@cs.petrsu.ru Abstract

More information

Coupling the User Interfaces of a Multiuser Program

Coupling the User Interfaces of a Multiuser Program Coupling the User Interfaces of a Multiuser Program PRASUN DEWAN University of North Carolina at Chapel Hill RAJIV CHOUDHARY Intel Corporation We have evelope a new moel for coupling the user-interfaces

More information

Scalable Deterministic Scheduling for WDM Slot Switching Xhaul with Zero-Jitter

Scalable Deterministic Scheduling for WDM Slot Switching Xhaul with Zero-Jitter FDL sel. VOA SOA 100 Regular papers ONDM 2018 Scalable Deterministic Scheuling for WDM Slot Switching Xhaul with Zero-Jitter Bogan Uscumlic 1, Dominique Chiaroni 1, Brice Leclerc 1, Thierry Zami 2, Annie

More information

Bends, Jogs, And Wiggles for Railroad Tracks and Vehicle Guide Ways

Bends, Jogs, And Wiggles for Railroad Tracks and Vehicle Guide Ways Ben, Jogs, An Wiggles for Railroa Tracks an Vehicle Guie Ways Louis T. Klauer Jr., PhD, PE. Work Soft 833 Galer Dr. Newtown Square, PA 19073 lklauer@wsof.com Preprint, June 4, 00 Copyright 00 by Louis

More information

ACE: And/Or-parallel Copying-based Execution of Logic Programs

ACE: And/Or-parallel Copying-based Execution of Logic Programs ACE: An/Or-parallel Copying-base Execution of Logic Programs Gopal GuptaJ Manuel Hermenegilo* Enrico PontelliJ an Vítor Santos Costa' Abstract In this paper we present a novel execution moel for parallel

More information

Modifying ROC Curves to Incorporate Predicted Probabilities

Modifying ROC Curves to Incorporate Predicted Probabilities Moifying ROC Curves to Incorporate Preicte Probabilities Cèsar Ferri DSIC, Universitat Politècnica e València Peter Flach Department of Computer Science, University of Bristol José Hernánez-Orallo DSIC,

More information

THE APPLICATION OF ARTICLE k-th SHORTEST TIME PATH ALGORITHM

THE APPLICATION OF ARTICLE k-th SHORTEST TIME PATH ALGORITHM International Journal of Physics an Mathematical Sciences ISSN: 2277-2111 (Online) 2016 Vol. 6 (1) January-March, pp. 24-6/Mao an Shi. THE APPLICATION OF ARTICLE k-th SHORTEST TIME PATH ALGORITHM Hua Mao

More information

Divide-and-Conquer Algorithms

Divide-and-Conquer Algorithms Supplment to A Practical Guie to Data Structures an Algorithms Using Java Divie-an-Conquer Algorithms Sally A Golman an Kenneth J Golman Hanout Divie-an-conquer algorithms use the following three phases:

More information

Shift-map Image Registration

Shift-map Image Registration Shift-map Image Registration Svärm, Linus; Stranmark, Petter Unpublishe: 2010-01-01 Link to publication Citation for publishe version (APA): Svärm, L., & Stranmark, P. (2010). Shift-map Image Registration.

More information

Classifying Facial Expression with Radial Basis Function Networks, using Gradient Descent and K-means

Classifying Facial Expression with Radial Basis Function Networks, using Gradient Descent and K-means Classifying Facial Expression with Raial Basis Function Networks, using Graient Descent an K-means Neil Allrin Department of Computer Science University of California, San Diego La Jolla, CA 9237 nallrin@cs.ucs.eu

More information

A shortest path algorithm in multimodal networks: a case study with time varying costs

A shortest path algorithm in multimodal networks: a case study with time varying costs A shortest path algorithm in multimoal networks: a case stuy with time varying costs Daniela Ambrosino*, Anna Sciomachen* * Department of Economics an Quantitative Methos (DIEM), University of Genoa Via

More information

On the Role of Multiply Sectioned Bayesian Networks to Cooperative Multiagent Systems

On the Role of Multiply Sectioned Bayesian Networks to Cooperative Multiagent Systems On the Role of Multiply Sectione Bayesian Networks to Cooperative Multiagent Systems Y. Xiang University of Guelph, Canaa, yxiang@cis.uoguelph.ca V. Lesser University of Massachusetts at Amherst, USA,

