Writing Libraries in MPI*

Size: px
Start display at page:

Download "Writing Libraries in MPI*"

Transcription

1 Writing Libraries in MPI* Anthony Skjellumt Nathan E. Doss Purushotham V. Bangaloret Computer Siene Departmentt & NSF Engineering Researh Center for Computational Field Simulation Mississippi State University Abstrat Abstrat MPI is the new de fato standard for multiomputer and luster message passing. This shod paper desribes the issues underlying the development of parallel libraries with MPI. We motivate the need for parallel libraries, and explain why ommon messagepassing systems do not support them. We desribe added features needed to support libraries, and what these features are alled in MPI. We mention additional features of MPI that support libraries, but whih are beyond the sope of this initial paper on this topi. We provide guidelines in terms of "do's" and UdonYsl' for would-be MPI library writers. We present a linear algebra library in skeletal form on a twodimensional virtual topology, inluding ode fragments. We summarize by mentioning related work, and omment on future ode development of parallel libraries under MPI, as well as porting efforts of existing odes lo the system. 1 Introdution MPI is the new de fato standard for multiomputer and luster mesaage passing [l]. Besides portability and higher performane, one of the primary motivations for MPI is to support parallel libraries, and, onsequently, to support the reation of large-sale multiomputer software. We desribe basi issues in the development of parallel libraries using MPI. We motivate the need for parallel libraries as suh, and explain why ommon message-passing systems do not support them well. We desribe the kind of message-passing features that 'Work supported by the NSF Engineering Rerearh Center for Computational Field Simulation, Missisaippi State University. are needed to support libraries, what MPI alls suh features, and how they are related to one another. We mention additional features of MPI that support litbraries, but whih are beyond the =ope of this initial paper on this topi. We provide guidelines in terms of "do's" and Udon'ts" for would-be MPI library writers. Given MPI features for library safety, MPI library writers will be able to expet reliable message-passing behavior without sarifiing performane. Furthermore, MPI provides a higher level of abstration than ommon message-passing systems, whih will help library writm develop more effiient and understandable odes simultaneously. MPI was designed to support primitives that would map effiiently to the existing and expeted high performane message passing hardware of multiomputers and gigabit luster networks. MPI library writers will be able to aess suh hardware apabilities in portable ode. To onretize the disussion, we present a linear algebra library in skeletal form on a two-dimensional virtual topology, inluding ode fragments. These exa-mples are simplified from other work we have done in the area of salable libraries, so they represent realisti implementations. We think that this extended example will prove instrutive to those who are planning to d.evelop distributed data strutures for mathematial libraries ["I. We summarize by mentioning related work, and omment on future ode development of parallel libraries under MPI, as well as porting efforts of existing library odes to the system. 2 Motivation This setion motivates the need for parallel libraries. It sets expetations for what libraries need to work effetively in a distributed memory environment, and explains why ommon message-passing sys V94 $3.00 Q 1994 IEEE 166

2 term do not support libraries well. MPI's solutions to these needs are disussed. 2.1 Why Parallel Libraries? Parallel libraries are needed to enapsulate the ponderous details assoiated with distributed realizations of key algorithms. The programmer needs: 0 portable, reusable, high quality odes, 0 easily available, reliable odes, 0 higher level portability interfaes, 0 the ability to hide oding omplexities from other programmers and users. Libraries ensure onsistent quality of suh proedures, and provide a "higher level" of portability than MPI itself an or should provide. Libraries prevent eah programmer from repeating the basi work of defining onsistent data strutures, data layouts, and methods that implement key algorithms (suh as matrix operations). Sine the best libraries ome with several variations on parallel systems (different data layouts, different strategies dependent on sirse of system, or type of floating point), this too needs to be hidden from the user. 2.2 What's Missing? The following features are missing from ommon vendor and portability systems: 0 ontexts of message passing (isolation), 0 olletive ommuniation funtionality, defined with proess-group sope, 0 virtual-topology naming of proesses. The subsequent paragraphs elaborate on eah of these topis, and why libraries need these features Communiation Contexts - Isolation In the sequential Fortran and C environments, it is reasonably easy to reate libraries, beause the stak-oriented proedural programming model has well-defined onditions about reasonable versus erroneous programs. However, in the distributed-memory, message-passing environment, it is diffiult to write libraries with either vendor or ommon portability systems. This seemingly strong inditment is baked up as follows: There is no reasonable way for libraries to isolate themselves from the on-going point-to-point message passing present in a running appliation. Message tags, the sole meane for designating restritions on message delivery, ar inadeient for isolation. For instane, more than one library (or invoation of the same library) ould use the mame tags. Systems that lak soure seletivity have to use tags to reate deterministi olletive operations; in suh a situation, olletive and point-to-point messages an be misqueued unless the system reserves tags for this purpose. Finally, wildard reeipt-seletivity on tags destroys any promise of real protetion that tags ould otherwise afford Colletive Operations Colletive operations, suh as redue, broadast, and so on, are not defined in all ommon systems, and when defined are most often only available over the full set of proessors (or proesses) alloated to the user. Libraries need to work on different olletions of proesses, and to have available olletive operations that only impat the proesses that are about the olletive operation; uninvolved proesses want to remain uninvolved. Furthermore, the olletive operations have to work reliably, and without rae onditions Virtual-Topology Naming Libraries don't want to desribe point-to-point ommuniation in terms of hardware-dependent names; in fat, many algorithms are more natural if desribed in terms of point-to-point alls relative to a virtualtopology naming sheme. Virtual-topology naming might reflet row or olumn parallelism in matrix op erations, for instane [SI. So, we have established three key areas that will support the reation of libraries: 0 Safe ommuniation spae (isolation), 0 Support for olletive operation, not just pointto-point; group sope is needed for olletive o p erations, 0 Abstrat names for proesses are based on virtual topologies, or at least rank-in-group names. 2.3 MPI's Support for Libraries In MPI, the features that implement the important onepts disussed in the previous setion are as follows: 167

3 Proess Groups define an ordered olletion of proenses, eah with a rank. Proess groups define the low-level names for inter-proeee ommuniation. For point-to-point message passing, proaa group define the rank of sender and reipient. For olletive ommuniation, proess group define the set of partiipants. Virtual Topologies are the final link in the naming proess for libraries, providing an alternative to simple rank naming for libraries. Cartesian and general graph topologies are supported. The user an also define his or her own topology strategies. Contexts provide the ability to have separate safe "universes" of message-passing for proess group in MPI. (A ontext is a onept made up of a proess group and other information, it does not speify an implementation.) We talk about ontexts in MPI, but they are atually implemented with ommuniators. Communiators enapsulate ontexts, groups, and virtual topologies in an objet that provides the appropriate sope for all ommuniation op erations in MPI. Communiators bind proess group and ontexts together to form a safe ommuniation spae within the group. In summary, the use of separate ommuniation ontexts by distint libraries (or distint library invoations) will insulate ommuniation internal to the library exeution from external ommuniation (group safety). This allows the invoation of the library even if there are pending ommuniations, and avoids the need to synhronize eah entry into and exit from library ode. 2.4 Additional MPI features for libraries MPI also provides several other features whih are useful for building salable libraries: 0 Dupliation of ommuniators allows libraries to get additional ommuniation spae easily. (Good implementations may/will ahe suh additional ommuniation spae, resulting in signifiant performane benefits.) 0 Systemati, aligned subsetting of ommuniators allows ommuniating SPMD proesses to define hierarhies of ommuniating sub-groups. 0 Cahing of attributes on ommuniators allows users to define new operations that have the same oneptual properties as built-in MPI olletive operations. This is important for extending servies not antiipated by MPI. 3 Guidelines for Library Writers In this setion, we bulletize important ideas for library writers based on our previous experiene with the Multiomputer Toolbox and Zipode [3]. Pre grammers who have tried to write libraries with ommon systems will appreiate the simpliity of writing message-passing ode with MPI. 3.1 DO'S 0 Define persistent distributed strutures, based on ommuniators, 0 Dupliate ommuniators for safe ommuniation spae when entering alls to parallel libraries, 0 Use MPI-defined virtual topology information to make algorithms easier to understand, 0 Augment virtual topology tehnology as needed (e.g., topology hierarhies, as in setion 4), 0 Use attribute ahing when adding additional olletive operations to a ommuniator. 3.2 Don'ts 0 Publish how they use the message resoures (e.g., tag allotments), 0 Synhronize exessively at the entry and exit of parallel alls, 0 Guarantee that ommuniators be "quiesent" before they an do olletive operations, 0 Restrit the library programming model to single invoation per proess tehnology, 0 Restrit the library programming model to nonoverlapping proess group assumptions, 0 Write libraries that work only over "all" proessors (proesses) of a user's alloation. 168

