Data Layout Optimization for Petascale File Systems
|
|
- Lora Austin
- 8 years ago
- Views:
Transcription
1 Data Layout Optimizatio for Petacale File Sytem Xia-He Su Illioi Ititute of Techology Yog Che Illioi Ititute of Techology Yalog Yi Illioi Ititute of Techology ABSTRACT I thi tudy, the author propoe a imple performace model to promote a better itegratio betwee the parallel I/O middleware layer ad parallel file ytem. They how that applicatiopecific data layout optimizatio ca improve overall data acce delay coiderably for may applicatio. Implemetatio reult uder MPI-IO middleware ad PVFS file ytem cofirm the correcte ad effectivee of their approach, ad demotrate the potetial of data layout optimizatio i petacale data torage. Categorie ad Subject Decriptor B..3 [Iput/Output ad Data Commuicatio]: Itercoectio (Subytem) parallel I/O. D.. [Operatig Sytem]: Storage Maagemet allocatio/deallocatio trategie, ecodary torage. Geeral Term Meauremet, Performace, Experimetatio. Keyword Data layout, parallel file ytem, parallel I/O 1. INTRODUCTION High-performace computig (HPC) ha croed the Petaflop mark ad i movig forward to reach the Exaflop rage [15]. However, while computig reource are makig rapid progre, there i a igificat gap betwee proceig capacity ad dataacce performace. Due to thi gap, although proceig reource are available, they have to tay idle waitig for data to arrive, which lead to a evere overall performace degradatio. Figure 1 how the umber of CPU cycle required to acce cache memory (SRAM), mai memory (DRAM), ad dik torage []. It ca be ee that the umber of cycle for acceig dik i hudred of thouad of time lower. Thi tred i predicted to cotiue i the ear future. I the meatime, applicatio are becomig more ad more data iteive. Due to the growig performace diparity ad emergig data iteive applicatio, I/O ad torage have become a critical performace bottleeck i HPC machie, epecially whe we are dealig with petacale Permiio to make digital or hard copie of all or part of thi work for peroal or claroom ue i grated without fee provided that copie are ot made or ditributed for profit or commercial advatage ad that copie bear thi otice ad the full citatio o the firt page. To copy otherwie, to republih, to pot o erver or to reditribute to lit, require prior pecific permiio ad/or a fee. Supercomputig PDSW'9, Nov. 15, 9. Portlad, OR, USA. Copyright 9 ACM /9/11... $1. data torage. Data layout mechaim decide how data i ditributed amog multiple file erver. It i a crucial factor that decide the data acce latecy ad the I/O ubytem performace for highperformace computig. The recet work i log-like reorderig of data [1][7] ha demotrated the importace ad performace improvemet by arragig data i a proper maer. However, hitorically, parallel I/O middleware ytem, uch a ROMIO [], ad parallel file ytem are developed eparately with a implified modular deig i mid. Parallel I/O middleware ytem ofte aume the uderlyig i a big file ytem, ad, o the other had, parallel file ytem ofte rely o the I/O middleware for data acce optimizatio ad do little i data layout optimizatio. I thi tudy, we argue that purely depedig o I/O middleware for data retrieval optimizatio i cotly ad may ot be effective i may ituatio. We argue that if we pa ome of the applicatio-pecific I/O requet iformatio to file ytem for data layout optimizatio, the reult could be much better. Exitig parallel file ytem, uch a PVFS [3], Lutre [5], ad GPFS [11] provide high badwidth for imple, well-formed, ad geeric I/O acce characteritic, but their performace varie from applicatio to applicatio [][]. Tuig data layout accordig to pecific I/O acce patter for a parallel I/O ytem i a eceity. Thi tuig require udertadig file ytem abtractio, gaiig kowledge of dik torage, kowig the deig of high-level librarie, ad makig itelliget deciio. While PVFS ad high-level parallel I/O librarie, uch a MPI-IO [13] ad HDF-5 [] provide ome fuctioality to cutomize data layout accordig to pecific I/O workload, few kow how to ue them effectively. Cycle 1M 1M 1 1K 1K Data Acce Time i CPU Cycle Year SRAM DRAM Dik Figure 1. Compario of data acce latecy. I thi reearch, we tudy data layout optimizatio of parallel file ytem. We how that, with the coideratio of applicatio-