More information

Generalized Edge Coloring for Channel Assignment in Wireless Networks

Generalized Edge Coloring for Channel Assignment in Wireless Networks TR-IIS-05-021 Generalize Ege Coloring for Channel Assignment in Wireless Networks Chun-Chen Hsu, Pangfeng Liu, Da-Wei Wang, Jan-Jan Wu December 2005 Technical Report No. TR-IIS-05-021 http://www.iis.sinica.eu.tw/lib/techreport/tr2005/tr05.html

More information

Politecnico di Torino. Porto Institutional Repository

Politecnico di Torino. Porto Institutional Repository Politecnico i Torino Porto Institutional Repository [Proceeing] Automatic March tests generation for multi-port SRAMs Original Citation: Benso A., Bosio A., i Carlo S., i Natale G., Prinetto P. (26). Automatic

More information

Characterizing Decoding Robustness under Parametric Channel Uncertainty

Characterizing Decoding Robustness under Parametric Channel Uncertainty Characterizing Decoing Robustness uner Parametric Channel Uncertainty Jay D. Wierer, Wahee U. Bajwa, Nigel Boston, an Robert D. Nowak Abstract This paper characterizes the robustness of ecoing uner parametric

More information

Efficient State-Diagram Construction Methods for Software Pipelining

Efficient State-Diagram Construction Methods for Software Pipelining Efficient State-Diagram Construction Methods for Software Pipelining Chihong Zhang 1, Ramaswamy Govindarajan 2,SeanRyan 1,and Guang R. Gao 1 1 Dept. of Electric and Computer Engineering, Univ. of Delaware

More information

THE BAYESIAN RECEIVER OPERATING CHARACTERISTIC CURVE AN EFFECTIVE APPROACH TO EVALUATE THE IDS PERFORMANCE

THE BAYESIAN RECEIVER OPERATING CHARACTERISTIC CURVE AN EFFECTIVE APPROACH TO EVALUATE THE IDS PERFORMANCE БСУ Международна конференция - 2 THE BAYESIAN RECEIVER OPERATING CHARACTERISTIC CURVE AN EFFECTIVE APPROACH TO EVALUATE THE IDS PERFORMANCE Evgeniya Nikolova, Veselina Jecheva Burgas Free University Abstract:

More information

Shift-map Image Registration

Shift-map Image Registration Shift-map Image Registration Linus Svärm Petter Stranmark Centre for Mathematical Sciences, Lun University {linus,petter}@maths.lth.se Abstract Shift-map image processing is a new framework base on energy

More information

Wireless Sensing and Structural Control Strategies

Wireless Sensing and Structural Control Strategies Wireless Sensing an Structural Control Strategies Kincho H. Law 1, Anrew Swartz 2, Jerome P. Lynch 3, Yang Wang 4 1 Dept. of Civil an Env. Engineering, Stanfor University, Stanfor, CA 94305, USA 2 Dept.

More information

Chapter 9 Memory Management

Chapter 9 Memory Management Contents 1. Introuction 2. Computer-System Structures 3. Operating-System Structures 4. Processes 5. Threas 6. CPU Scheuling 7. Process Synchronization 8. Dealocks 9. Memory Management 10.Virtual Memory

More information

Dual Arm Robot Research Report

Dual Arm Robot Research Report Dual Arm Robot Research Report Analytical Inverse Kinematics Solution for Moularize Dual-Arm Robot With offset at shouler an wrist Motivation an Abstract Generally, an inustrial manipulator such as PUMA

More information

Threshold Based Data Aggregation Algorithm To Detect Rainfall Induced Landslides

Threshold Based Data Aggregation Algorithm To Detect Rainfall Induced Landslides Threshol Base Data Aggregation Algorithm To Detect Rainfall Inuce Lanslies Maneesha V. Ramesh P. V. Ushakumari Department of Computer Science Department of Mathematics Amrita School of Engineering Amrita

More information

AnyTraffic Labeled Routing

AnyTraffic Labeled Routing AnyTraffic Labele Routing Dimitri Papaimitriou 1, Pero Peroso 2, Davie Careglio 2 1 Alcatel-Lucent Bell, Antwerp, Belgium Email: imitri.papaimitriou@alcatel-lucent.com 2 Universitat Politècnica e Catalunya,