4 4 A 2D-Grid Vetor Class Library In this setion we develop a "vetor" library for linear algebra on a twedimensional logial topology and demonstrate two variants of "dot produt" for aligned, dietributed vetors. 4.1 Vetor Layout In a 2D logial grid, vetors may either be olumn or row vetors. We assume that eah row (reap, olumn) vetor is repliated over eah row (resp, olumn) of a logial grid. Figure 1 shows how two row vetors might be dietributed in a grid. Eah vetor is broken into two piees and distributed along the olumns of the grid. Eah proess row has a opy of both vetors. Figure 1: This figure shows the layout of two row vetors over a 2 x 2 logial proess grid. Column vetors are laid out in an analogous, fashion, orthogonal to the row vetors. 4.2 Creating a Logial Grid MPI provides a set of routines whih an be used to reate virtual topologies. In this first example, we show how a library writer might reate a logial 2D grid struture without using the topology routines provided by MPI. Here is the persistent data struture needed to reate a logial grid along with a simple funtion to initialise a grid. /* from tbx-1ibrary.h */ typedef strut tbx-2d-grid < int P, Q; int p, q; MPI-Comm grid-omm; MPI-Comm row-omm; MPI-omm ol-omm; 1 TBX-2D-GRID; /* from tbx,grid. */ /* global shape */ /* position */ TBX-GRID *tbx,build,grid(mpi-comm omm, int P, Q) TBX-2D-GRID *grid; MPI-Comm row, ol; int PD q; /* Determine row and olumn position */ HPI-Comm-rank(omm, hay-rank) ; p = my-rank / 9; q = my-rank X 9; /* Split omm into row and ol omms */ MPI-Comm-split (omm, p, q, &row) ; MPI-Comm-split (omm, q, p, &ol) ; /* Make new grid */ grid = BEW-CLEAR-ITEH(TBX-2D-GRID, 1) ; /* Fill in new grid struture */ grid->grid-omm = omm; grid->row-omm = row; grid->ol-omm = ol; grid->p = P; grid- >q = 9; grid->p = p; grid->q = 9; /* Return the newly built grid */ return (grid); Vetor Struture and Creation Vetors are defined by the following data struture. As a distributed objet, they ontain loal information inluding global and loal lengths, loal real data storage, and a type field that indiates if they are rowor olumn-distributed. In addition, the data struture is based on a logial grid topology data struture, and a, data distribution topology. The former enapsulates 169

5 ommuniators, the latter enapsulates funtions that map global indies to loal indies and vie versa. See Figure 2. /* from tbx-1ibrary.h */ #define TBX-ROW 0 #define TBX-COL 1 typedef strut tbx-dvetor int length; /* full length of vetor */ int loal-length;/* loal vetor length */ int type; /* row or olumn type */ double *data; /* vetor data */ TBX-2D-GRID *grid; TBX-DISTRIB *dis; /* logial grid */ /* how to map data */ /* funtions working on vetors */ double (*dot) (strut tbx-dvetor *x, *y); /*... */ > TBX-DVECTOR; The following funtion is the onstrutor for a die tributed vetor. /* from tbx-dvet0r. */ /* * length = total length of vetor type = row or olumn vetor? * grid = grid whih vetor belongs to * dis = distribution funtions * (how oeffiients map) */ TBX-DVECTOR *new-dvetor(int length, type, TBX-GRID *grid, TBX-DISTRIB *dis) TBX-DVECTOR *ve; int loal-length; /* Create a new vetor */ ve = BEW-CLEAR-ITEM(TBX-DVECTOR,~); /* figure the loal length - i.e., how */ /* many oeffiients (and whih ones) */ /* map to this proess */ loal-length = (*dis -> loal-length) ( grid, dis, length, type); /* Fill in the vetor struture */ ve->length = length; ve->loal-length = loal-length; ve->dat a - HEW-CLEAR-ITEM(double, loal-length) ; ve->type = type; ve->grid = grid; ve->dis = dis; /* assign values to default fns */ ve->dot = ddot-strided; /*... */ 1 Grid to whih vetor belongs Total length of vetor Vetor type -row or olumn I Length of loal part of vetor ~peralions on vetors FUC~~OIW that imp& vetor operations Figure 2: This figure shows the struture of a distributed vetor. 4.4 Dot Produt Funtions High-quality libraries will have to provide more than one version of a funtion that would perform effiiently over different onditions (we all suh a set of methods a poly-algorithm). For example, here we provide two dot produt funtions (strided and nonstrided). The strided version has a redued omputational load at the expense of more ommuniation. Conversely, the non-strided version is omputationally 170

6 more intensive but fewer messages are needed, beause the algorithm exploits data redundany. The most effiient funtion is determined by omparing issues suh aa message lateny and bandwidth, and floating point performane, all with referene to the size of the vetor, and the logial grid dimensions P, Q Non- Strided The non-strided funtion works independently in proess rows (resp, olumns) for row-distributed (resp, olumn-distributed) vetors. First, the loal ontribution to the dot produt is omputed in eah proess, and then, a "ombine" funtion, alled MPI-Allredue() sums the partial sums, leaving the dot produt in eah proess. The olletive ommuniation ours independently in rows (resp, olumns). /* from tbx-dvet0r. */ double ddot-nostride(tbx-dvector *x, *y) int i; double sum = 0.0; /* Chek for ompatible types */ if (x->type!= y->type) C /* erroneous */ 3 /* Sum up my portion of the vetor */ for (i = 0; i < x->loal-length; i++) sum += x->data[i] * y->datacil; /* Get row or olumn sum based on type */ if (x->type == TBX-ROW) MPI-Allredue(Btsum, &sum, 1, MPI-DOUBLE, MPI-SUM, x->grid->rou-omm) ; else MPI-Allredue(&sum, &sum, I, MPI-DOUBLE, MPI-SUM, x->grid->ol-omm); /* Return the result */ return (sum); Strided The strided version of dot produt relies on the fat that there is data repliation. Hene, only a fration of eah sum need be omputed in any proess. However, given striding, the ommuniation required to form the global sum requires MPI-Allredue() to work over all proesses speified by the logial grid, rather than independently in rows or olumns, as above. /* from tbx-dvet0r. */ double ddot-stride(tbx-dvector *x, *y) int i, stride; double sum = 0.0; /* Chek for ompatible types */ if (x->type!= y->type) C /* erroneous */ 3 /* Determine the stride based on type */ if (x->type == TBX-ROW) stride = x->grid->p; else stride = x->grid->q; /* sum up my part */ for ( i = r->grid->p; i < x->loal-length; i += stride) sum += x->data[i] * y->data[il; /* Get the sum of all parts */ MPI-Allredue(Csum, &sum, 1. MPI-DOUBLE, MPI-SUM, x->grid->grid-omm) ; /* Return result */ return( sum) ; 1, 5 The Example Revisited In this setion we revise the "vetor" library for linear algebra on a two-dimensional logial topology and demonstrate two variants of "dot produt" for distributed vetors. This time, we use the MPI-defined Cartesian topologies to help. As before, in a 2D logial grid, vetors may either be olumn or row vetors. Eherything else desribed in the first example remains the same. 5.1 Creating a Logial Grid MPI provides a set of routines whih an be used to reate virtual topologies. The funtion MPI-Make-art () is used for Cartesian topologies. The struture tbr-2d-grid is unhanged from before. However, the funtion tbx-build-grid() makes use of the MPI Cartesian virtual topology funtions, and is somewhat different. 171

7 #define B-DIMS 2 /* from tbr-grid. */ TBX-GRID *tbx,build-grid(mpi-comm omm, int P, 9) I TBX-2D-GRID *grid; NPI-Comm row, ol; int my-r-, P, q; HPI-Comm omm-2d; int dims[lf-dims], /* hold dimensions */ loal[lf-dims], /* loal position */ period[b-dihs] ; /* aperiodi flags */ /* Generate a new ommuniator with virtual topology added: */ dims[o] = P; dims[ll = 9; periodco] = period[l] = FALSE; MPI-Make-art(omm, lf-dims, dims, period, TRUE, Lomm-2d) ; /* Split orm into row and ol omms */ MPI-Comm,rank(omm, &my-rank) ; /* map bak to topology oordinates: */ HPI-Cart-oords(omm~2d,my-ra~k, lf-dims, &loal); p = loal Col ; q = 1oalCil; to send a mail message to netlibooml.gov ontaining the line send index from HPI. Alternatively, one an anonymous ftp the soure from netlib20s.utk.edu. To learn more about full-sale parallel libraries and general distributed data strutures along the lines desribed above, see one or more of the following referenes onerning the Multiomputer Toolbox [4, 5, 6, 7, Conlusions We introdued MPI, the new de fato standard for multiomputer and luster message passing [l]. We desribed the issues underlying the development of parallel libraries with MPI. In fat, we began by motivating the need for parallel libraries per se, and we explained why ommon message-paseing systems do not support them. We desribed added features needed to support suh libraries, and what these features are alled in MPI. We mentioned additional features of MPI that support libraries, but whih are beyond the sope of this initial paper on this topi (see [7]). We provided basi "do's" and Udon'ts" for wouldbe MPI library writers. We presented a linear algebra library in skeletal form on a two-dimensional virtual topology, inluding ode fragments. We summarized by mentioning related work, and omment on future ode development of parallel libraries under MPI, as well as porting efforts of existing odes to the system. /* Make new grid */ grid = lfew-clear-item(tbx-2d-grid, 1); /* Fill in new grid struture */ grid->grid-omm = omm-2d; grid->row-omm = row; grid->ol-omm = ol; grid->p = P; grid->q = 9; grid->p = p; grid->q = 9; /* Return the newly built grid */ return (grid); 1 6 Related Work To find out more about MPI, see the doument [l] available on netlib. A quik way to aess this is Referenes Message Passing Interfae Forum. Doument for a Standard Message-Passing Interfae. Tehnial Report Tehnial Report No. CS , University of Tennessee, November Available on net lib. William Gropp and Ewing Lusk. An Abstrat Devie Definition to Support the Implementation of a High-Level Point-to-Point Message-Passing Interfae. Tehnial Report MCS-P , Mathematis and Computer Siene Division, Argonne National Laboratory, Argonne, IL 60439, Anthony Skjellum. The Design and Evolution of Zipode. Parallel Computing, (Invited Paper, to appear in Speial Issue on Message Passing). Anthony Skjellum. The Multiomputer Toolbox: Current and Future Diretions. In Anthony Skjel- 172

8 lum and Donna S. Reese, editors, Proeedings of the Salable Parallel Libraries Conferene. IEEE Computer Soiety Preee, Otober [5] Anthony Skjellum, Steven F. Ashby, Peter N. Brown, Milo R. Dorr, and Alan C. Hindmarsh. The Multiomputer Toolbox. In G. L. Struble et al., editors, Laboratory Direted Researh and Deuelopment FY91 - LLNL, pages Lawrene Livermore National Laboratory, August UCRL (Rev 1). [6] Anthony Skjellum and Chuk H. Baldwin. The Multiomputer Toolbox: Salable Parallel Libraries for Large-sale Conurrent Appliations. Tehnial Report UCRL-JC , Lawrene Livermore National Laboratory, Deember [7] Anthony Skjellum et al. Salable Libraries Using MPI: Abstration, Performane, Portability. In preparation, [SI Anthony Skjellum, Alvin P. Leung, Charles H. Still Steven G. Smith, Robert D. Falgout, and Chuk H. Baldwin. The Multiomputer Toolbox - First- Generation Salable Libraries. In Proeedings of HICSS-27. IEEE Computer Soiety Press, HICSS-27 Minitrak on Tools and Languages for Transportable Parallel Appliations. [9] Steven G. Smith, Robert D. Falgout, Charles H. Still, and Anthony Skjellum. High-level Messagepassing Construts for Zipode 1.0: Design and Implementation. In Proeedings of the Salable Parallel Libraries Conferene. IEEE Computer Soiety Press,

Pipelined Multipliers for Reconfigurable Hardware

Pipelined Multipliers for Reconfigurable Hardware Pipelined Multipliers for Reonfigurable Hardware Mithell J. Myjak and José G. Delgado-Frias Shool of Eletrial Engineering and Computer Siene, Washington State University Pullman, WA 99164-2752 USA {mmyjak,

More information

Outline: Software Design

Outline: Software Design Outline: Software Design. Goals History of software design ideas Design priniples Design methods Life belt or leg iron? (Budgen) Copyright Nany Leveson, Sept. 1999 A Little History... At first, struggling

More information

Extracting Partition Statistics from Semistructured Data

Extracting Partition Statistics from Semistructured Data Extrating Partition Statistis from Semistrutured Data John N. Wilson Rihard Gourlay Robert Japp Mathias Neumüller Department of Computer and Information Sienes University of Strathlyde, Glasgow, UK {jnw,rsg,rpj,mathias}@is.strath.a.uk

More information

Automatic Physical Design Tuning: Workload as a Sequence Sanjay Agrawal Microsoft Research One Microsoft Way Redmond, WA, USA +1-(425)

Automatic Physical Design Tuning: Workload as a Sequence Sanjay Agrawal Microsoft Research One Microsoft Way Redmond, WA, USA +1-(425) Automati Physial Design Tuning: Workload as a Sequene Sanjay Agrawal Mirosoft Researh One Mirosoft Way Redmond, WA, USA +1-(425) 75-357 sagrawal@mirosoft.om Eri Chu * Computer Sienes Department University

More information

On - Line Path Delay Fault Testing of Omega MINs M. Bellos 1, E. Kalligeros 1, D. Nikolos 1,2 & H. T. Vergos 1,2

On - Line Path Delay Fault Testing of Omega MINs M. Bellos 1, E. Kalligeros 1, D. Nikolos 1,2 & H. T. Vergos 1,2 On - Line Path Delay Fault Testing of Omega MINs M. Bellos, E. Kalligeros, D. Nikolos,2 & H. T. Vergos,2 Dept. of Computer Engineering and Informatis 2 Computer Tehnology Institute University of Patras,