More information

Distributed Line Graphs: A Universal Technique for Designing DHTs Based on Arbitrary Regular Graphs

Distributed Line Graphs: A Universal Technique for Designing DHTs Based on Arbitrary Regular Graphs IEEE TRANSACTIONS ON KNOWLEDE AND DATA ENINEERIN, MANUSCRIPT ID Distribute Line raphs: A Universal Technique for Designing DHTs Base on Arbitrary Regular raphs Yiming Zhang an Ling Liu, Senior Member,

More information

Offloading Cellular Traffic through Opportunistic Communications: Analysis and Optimization

Offloading Cellular Traffic through Opportunistic Communications: Analysis and Optimization 1 Offloaing Cellular Traffic through Opportunistic Communications: Analysis an Optimization Vincenzo Sciancalepore, Domenico Giustiniano, Albert Banchs, Anreea Picu arxiv:1405.3548v1 [cs.ni] 14 May 24

More information

Interior Permanent Magnet Synchronous Motor (IPMSM) Adaptive Genetic Parameter Estimation

Interior Permanent Magnet Synchronous Motor (IPMSM) Adaptive Genetic Parameter Estimation Interior Permanent Magnet Synchronous Motor (IPMSM) Aaptive Genetic Parameter Estimation Java Rezaie, Mehi Gholami, Reza Firouzi, Tohi Alizaeh, Karim Salashoor Abstract - Interior permanent magnet synchronous

More information

Waleed K. Al-Assadi. Anura P. Jayasumana. Yashwant K. Malaiya y. February Colorado State University

Waleed K. Al-Assadi. Anura P. Jayasumana. Yashwant K. Malaiya y. February Colorado State University Dierential I DDQ Testable Static RAM Architecture Walee K. Al-Assai Anura P. Jayasumana Yashwant K. Malaiya y Technical Report CS-96-102 February 1996 Department of Electrical Engineering/ y Department

More information

NEW METHOD FOR FINDING A REFERENCE POINT IN FINGERPRINT IMAGES WITH THE USE OF THE IPAN99 ALGORITHM 1. INTRODUCTION 2.

NEW METHOD FOR FINDING A REFERENCE POINT IN FINGERPRINT IMAGES WITH THE USE OF THE IPAN99 ALGORITHM 1. INTRODUCTION 2. JOURNAL OF MEDICAL INFORMATICS & TECHNOLOGIES Vol. 13/009, ISSN 164-6037 Krzysztof WRÓBEL, Rafał DOROZ * fingerprint, reference point, IPAN99 NEW METHOD FOR FINDING A REFERENCE POINT IN FINGERPRINT IMAGES

More information

Automatic Counterflow Pipeline Synthesis

Automatic Counterflow Pipeline Synthesis Automatic Counterflow Pipeline Synthesis Bruce R. Childers, Jack W. Davidson Computer Science Department University of Virginia Charlottesville, Virginia 22901 {brc2m, jwd}@cs.virginia.edu Abstract The

More information

Disjoint Multipath Routing in Dual Homing Networks using Colored Trees

Disjoint Multipath Routing in Dual Homing Networks using Colored Trees Disjoint Multipath Routing in Dual Homing Networks using Colore Trees Preetha Thulasiraman, Srinivasan Ramasubramanian, an Marwan Krunz Department of Electrical an Computer Engineering University of Arizona,

More information

Handling missing values in kernel methods with application to microbiology data

Handling missing values in kernel methods with application to microbiology data an Machine Learning. Bruges (Belgium), 24-26 April 2013, i6oc.com publ., ISBN 978-2-87419-081-0. Available from http://www.i6oc.com/en/livre/?gcoi=28001100131010. Hanling missing values in kernel methos

More information

Robust PIM-SM Multicasting using Anycast RP in Wireless Ad Hoc Networks

Robust PIM-SM Multicasting using Anycast RP in Wireless Ad Hoc Networks Robust PIM-SM Multicasting using Anycast RP in Wireless A Hoc Networks Jaewon Kang, John Sucec, Vikram Kaul, Sunil Samtani an Mariusz A. Fecko Applie Research, Telcoria Technologies One Telcoria Drive,

More information

MORA: a Movement-Based Routing Algorithm for Vehicle Ad Hoc Networks

MORA: a Movement-Based Routing Algorithm for Vehicle Ad Hoc Networks : a Movement-Base Routing Algorithm for Vehicle A Hoc Networks Fabrizio Granelli, Senior Member, Giulia Boato, Member, an Dzmitry Kliazovich, Stuent Member Abstract Recent interest in car-to-car communications

More information

Compiler Optimisation

Compiler Optimisation Compiler Optimisation Michael O Boyle mob@inf.e.ac.uk Room 1.06 January, 2014 1 Two recommene books for the course Recommene texts Engineering a Compiler Engineering a Compiler by K. D. Cooper an L. Torczon.