More information

mahines. HBSP enhanes the appliability of the BSP model by inorporating parameters that reet the relative speeds of the heterogeneous omputing omponen

mahines. HBSP enhanes the appliability of the BSP model by inorporating parameters that reet the relative speeds of the heterogeneous omputing omponen The Heterogeneous Bulk Synhronous Parallel Model Tiani L. Williams and Rebea J. Parsons Shool of Computer Siene University of Central Florida Orlando, FL 32816-2362 fwilliams,rebeag@s.uf.edu Abstrat. Trends

More information

Exploring the Commonality in Feature Modeling Notations

Exploring the Commonality in Feature Modeling Notations Exploring the Commonality in Feature Modeling Notations Miloslav ŠÍPKA Slovak University of Tehnology Faulty of Informatis and Information Tehnologies Ilkovičova 3, 842 16 Bratislava, Slovakia miloslav.sipka@gmail.om

More information

NONLINEAR BACK PROJECTION FOR TOMOGRAPHIC IMAGE RECONSTRUCTION. Ken Sauer and Charles A. Bouman

NONLINEAR BACK PROJECTION FOR TOMOGRAPHIC IMAGE RECONSTRUCTION. Ken Sauer and Charles A. Bouman NONLINEAR BACK PROJECTION FOR TOMOGRAPHIC IMAGE RECONSTRUCTION Ken Sauer and Charles A. Bouman Department of Eletrial Engineering, University of Notre Dame Notre Dame, IN 46556, (219) 631-6999 Shool of

More information

Abstract. Key Words: Image Filters, Fuzzy Filters, Order Statistics Filters, Rank Ordered Mean Filters, Channel Noise. 1.

Abstract. Key Words: Image Filters, Fuzzy Filters, Order Statistics Filters, Rank Ordered Mean Filters, Channel Noise. 1. Fuzzy Weighted Rank Ordered Mean (FWROM) Filters for Mixed Noise Suppression from Images S. Meher, G. Panda, B. Majhi 3, M.R. Meher 4,,4 Department of Eletronis and I.E., National Institute of Tehnology,

More information

COST PERFORMANCE ASPECTS OF CCD FAST AUXILIARY MEMORY

COST PERFORMANCE ASPECTS OF CCD FAST AUXILIARY MEMORY COST PERFORMANCE ASPECTS OF CCD FAST AUXILIARY MEMORY Dileep P, Bhondarkor Texas Instruments Inorporated Dallas, Texas ABSTRACT Charge oupled devies (CCD's) hove been mentioned as potential fast auxiliary

More information

A DYNAMIC ACCESS CONTROL WITH BINARY KEY-PAIR

A DYNAMIC ACCESS CONTROL WITH BINARY KEY-PAIR Malaysian Journal of Computer Siene, Vol 10 No 1, June 1997, pp 36-41 A DYNAMIC ACCESS CONTROL WITH BINARY KEY-PAIR Md Rafiqul Islam, Harihodin Selamat and Mohd Noor Md Sap Faulty of Computer Siene and

More information

The Minimum Redundancy Maximum Relevance Approach to Building Sparse Support Vector Machines

The Minimum Redundancy Maximum Relevance Approach to Building Sparse Support Vector Machines The Minimum Redundany Maximum Relevane Approah to Building Sparse Support Vetor Mahines Xiaoxing Yang, Ke Tang, and Xin Yao, Nature Inspired Computation and Appliations Laboratory (NICAL), Shool of Computer

More information

1. Introduction. 2. The Probable Stope Algorithm

1. Introduction. 2. The Probable Stope Algorithm 1. Introdution Optimization in underground mine design has reeived less attention than that in open pit mines. This is mostly due to the diversity o underground mining methods and omplexity o underground

More information

A Novel Validity Index for Determination of the Optimal Number of Clusters

A Novel Validity Index for Determination of the Optimal Number of Clusters IEICE TRANS. INF. & SYST., VOL.E84 D, NO.2 FEBRUARY 2001 281 LETTER A Novel Validity Index for Determination of the Optimal Number of Clusters Do-Jong KIM, Yong-Woon PARK, and Dong-Jo PARK, Nonmembers

More information

Algorithms, Mechanisms and Procedures for the Computer-aided Project Generation System

Algorithms, Mechanisms and Procedures for the Computer-aided Project Generation System Algorithms, Mehanisms and Proedures for the Computer-aided Projet Generation System Anton O. Butko 1*, Aleksandr P. Briukhovetskii 2, Dmitry E. Grigoriev 2# and Konstantin S. Kalashnikov 3 1 Department

More information

A Dual-Hamiltonian-Path-Based Multicasting Strategy for Wormhole-Routed Star Graph Interconnection Networks

A Dual-Hamiltonian-Path-Based Multicasting Strategy for Wormhole-Routed Star Graph Interconnection Networks A Dual-Hamiltonian-Path-Based Multiasting Strategy for Wormhole-Routed Star Graph Interonnetion Networks Nen-Chung Wang Department of Information and Communiation Engineering Chaoyang University of Tehnology,

More information

Contents Contents...I List of Tables...VIII List of Figures...IX 1. Introduction Information Retrieval... 8

Contents Contents...I List of Tables...VIII List of Figures...IX 1. Introduction Information Retrieval... 8 Contents Contents...I List of Tables...VIII List of Figures...IX 1. Introdution... 1 1.1. Internet Information...2 1.2. Internet Information Retrieval...3 1.2.1. Doument Indexing...4 1.2.2. Doument Retrieval...4

More information

HEXA: Compact Data Structures for Faster Packet Processing

HEXA: Compact Data Structures for Faster Packet Processing Washington University in St. Louis Washington University Open Sholarship All Computer Siene and Engineering Researh Computer Siene and Engineering Report Number: 27-26 27 HEXA: Compat Data Strutures for

More information

Learning Convention Propagation in BeerAdvocate Reviews from a etwork Perspective. Abstract

Learning Convention Propagation in BeerAdvocate Reviews from a etwork Perspective. Abstract CS 9 Projet Final Report: Learning Convention Propagation in BeerAdvoate Reviews from a etwork Perspetive Abstrat We look at the way onventions propagate between reviews on the BeerAdvoate dataset, and

More information

Gray Codes for Reflectable Languages

Gray Codes for Reflectable Languages Gray Codes for Refletable Languages Yue Li Joe Sawada Marh 8, 2008 Abstrat We lassify a type of language alled a refletable language. We then develop a generi algorithm that an be used to list all strings

More information

A Load-Balanced Clustering Protocol for Hierarchical Wireless Sensor Networks

A Load-Balanced Clustering Protocol for Hierarchical Wireless Sensor Networks International Journal of Advanes in Computer Networks and Its Seurity IJCNS A Load-Balaned Clustering Protool for Hierarhial Wireless Sensor Networks Mehdi Tarhani, Yousef S. Kavian, Saman Siavoshi, Ali

More information

- 1 - S 21. Directory-based Administration of Virtual Private Networks: Policy & Configuration. Charles A Kunzinger.

- 1 - S 21. Directory-based Administration of Virtual Private Networks: Policy & Configuration. Charles A Kunzinger. - 1 - S 21 Diretory-based Administration of Virtual Private Networks: Poliy & Configuration Charles A Kunzinger kunzinge@us.ibm.om - 2 - Clik here Agenda to type page title What is a VPN? What is VPN Poliy?

More information

arxiv: v1 [cs.db] 13 Sep 2017

arxiv: v1 [cs.db] 13 Sep 2017 An effiient lustering algorithm from the measure of loal Gaussian distribution Yuan-Yen Tai (Dated: May 27, 2018) In this paper, I will introdue a fast and novel lustering algorithm based on Gaussian distribution

More information

Fuzzy Meta Node Fuzzy Metagraph and its Cluster Analysis

Fuzzy Meta Node Fuzzy Metagraph and its Cluster Analysis Journal of Computer Siene 4 (): 9-97, 008 ISSN 549-3636 008 Siene Publiations Fuzzy Meta Node Fuzzy Metagraph and its Cluster Analysis Deepti Gaur, Aditya Shastri and Ranjit Biswas Department of Computer

More information

Partial Character Decoding for Improved Regular Expression Matching in FPGAs

Partial Character Decoding for Improved Regular Expression Matching in FPGAs Partial Charater Deoding for Improved Regular Expression Mathing in FPGAs Peter Sutton Shool of Information Tehnology and Eletrial Engineering The University of Queensland Brisbane, Queensland, 4072, Australia

More information

Make your process world

Make your process world Automation platforms Modion Quantum Safety System Make your proess world a safer plae You are faing omplex hallenges... Safety is at the heart of your proess In order to maintain and inrease your ompetitiveness,

More information

Cluster-Based Cumulative Ensembles

Cluster-Based Cumulative Ensembles Cluster-Based Cumulative Ensembles Hanan G. Ayad and Mohamed S. Kamel Pattern Analysis and Mahine Intelligene Lab, Eletrial and Computer Engineering, University of Waterloo, Waterloo, Ontario N2L 3G1,

More information

Runtime Support for OOLs Part II Comp 412

Runtime Support for OOLs Part II Comp 412 COMP 412 FALL 2017 Runtime Support for OOLs Part II Comp 412 soure IR Front End Optimizer Bak End IR target Copright 2017, Keith D. Cooper & Linda Torzon, all rights reserved. Students enrolled in Comp

More information

Performance Benchmarks for an Interactive Video-on-Demand System

Performance Benchmarks for an Interactive Video-on-Demand System Performane Benhmarks for an Interative Video-on-Demand System. Guo,P.G.Taylor,E.W.M.Wong,S.Chan,M.Zukerman andk.s.tang ARC Speial Researh Centre for Ultra-Broadband Information Networks (CUBIN) Department

More information

Effecting Parallel Graph Eigensolvers Through Library Composition

Effecting Parallel Graph Eigensolvers Through Library Composition Effeting Parallel Graph Eigensolvers Through Library Composition Alex Breuer, Peter Gottshling, Douglas Gregor, Andrew Lumsdaine Open Systems Laboratory Indiana University Bloomington, IN 47405 {abreuer,pgottsh,dgregor,lums@osl.iu.edu

More information

We don t need no generation - a practical approach to sliding window RLNC

We don t need no generation - a practical approach to sliding window RLNC We don t need no generation - a pratial approah to sliding window RLNC Simon Wunderlih, Frank Gabriel, Sreekrishna Pandi, Frank H.P. Fitzek Deutshe Telekom Chair of Communiation Networks, TU Dresden, Dresden,

More information

1. The collection of the vowels in the word probability. 2. The collection of real numbers that satisfy the equation x 9 = 0.

1. The collection of the vowels in the word probability. 2. The collection of real numbers that satisfy the equation x 9 = 0. C HPTER 1 SETS I. DEFINITION OF SET We begin our study of probability with the disussion of the basi onept of set. We assume that there is a ommon understanding of what is meant by the notion of a olletion

More information

CleanUp: Improving Quadrilateral Finite Element Meshes

CleanUp: Improving Quadrilateral Finite Element Meshes CleanUp: Improving Quadrilateral Finite Element Meshes Paul Kinney MD-10 ECC P.O. Box 203 Ford Motor Company Dearborn, MI. 8121 (313) 28-1228 pkinney@ford.om Abstrat: Unless an all quadrilateral (quad)

More information

Algorithms for External Memory Lecture 6 Graph Algorithms - Weighted List Ranking

Algorithms for External Memory Lecture 6 Graph Algorithms - Weighted List Ranking Algorithms for External Memory Leture 6 Graph Algorithms - Weighted List Ranking Leturer: Nodari Sithinava Sribe: Andi Hellmund, Simon Ohsenreither 1 Introdution & Motivation After talking about I/O-effiient

More information

A {k, n}-secret Sharing Scheme for Color Images

A {k, n}-secret Sharing Scheme for Color Images A {k, n}-seret Sharing Sheme for Color Images Rastislav Luka, Konstantinos N. Plataniotis, and Anastasios N. Venetsanopoulos The Edward S. Rogers Sr. Dept. of Eletrial and Computer Engineering, University

More information

Batch Auditing for Multiclient Data in Multicloud Storage

Batch Auditing for Multiclient Data in Multicloud Storage Advaned Siene and Tehnology Letters, pp.67-73 http://dx.doi.org/0.4257/astl.204.50. Bath Auditing for Multilient Data in Multiloud Storage Zhihua Xia, Xinhui Wang, Xingming Sun, Yafeng Zhu, Peng Ji and

More information

Parallelizing Frequent Web Access Pattern Mining with Partial Enumeration for High Speedup

Parallelizing Frequent Web Access Pattern Mining with Partial Enumeration for High Speedup Parallelizing Frequent Web Aess Pattern Mining with Partial Enumeration for High Peiyi Tang Markus P. Turkia Department of Computer Siene Department of Computer Siene University of Arkansas at Little Rok

More information

Methods for Multi-Dimensional Robustness Optimization in Complex Embedded Systems

Methods for Multi-Dimensional Robustness Optimization in Complex Embedded Systems Methods for Multi-Dimensional Robustness Optimization in Complex Embedded Systems Arne Hamann, Razvan Rau, Rolf Ernst Institute of Computer and Communiation Network Engineering Tehnial University of Braunshweig,

More information

RAC 2 E: Novel Rendezvous Protocol for Asynchronous Cognitive Radios in Cooperative Environments

RAC 2 E: Novel Rendezvous Protocol for Asynchronous Cognitive Radios in Cooperative Environments 21st Annual IEEE International Symposium on Personal, Indoor and Mobile Radio Communiations 1 RAC 2 E: Novel Rendezvous Protool for Asynhronous Cognitive Radios in Cooperative Environments Valentina Pavlovska,

More information

System-Level Parallelism and Throughput Optimization in Designing Reconfigurable Computing Applications

System-Level Parallelism and Throughput Optimization in Designing Reconfigurable Computing Applications System-Level Parallelism and hroughput Optimization in Designing Reonfigurable Computing Appliations Esam El-Araby 1, Mohamed aher 1, Kris Gaj 2, arek El-Ghazawi 1, David Caliga 3, and Nikitas Alexandridis

More information

SSD Based First Layer File System for the Next Generation Super-computer

SSD Based First Layer File System for the Next Generation Super-computer SSD Based First Layer File System for the Next Generation Super-omputer Shinji Sumimoto, Ph.D. Next Generation Tehnial Computing Unit FUJITSU LIMITED Sept. 24 th, 2018 0 Outline of This Talk A64FX: High

More information

Plot-to-track correlation in A-SMGCS using the target images from a Surface Movement Radar

Plot-to-track correlation in A-SMGCS using the target images from a Surface Movement Radar Plot-to-trak orrelation in A-SMGCS using the target images from a Surfae Movement Radar G. Golino Radar & ehnology Division AMS, Italy ggolino@amsjv.it Abstrat he main topi of this paper is the formulation

More information

Self-Adaptive Parent to Mean-Centric Recombination for Real-Parameter Optimization

Self-Adaptive Parent to Mean-Centric Recombination for Real-Parameter Optimization Self-Adaptive Parent to Mean-Centri Reombination for Real-Parameter Optimization Kalyanmoy Deb and Himanshu Jain Department of Mehanial Engineering Indian Institute of Tehnology Kanpur Kanpur, PIN 86 {deb,hjain}@iitk.a.in

More information

COMP 181. Prelude. Intermediate representations. Today. Types of IRs. High-level IR. Intermediate representations and code generation