More information

k-nn Graph Construction: a Generic Online Approach

k-nn Graph Construction: a Generic Online Approach k-nn Graph Construction: a Generic Online Approach Wan-Lei Zhao arxiv:80.00v [cs.ir] Sep 08 Abstract Nearest neighbor search an k-nearest neighbor graph construction are two funamental issues arise from

More information

Intensive Hypercube Communication: Prearranged Communication in Link-Bound Machines 1 2

Intensive Hypercube Communication: Prearranged Communication in Link-Bound Machines 1 2 This paper appears in J. of Parallel an Distribute Computing 10 (1990), pp. 167 181. Intensive Hypercube Communication: Prearrange Communication in Link-Boun Machines 1 2 Quentin F. Stout an Bruce Wagar

More information

Design of Controller for Crawling to Sitting Behavior of Infants

Design of Controller for Crawling to Sitting Behavior of Infants Design of Controller for Crawling to Sitting Behavior of Infants A Report submitte for the Semester Project To be accepte on: 29 June 2007 by Neha Priyaarshini Garg Supervisors: Luovic Righetti Prof. Auke

More information

Research Article Research on Law s Mask Texture Analysis System Reliability

Research Article Research on Law s Mask Texture Analysis System Reliability Research Journal of Applie Sciences, Engineering an Technology 7(19): 4002-4007, 2014 DOI:10.19026/rjaset.7.761 ISSN: 2040-7459; e-issn: 2040-7467 2014 Maxwell Scientific Publication Corp. Submitte: November

More information

Learning Polynomial Functions. by Feature Construction

Learning Polynomial Functions. by Feature Construction I Proceeings of the Eighth International Workshop on Machine Learning Chicago, Illinois, June 27-29 1991 Learning Polynomial Functions by Feature Construction Richar S. Sutton GTE Laboratories Incorporate

More information

Verifying performance-based design objectives using assemblybased vulnerability

Verifying performance-based design objectives using assemblybased vulnerability Verying performance-base esign objectives using assemblybase vulnerability K.A. Porter Calornia Institute of Technology, Pasaena, Calornia, USA A.S. Kiremijian Stanfor University, Stanfor, Calornia, USA

More information

6.823 Computer System Architecture. Problem Set #3 Spring 2002

6.823 Computer System Architecture. Problem Set #3 Spring 2002 6.823 Computer System Architecture Problem Set #3 Spring 2002 Stuents are strongly encourage to collaborate in groups of up to three people. A group shoul han in only one copy of the solution to the problem

More information

Optimal Routing and Scheduling for Deterministic Delay Tolerant Networks

Optimal Routing and Scheduling for Deterministic Delay Tolerant Networks Optimal Routing an Scheuling for Deterministic Delay Tolerant Networks Davi Hay Dipartimento i Elettronica olitecnico i Torino, Italy Email: hay@tlc.polito.it aolo Giaccone Dipartimento i Elettronica olitecnico

More information

Kinematic Analysis of a Family of 3R Manipulators

Kinematic Analysis of a Family of 3R Manipulators Kinematic Analysis of a Family of R Manipulators Maher Baili, Philippe Wenger an Damien Chablat Institut e Recherche en Communications et Cybernétique e Nantes, UMR C.N.R.S. 6597 1, rue e la Noë, BP 92101,

More information

Optimal Oblivious Path Selection on the Mesh

Optimal Oblivious Path Selection on the Mesh Optimal Oblivious Path Selection on the Mesh Costas Busch Malik Magon-Ismail Jing Xi Department of Computer Science Rensselaer Polytechnic Institute Troy, NY 280, USA {buschc,magon,xij2}@cs.rpi.eu Abstract