COMP 181. Prelude. Intermediate representations. Today. Types of IRs. High-level IR. Intermediate representations and code generation Prelude COMP 181 Intermediate representations and ode generation November, 009 What is this devie? Large Hadron Collider What is a hadron? Subatomi partile made up of quarks bound by the strong fore What

More information

Cross-layer Resource Allocation on Broadband Power Line Based on Novel QoS-priority Scheduling Function in MAC Layer

Cross-layer Resource Allocation on Broadband Power Line Based on Novel QoS-priority Scheduling Function in MAC Layer Communiations and Networ, 2013, 5, 69-73 http://dx.doi.org/10.4236/n.2013.53b2014 Published Online September 2013 (http://www.sirp.org/journal/n) Cross-layer Resoure Alloation on Broadband Power Line Based

More information

Staircase Join: Teach a Relational DBMS to Watch its (Axis) Steps

Staircase Join: Teach a Relational DBMS to Watch its (Axis) Steps Stairase Join: Teah a Relational DBMS to Wath its (Axis) Steps Torsten Grust Maurie van Keulen Jens Teubner University of Konstanz Department of Computer and Information Siene P.O. Box D 88, 78457 Konstanz,

More information

Reading Object Code. A Visible/Z Lesson

Reading Object Code. A Visible/Z Lesson Reading Objet Code A Visible/Z Lesson The Idea: When programming in a high-level language, we rarely have to think about the speifi ode that is generated for eah instrution by a ompiler. But as an assembly

More information

Space- and Time-Efficient BDD Construction via Working Set Control

Space- and Time-Efficient BDD Construction via Working Set Control Spae- and Time-Effiient BDD Constrution via Working Set Control Bwolen Yang Yirng-An Chen Randal E. Bryant David R. O Hallaron Computer Siene Department Carnegie Mellon University Pittsburgh, PA 15213.

More information

timestamp, if silhouette(x, y) 0 0 if silhouette(x, y) = 0, mhi(x, y) = and mhi(x, y) < timestamp - duration mhi(x, y), else

timestamp, if silhouette(x, y) 0 0 if silhouette(x, y) = 0, mhi(x, y) = and mhi(x, y) < timestamp - duration mhi(x, y), else 3rd International Conferene on Multimedia Tehnolog(ICMT 013) An Effiient Moving Target Traking Strateg Based on OpenCV and CAMShift Theor Dongu Li 1 Abstrat Image movement involved bakground movement and

More information

Query Evaluation Overview. Query Optimization: Chap. 15. Evaluation Example. Cost Estimation. Query Blocks. Query Blocks

Query Evaluation Overview. Query Optimization: Chap. 15. Evaluation Example. Cost Estimation. Query Blocks. Query Blocks Query Evaluation Overview Query Optimization: Chap. 15 CS634 Leture 12 SQL query first translated to relational algebra (RA) Atually, some additional operators needed for SQL Tree of RA operators, with

More information

Dr.Hazeem Al-Khafaji Dept. of Computer Science, Thi-Qar University, College of Science, Iraq

Dr.Hazeem Al-Khafaji Dept. of Computer Science, Thi-Qar University, College of Science, Iraq Volume 4 Issue 6 June 014 ISSN: 77 18X International Journal of Advaned Researh in Computer Siene and Software Engineering Researh Paper Available online at: www.ijarsse.om Medial Image Compression using

More information

COSSIM An Integrated Solution to Address the Simulator Gap for Parallel Heterogeneous Systems

COSSIM An Integrated Solution to Address the Simulator Gap for Parallel Heterogeneous Systems COSSIM An Integrated Solution to Address the Simulator Gap for Parallel Heterogeneous Systems Andreas Brokalakis Synelixis Solutions Ltd, Greee brokalakis@synelixis.om Nikolaos Tampouratzis Teleommuniation

More information

Automated System for the Study of Environmental Loads Applied to Production Risers Dustin M. Brandt 1, Celso K. Morooka 2, Ivan R.

Automated System for the Study of Environmental Loads Applied to Production Risers Dustin M. Brandt 1, Celso K. Morooka 2, Ivan R. EngOpt 2008 - International Conferene on Engineering Optimization Rio de Janeiro, Brazil, 01-05 June 2008. Automated System for the Study of Environmental Loads Applied to Prodution Risers Dustin M. Brandt

More information

Flow Demands Oriented Node Placement in Multi-Hop Wireless Networks

Flow Demands Oriented Node Placement in Multi-Hop Wireless Networks Flow Demands Oriented Node Plaement in Multi-Hop Wireless Networks Zimu Yuan Institute of Computing Tehnology, CAS, China {zimu.yuan}@gmail.om arxiv:153.8396v1 [s.ni] 29 Mar 215 Abstrat In multi-hop wireless

More information

An Efficient and Scalable Approach to CNN Queries in a Road Network

An Efficient and Scalable Approach to CNN Queries in a Road Network An Effiient and Salable Approah to CNN Queries in a Road Network Hyung-Ju Cho Chin-Wan Chung Dept. of Eletrial Engineering & Computer Siene Korea Advaned Institute of Siene and Tehnology 373- Kusong-dong,

More information

Exploiting Enriched Contextual Information for Mobile App Classification

Exploiting Enriched Contextual Information for Mobile App Classification Exploiting Enrihed Contextual Information for Mobile App Classifiation Hengshu Zhu 1 Huanhuan Cao 2 Enhong Chen 1 Hui Xiong 3 Jilei Tian 2 1 University of Siene and Tehnology of China 2 Nokia Researh Center

More information

Reading Object Code. A Visible/Z Lesson

Reading Object Code. A Visible/Z Lesson Reading Objet Code A Visible/Z Lesson The Idea: When programming in a high-level language, we rarely have to think about the speifi ode that is generated for eah instrution by a ompiler. But as an assembly

More information

PathRings. Manual. Version 1.0. Yongnan Zhu December 16,

PathRings. Manual. Version 1.0. Yongnan Zhu   December 16, PathRings Version 1.0 Manual Yongnan Zhu E-mail: yongnan@umb.edu Deember 16, 2014-1 - PathRings This tutorial introdues PathRings, the user interfae and the interation. For better to learn, you will need

More information

arxiv: v1 [cs.gr] 10 Apr 2015

arxiv: v1 [cs.gr] 10 Apr 2015 REAL-TIME TOOL FOR AFFINE TRANSFORMATIONS OF TWO DIMENSIONAL IFS FRACTALS ELENA HADZIEVA AND MARIJA SHUMINOSKA arxiv:1504.02744v1 s.gr 10 Apr 2015 Abstrat. This work introdues a novel tool for interative,

More information

13.1 Numerical Evaluation of Integrals Over One Dimension

13.1 Numerical Evaluation of Integrals Over One Dimension 13.1 Numerial Evaluation of Integrals Over One Dimension A. Purpose This olletion of subprograms estimates the value of the integral b a f(x) dx where the integrand f(x) and the limits a and b are supplied

More information

Total 100

Total 100 CS331 SOLUTION Problem # Points 1 10 2 15 3 25 4 20 5 15 6 15 Total 100 1. ssume you are dealing with a ompiler for a Java-like language. For eah of the following errors, irle whih phase would normally

More information

Acoustic Links. Maximizing Channel Utilization for Underwater

Acoustic Links. Maximizing Channel Utilization for Underwater Maximizing Channel Utilization for Underwater Aousti Links Albert F Hairris III Davide G. B. Meneghetti Adihele Zorzi Department of Information Engineering University of Padova, Italy Email: {harris,davide.meneghetti,zorzi}@dei.unipd.it

More information

DETECTION METHOD FOR NETWORK PENETRATING BEHAVIOR BASED ON COMMUNICATION FINGERPRINT

DETECTION METHOD FOR NETWORK PENETRATING BEHAVIOR BASED ON COMMUNICATION FINGERPRINT DETECTION METHOD FOR NETWORK PENETRATING BEHAVIOR BASED ON COMMUNICATION FINGERPRINT 1 ZHANGGUO TANG, 2 HUANZHOU LI, 3 MINGQUAN ZHONG, 4 JIAN ZHANG 1 Institute of Computer Network and Communiation Tehnology,

More information

A Multi-Head Clustering Algorithm in Vehicular Ad Hoc Networks

A Multi-Head Clustering Algorithm in Vehicular Ad Hoc Networks International Journal of Computer Theory and Engineering, Vol. 5, No. 2, April 213 A Multi-Head Clustering Algorithm in Vehiular Ad Ho Networks Shou-Chih Lo, Yi-Jen Lin, and Jhih-Siao Gao Abstrat Clustering

More information

Computing Pool: a Simplified and Practical Computational Grid Model

Computing Pool: a Simplified and Practical Computational Grid Model Computing Pool: a Simplified and Pratial Computational Grid Model Peng Liu, Yao Shi, San-li Li Institute of High Performane Computing, Department of Computer Siene and Tehnology, Tsinghua University, Beijing,

More information

Accommodations of QoS DiffServ Over IP and MPLS Networks

Accommodations of QoS DiffServ Over IP and MPLS Networks Aommodations of QoS DiffServ Over IP and MPLS Networks Abdullah AlWehaibi, Anjali Agarwal, Mihael Kadoh and Ahmed ElHakeem Department of Eletrial and Computer Department de Genie Eletrique Engineering

More information

Institute for Computer Applications in Science and Engineering NASA Langley Research Center Hampton, VIrginia

Institute for Computer Applications in Science and Engineering NASA Langley Research Center Hampton, VIrginia https://ntrs.nasa.gov/searh.jsp?r=19910002904 2017-10-28T01:31:13+00:00Z NA5A- u- 1

More information

SAND Unlimited Release Printed November 1995 Updated November 29, :26 PM EXODUS II: A Finite Element Data Model

SAND Unlimited Release Printed November 1995 Updated November 29, :26 PM EXODUS II: A Finite Element Data Model SAND92-2137 Unlimited Release Printed November 1995 Updated November 29, 2006 12:26 PM EXODUS II: A Finite Element Data Model Gregory D. Sjaardema (updated version) Larry A. Shoof, Vitor R. Yarberry Computational

More information

Detection and Recognition of Non-Occluded Objects using Signature Map

Detection and Recognition of Non-Occluded Objects using Signature Map 6th WSEAS International Conferene on CIRCUITS, SYSTEMS, ELECTRONICS,CONTROL & SIGNAL PROCESSING, Cairo, Egypt, De 9-31, 007 65 Detetion and Reognition of Non-Oluded Objets using Signature Map Sangbum Park,

More information

Rotation Invariant Spherical Harmonic Representation of 3D Shape Descriptors

Rotation Invariant Spherical Harmonic Representation of 3D Shape Descriptors Eurographis Symposium on Geometry Proessing (003) L. Kobbelt, P. Shröder, H. Hoppe (Editors) Rotation Invariant Spherial Harmoni Representation of 3D Shape Desriptors Mihael Kazhdan, Thomas Funkhouser,

More information

Tackling IPv6 Address Scalability from the Root

Tackling IPv6 Address Scalability from the Root Takling IPv6 Address Salability from the Root Mei Wang Ashish Goel Balaji Prabhakar Stanford University {wmei, ashishg, balaji}@stanford.edu ABSTRACT Internet address alloation shemes have a huge impat

More information

Tree Awareness for Relational DBMS Kernels: Staircase Join

Tree Awareness for Relational DBMS Kernels: Staircase Join Tree Awareness for Relational DBMS Kernels: Stairase Join Torsten Grust 1 and Maurie van Keulen 2 1 Department of Computer and Information Siene, University of Konstanz, P.O. Box D188, 78457 Konstanz,

More information

Torpedo Trajectory Visual Simulation Based on Nonlinear Backstepping Control

Torpedo Trajectory Visual Simulation Based on Nonlinear Backstepping Control orpedo rajetory Visual Simulation Based on Nonlinear Bakstepping Control Peng Hai-jun 1, Li Hui-zhou Chen Ye 1, 1. Depart. of Weaponry Eng, Naval Univ. of Engineering, Wuhan 400, China. Depart. of Aeronautial

More information

Evaluation of Benchmark Performance Estimation for Parallel. Fortran Programs on Massively Parallel SIMD and MIMD. Computers.

Evaluation of Benchmark Performance Estimation for Parallel. Fortran Programs on Massively Parallel SIMD and MIMD. Computers. Evaluation of Benhmark Performane Estimation for Parallel Fortran Programs on Massively Parallel SIMD and MIMD Computers Thomas Fahringer Dept of Software Tehnology and Parallel Systems University of Vienna

More information

Weak Dependence on Initialization in Mixture of Linear Regressions

Weak Dependence on Initialization in Mixture of Linear Regressions Proeedings of the International MultiConferene of Engineers and Computer Sientists 8 Vol I IMECS 8, Marh -6, 8, Hong Kong Weak Dependene on Initialization in Mixture of Linear Regressions Ryohei Nakano

More information

Performance of Histogram-Based Skin Colour Segmentation for Arms Detection in Human Motion Analysis Application

Performance of Histogram-Based Skin Colour Segmentation for Arms Detection in Human Motion Analysis Application World Aademy of Siene, Engineering and Tehnology 8 009 Performane of Histogram-Based Skin Colour Segmentation for Arms Detetion in Human Motion Analysis Appliation Rosalyn R. Porle, Ali Chekima, Farrah

More information

FUZZY WATERSHED FOR IMAGE SEGMENTATION

FUZZY WATERSHED FOR IMAGE SEGMENTATION FUZZY WATERSHED FOR IMAGE SEGMENTATION Ramón Moreno, Manuel Graña Computational Intelligene Group, Universidad del País Vaso, Spain http://www.ehu.es/winto; {ramon.moreno,manuel.grana}@ehu.es Abstrat The

More information

Direct-Mapped Caches

Direct-Mapped Caches A Case for Diret-Mapped Cahes Mark D. Hill University of Wisonsin ahe is a small, fast buffer in whih a system keeps those parts, of the ontents of a larger, slower memory that are likely to be used soon.

More information

Recursion examples: Problem 2. (More) Recursion and Lists. Tail recursion. Recursion examples: Problem 2. Recursion examples: Problem 3

Recursion examples: Problem 2. (More) Recursion and Lists. Tail recursion. Recursion examples: Problem 2. Recursion examples: Problem 3 Reursion eamples: Problem 2 (More) Reursion and s Reursive funtion to reverse a string publi String revstring(string str) { if(str.equals( )) return str; return revstring(str.substring(1, str.length()))

More information

APPENDIX A MPI. A.1.2. Historical perspective Many developers contributed to the MPI library, as shown in Fig. A.1.