More information

Algorithm for Intermodal Optimal Multidestination Tour with Dynamic Travel Times

Algorithm for Intermodal Optimal Multidestination Tour with Dynamic Travel Times Algorithm for Intermoal Optimal Multiestination Tour with Dynamic Travel Times Neema Nassir, Alireza Khani, Mark Hickman, an Hyunsoo Noh This paper presents an efficient algorithm that fins the intermoal

More information

A Plane Tracker for AEC-automation Applications

A Plane Tracker for AEC-automation Applications A Plane Tracker for AEC-automation Applications Chen Feng *, an Vineet R. Kamat Department of Civil an Environmental Engineering, University of Michigan, Ann Arbor, USA * Corresponing author (cforrest@umich.eu)

More information

Design of Policy-Aware Differentially Private Algorithms

Design of Policy-Aware Differentially Private Algorithms Design of Policy-Aware Differentially Private Algorithms Samuel Haney Due University Durham, NC, USA shaney@cs.ue.eu Ashwin Machanavajjhala Due University Durham, NC, USA ashwin@cs.ue.eu Bolin Ding Microsoft

More information

NAND flash memory is widely used as a storage

NAND flash memory is widely used as a storage 1 : Buffer-Aware Garbage Collection for Flash-Base Storage Systems Sungjin Lee, Dongkun Shin Member, IEEE, an Jihong Kim Member, IEEE Abstract NAND flash-base storage evice is becoming a viable storage

More information

Comparison of Wireless Network Simulators with Multihop Wireless Network Testbed in Corridor Environment

Comparison of Wireless Network Simulators with Multihop Wireless Network Testbed in Corridor Environment Comparison of Wireless Network Simulators with Multihop Wireless Network Testbe in Corrior Environment Rabiullah Khattak, Anna Chaltseva, Laurynas Riliskis, Ulf Boin, an Evgeny Osipov Department of Computer

More information

Random Clustering for Multiple Sampling Units to Speed Up Run-time Sample Generation

Random Clustering for Multiple Sampling Units to Speed Up Run-time Sample Generation DEIM Forum 2018 I4-4 Abstract Ranom Clustering for Multiple Sampling Units to Spee Up Run-time Sample Generation uzuru OKAJIMA an Koichi MARUAMA NEC Solution Innovators, Lt. 1-18-7 Shinkiba, Koto-ku, Tokyo,

More information

Socially-optimal ISP-aware P2P Content Distribution via a Primal-Dual Approach

Socially-optimal ISP-aware P2P Content Distribution via a Primal-Dual Approach Socially-optimal ISP-aware P2P Content Distribution via a Primal-Dual Approach Jian Zhao, Chuan Wu The University of Hong Kong {jzhao,cwu}@cs.hku.hk Abstract Peer-to-peer (P2P) technology is popularly

More information

Politehnica University of Timisoara Mobile Computing, Sensors Network and Embedded Systems Laboratory. Testing Techniques

Politehnica University of Timisoara Mobile Computing, Sensors Network and Embedded Systems Laboratory. Testing Techniques Politehnica University of Timisoara Mobile Computing, Sensors Network an Embee Systems Laboratory ing Techniques What is testing? ing is the process of emonstrating that errors are not present. The purpose

More information

Modulo-Variable Expansion Sensitive Scheduling

Modulo-Variable Expansion Sensitive Scheduling Modulo-Variable Expansion Sensitive Scheduling Madhavi Gopal Vallurif and R. GovindarajanfJ f Supercomputer Education and Research Centre Department of Computer Science and Automation Indian Institute

More information

Reconstructing the Nonlinear Filter Function of LILI-128 Stream Cipher Based on Complexity

Reconstructing the Nonlinear Filter Function of LILI-128 Stream Cipher Based on Complexity Reconstructing the Nonlinear Filter Function of LILI-128 Stream Cipher Base on Complexity Xiangao Huang 1 Wei Huang 2 Xiaozhou Liu 3 Chao Wang 4 Zhu jing Wang 5 Tao Wang 1 1 College of Engineering, Shantou