APPENDIX A MPI. A.1.2. Historical perspective Many developers contributed to the MPI library, as shown in Fig. A.1. APPENDIX A MPI A.1. An MPI Primer A.1.1. What is MPI? MPI is the standard for multi-omputer and luster message passing introdued by the Message-Passing Interfae Forum in April 1994. The goal of MPI is

More information

the data. Structured Principal Component Analysis (SPCA)

the data. Structured Principal Component Analysis (SPCA) Strutured Prinipal Component Analysis Kristin M. Branson and Sameer Agarwal Department of Computer Siene and Engineering University of California, San Diego La Jolla, CA 9193-114 Abstrat Many tasks involving

More information

Reducing Runtime Complexity of Long-Running Application Services via Dynamic Profiling and Dynamic Bytecode Adaptation for Improved Quality of Service

Reducing Runtime Complexity of Long-Running Application Services via Dynamic Profiling and Dynamic Bytecode Adaptation for Improved Quality of Service Reduing Runtime Complexity of Long-Running Appliation Servies via Dynami Profiling and Dynami Byteode Adaptation for Improved Quality of Servie ABSTRACT John Bergin Performane Engineering Laboratory University

More information

Distributed Resource Allocation Strategies for Achieving Quality of Service in Server Clusters

Distributed Resource Allocation Strategies for Achieving Quality of Service in Server Clusters Proeedings of the 45th IEEE Conferene on Deision & Control Manhester Grand Hyatt Hotel an Diego, CA, UA, Deember 13-15, 2006 Distributed Resoure Alloation trategies for Ahieving Quality of ervie in erver

More information

Parametric Abstract Domains for Shape Analysis

Parametric Abstract Domains for Shape Analysis Parametri Abstrat Domains for Shape Analysis Xavier RIVAL (INRIA & Éole Normale Supérieure) Joint work with Bor-Yuh Evan CHANG (University of Maryland U University of Colorado) and George NECULA (University

More information

The recursive decoupling method for solving tridiagonal linear systems

The recursive decoupling method for solving tridiagonal linear systems Loughborough University Institutional Repository The reursive deoupling method for solving tridiagonal linear systems This item was submitted to Loughborough University's Institutional Repository by the/an

More information

次世代スーパーコンピュータ向け ファイルシステムについて

次世代スーパーコンピュータ向け ファイルシステムについて Gfarm シンポジウム 2018 次世代スーパーコンピュータ向け ファイルシステムについて Shinji Sumimoto, Ph.D. Next Generation Tehnial Computing Unit FUJITSU LIMITED Ot. 26 th, 2018 0 Outline of This Talk A64FX: High Performane Arm CPU Next Generation

More information

Multiple-Criteria Decision Analysis: A Novel Rank Aggregation Method

Multiple-Criteria Decision Analysis: A Novel Rank Aggregation Method 3537 Multiple-Criteria Deision Analysis: A Novel Rank Aggregation Method Derya Yiltas-Kaplan Department of Computer Engineering, Istanbul University, 34320, Avilar, Istanbul, Turkey Email: dyiltas@ istanbul.edu.tr

More information

Establishing Secure Ethernet LANs Using Intelligent Switching Hubs in Internet Environments

Establishing Secure Ethernet LANs Using Intelligent Switching Hubs in Internet Environments Establishing Seure Ethernet LANs Using Intelligent Swithing Hubs in Internet Environments WOEIJIUNN TSAUR AND SHIJINN HORNG Department of Eletrial Engineering, National Taiwan University of Siene and Tehnology,

More information

Announcements. Lecture Caching Issues for Multi-core Processors. Shared Vs. Private Caches for Small-scale Multi-core

Announcements. Lecture Caching Issues for Multi-core Processors. Shared Vs. Private Caches for Small-scale Multi-core Announements Your fous should be on the lass projet now Leture 17: Cahing Issues for Multi-ore Proessors This week: status update and meeting A short presentation on: projet desription (problem, importane,

More information

Graph-Based vs Depth-Based Data Representation for Multiview Images

Graph-Based vs Depth-Based Data Representation for Multiview Images Graph-Based vs Depth-Based Data Representation for Multiview Images Thomas Maugey, Antonio Ortega, Pasal Frossard Signal Proessing Laboratory (LTS), Eole Polytehnique Fédérale de Lausanne (EPFL) Email:

More information

Uplink Channel Allocation Scheme and QoS Management Mechanism for Cognitive Cellular- Femtocell Networks

Uplink Channel Allocation Scheme and QoS Management Mechanism for Cognitive Cellular- Femtocell Networks 62 Uplink Channel Alloation Sheme and QoS Management Mehanism for Cognitive Cellular- Femtoell Networks Kien Du Nguyen 1, Hoang Nam Nguyen 1, Hiroaki Morino 2 and Iwao Sasase 3 1 University of Engineering

More information

Run Time Environment. Implementing Object-Oriented Languages

Run Time Environment. Implementing Object-Oriented Languages Run Time Environment Implementing Objet-Oriented Languages Copright 2017, Pedro C. Diniz, all rights reserved. Students enrolled in the Compilers lass at the Universit of Southern California have epliit

More information

Boosted Random Forest

Boosted Random Forest Boosted Random Forest Yohei Mishina, Masamitsu suhiya and Hironobu Fujiyoshi Department of Computer Siene, Chubu University, 1200 Matsumoto-ho, Kasugai, Aihi, Japan {mishi, mtdoll}@vision.s.hubu.a.jp,

More information

EXODUS II: A Finite Element Data Model

EXODUS II: A Finite Element Data Model SAND92-2137 Unlimited Release Printed November 1995 Distribution Category UC-705 EXODUS II: A Finite Element Data Model Larry A. Shoof, Vitor R. Yarberry Computational Mehanis and Visualization Department

More information

'* ~rr' _ ~~ f' lee : eel. Series/1 []J 0 [[] "'l... !l]j1. IBM Series/1 FORTRAN IV. I ntrod uction ...

'* ~rr' _ ~~ f' lee : eel. Series/1 []J 0 [[] 'l... !l]j1. IBM Series/1 FORTRAN IV. I ntrod uction ... ---- --- - ----- - - - --_.- --- Series/1 GC34-0132-0 51-25 PROGRAM PRODUCT 1 IBM Series/1 FORTRAN IV I ntrod ution Program Numbers 5719-F01 5719-F03 0 lee : eel II 11111111111111111111111111111111111111111111111

More information

Facility Location: Distributed Approximation

Facility Location: Distributed Approximation Faility Loation: Distributed Approximation Thomas Mosibroda Roger Wattenhofer Distributed Computing Group PODC 2005 Where to plae ahes in the Internet? A distributed appliation that has to dynamially plae

More information

Multi-hop Fast Conflict Resolution Algorithm for Ad Hoc Networks

Multi-hop Fast Conflict Resolution Algorithm for Ad Hoc Networks Multi-hop Fast Conflit Resolution Algorithm for Ad Ho Networks Shengwei Wang 1, Jun Liu 2,*, Wei Cai 2, Minghao Yin 2, Lingyun Zhou 2, and Hui Hao 3 1 Power Emergeny Center, Sihuan Eletri Power Corporation,

More information

Chapter 2: Introduction to Maple V

Chapter 2: Introduction to Maple V Chapter 2: Introdution to Maple V 2-1 Working with Maple Worksheets Try It! (p. 15) Start a Maple session with an empty worksheet. The name of the worksheet should be Untitled (1). Use one of the standard

More information

Design Implications for Enterprise Storage Systems via Multi-Dimensional Trace Analysis

Design Implications for Enterprise Storage Systems via Multi-Dimensional Trace Analysis Design Impliations for Enterprise Storage Systems via Multi-Dimensional Trae Analysis Yanpei Chen, Kiran Srinivasan, Garth Goodson, Randy Katz University of California, Berkeley, NetApp In. {yhen2, randy}@ees.berkeley.edu,

More information

Volume 3, Issue 9, September 2013 International Journal of Advanced Research in Computer Science and Software Engineering

Volume 3, Issue 9, September 2013 International Journal of Advanced Research in Computer Science and Software Engineering Volume 3, Issue 9, September 2013 ISSN: 2277 128X International Journal of Advaned Researh in Computer Siene and Software Engineering Researh Paper Available online at: www.ijarsse.om A New-Fangled Algorithm

More information