More information

Latency-Sensitive Data Allocation for Cloud Storage

Latency-Sensitive Data Allocation for Cloud Storage Latency-Sensitive Data Allocation for Clou Storage Song Yang 1, Philipp Wieer 1, Muzzamil Aziz 1, Ramin Yahyapour 1, 2 an Xiaoming Fu 2 1 Gesellschaft für wissenschaftliche Datenverarbeitung mbh Göttingen

More information

Using Vector and Raster-Based Techniques in Categorical Map Generalization

Using Vector and Raster-Based Techniques in Categorical Map Generalization Thir ICA Workshop on Progress in Automate Map Generalization, Ottawa, 12-14 August 1999 1 Using Vector an Raster-Base Techniques in Categorical Map Generalization Beat Peter an Robert Weibel Department

More information

Research Article Inviscid Uniform Shear Flow past a Smooth Concave Body

Research Article Inviscid Uniform Shear Flow past a Smooth Concave Body International Engineering Mathematics Volume 04, Article ID 46593, 7 pages http://x.oi.org/0.55/04/46593 Research Article Invisci Uniform Shear Flow past a Smooth Concave Boy Abullah Mura Department of

More information

CS 106 Winter 2016 Craig S. Kaplan. Module 01 Processing Recap. Topics

CS 106 Winter 2016 Craig S. Kaplan. Module 01 Processing Recap. Topics CS 106 Winter 2016 Craig S. Kaplan Moule 01 Processing Recap Topics The basic parts of speech in a Processing program Scope Review of syntax for classes an objects Reaings Your CS 105 notes Learning Processing,

More information

Fast Window Based Stereo Matching for 3D Scene Reconstruction

Fast Window Based Stereo Matching for 3D Scene Reconstruction The International Arab Journal of Information Technology, Vol. 0, No. 3, May 203 209 Fast Winow Base Stereo Matching for 3D Scene Reconstruction Mohamma Mozammel Chowhury an Mohamma AL-Amin Bhuiyan Department

More information

Saving Register-File Leakage Power by Monitoring Instruction Sequence in ROB

Saving Register-File Leakage Power by Monitoring Instruction Sequence in ROB Saving Register-File Leakage Power by Monitoring Instruction Sequence in ROB Wann-Yun Shieh Department of Computer Science and Information Engineering Chang Gung University Tao-Yuan, Taiwan Hsin-Dar Chen

More information

ANALYTICAL OPTIMIZATION OF SIMPLE ROOF SHADING DEVICES. Khaled Nassar 1 and Mohamed Aly 1. line below should be left

ANALYTICAL OPTIMIZATION OF SIMPLE ROOF SHADING DEVICES. Khaled Nassar 1 and Mohamed Aly 1. line below should be left ANALYTICAL OPTIMIZATION OF SIMPLE ROOF SHADING DEVICES Khale Nassar 1 an Mohame Aly 1 1 The American University in Cairo line below shoul be left ABSTRACT Parametric an algorithmic esign tools have evelope

More information

Here are a couple of warnings to my students who may be here to get a copy of what happened on a day that you missed.

Here are a couple of warnings to my students who may be here to get a copy of what happened on a day that you missed. Preface Here are my online notes for my Calculus I course that I teach here at Lamar University. Despite the fact that these are my class notes, they shoul be accessible to anyone wanting to learn Calculus

More information

Implementation and Evaluation of NAS Parallel CG Benchmark on GPU Cluster with Proprietary Interconnect TCA

Implementation and Evaluation of NAS Parallel CG Benchmark on GPU Cluster with Proprietary Interconnect TCA Implementation an Evaluation of AS Parallel CG Benchmark on GPU Cluster with Proprietary Interconnect TCA Kazuya Matsumoto 1, orihisa Fujita 2, Toshihiro Hanawa 3, an Taisuke Boku 1,2 1 Center for Computational

More information

THE increasingly digitized power system offers more data,

THE increasingly digitized power system offers more data, 1 Cyber Risk Analysis of Combine Data Attacks Against Power System State Estimation Kaikai Pan, Stuent Member, IEEE, Anré Teixeira, Member, IEEE, Milos Cvetkovic, Member, IEEE, an Peter Palensky, Senior

More information

JigCell Model Connector: building large molecular network models from components

JigCell Model Connector: building large molecular network models from components Simulation Applications JigCell Moel Connector: builing large molecular network moels from components Simulation: Transactions of the Society for Moeling an Simulation International 2018, Vol. 94(11) 993

More information