2 pecific I/O requet, the data layout optimizatio ca be totally differet i a parallel file ytem. We preet a ytem-level applicatio-pecific data layout optimizatio trategy for petacale data torage. By ytem-level, we mea that the propoed approach i itegrated ito the file ytem ad i traparet to programmer ad uer. By applicatio-pecific, we mea that the propoed approach ca adapt to pecific data acce patter for a proper data layout. The cotributio of thi tudy i two fold. Firt, we how that the data layout optimizatio ha a igificat impact o petacale data torage performace. Secod, we demotrate with a imple performace model ad curret imple data layout fuctioalitie provided by PVFS that we ca achieve oticeable performace gai. While our reult are prelimiary, they demotrate the potetial of the data layout optimizatio approach.. APPLICATION-SPECIFIC DATA LAYOUT MODELING Modelig ad evaluatig the performace of data layout trategy i eetial i providig a applicatio-pecific data layout optimizatio. The covetioal roud-robi ditributio (referred to a imple tripig i ome exitig work) i i place i may of parallel file ytem [3][5][11]. However, uder parallel I/O ytem, thi imple ditributio may ot be the bet data layout ad ca be improved. We preet a imple data layout performace model herei. I thi model, we aume that the coectio betwee compute (I/O) ode ad file erver i ot a performace bottleeck ad that the igificat overhead i i acceig file erver. We further aume that each file erver performace ca be meaured a α+β, where α i the tart up time (latecy), i the data ize, ad β i the tramiio time of igle uit data (the reciprocal of tramiio rate). I thi model, we differetiate three data layout trategie, 1-D Horizotal Layout, 1-D Vertical Layout ad -D Layout. The 1-D Horizotal Layout (or 1-DH i hort) refer to the trategy that data i ditributed amog all available file erver i a traditioal roud-robi fahio. Thi layout matche with the exitig imple tripig or roud-robi trategy. The 1-D Vertical Layout (or 1- DV i hort) refer to the trategy that data to be acceed by each proce i tored o oe give file erver. The -D Layout (or -D i hort) i the trategy i which data to be acceed by each proce i tored o a ubet of file erver. Figure illutrate thee three trategie with a example. Aume that we have p computig (I/O) ode, file erver, where all computig ode participate i a SPMD form of parallel computig, with a block-cyclic or ome imilar, eve data partitioig. With 1-DH data layout, i.e., with imple tripig roud-robi layout where exactly / of the data are i ay of the file erver, the cot of acceig data of ize by oe proce ad p procee are: ( α + β) ad p ( α + β) = p p α + β (1) repectively. With the 1-DH layout, each proce accee it required data cocurretly, but multiple procee have to acce data oe by oe equetially; ad the data of each proce i ditributed over differet file erver evely. Thi trategy make accee i a equetial cocurret way. The value of Equatio (1) deped o the value of p,, α ad β. I ay cae, however, the 1-DH layout or the covetioal roud-robi layout may ot be the bet choice whe p. If we take the 1-DV layout, i.e. takig a cocurret equetial approach, we ca get a better p performace, with ( ) α + β. If p <, the the data ca be tored either o erver uig 1-DH layout or uig -D layout, where each of the p procee get /p file erver for data torage. For the former layout, the cot i α+β ad for the latter cae, the cot i α + β. p P P1 P P3 p p( α + ) p β = α + β α + β or p ( α + β) Figure. Data layout trategie. α + β p 3. APPLICATION-SPECIFIC DATA LAYOUT OPTIMIZATION With applicatio-pecific data layout modelig, we are able to guide data layout toward a better way by coiderig data acce characteritic. With the value of p, ad α, β, the proper data layout ca be determied with the aforemetioed data layout formula for a give parallel I/O requet. The data layout of a give applicatio ca the be determied baed o the weighted ummatio of the cot of it I/O requet. The above model i determiitic ad i ready to ue uder exitig parallel file ytem, uch a PVFS. I additio, i parallel I/O applicatio, it i commo that a applicatio accee multiple file ad each file i multiple occaio. We tore each file i a differet layout to improve performace. Whe a applicatio accee a file i multiple patter, it i eceary to fid a layout that i beeficial for all patter. For example, a file i read i cotiguou acce patter ad writte i a complex o-cotiguou patter. From may obervatio [][13], acceig data i o-cotiguou patter perform wore tha acceig cotiguouly. Storig data to facilitate o-cotiguou accee may deteriorate cotiguou acce performace. We have to fid a balace betwee performace beefit whe we decide o performace layout. Baed o patter aalyi, we ca utilize a trategy by aigig each patter a weight to repreet it cope for I/O performace improvemet.
3 Baed o the modelig ad obervatio, we defie a et of data layout heuritic a how i Table 1. Whe I/O acce characteritic are ukow or completely radom, we rely o 1- DH trategy or the default imple roud-robi trategy. Whe the degree of I/O cocurrecy i high, it i beeficial to ue 1-DV layout. 1-DH layout or -D layout ca be cofigured for low degree of cocurrecy. I cae of TCP Icat [1], it i better to tripe data amog a certai et of file erver itead of all available file erver, which i -D layout. File ytem uch a PVFS provide feature to exted ad create ew ditributio [9]. We utilize thee feature i geeratig ew applicatiopecific ditributio i our implemetatio. Table 1. Heuritic for Chooig Layout Acce Patter Feature Radom High degree of I/O cocurrecy Low degree of I/O cocurrecy Too may I/O erver o TCP/IP Data Layout Strategy 1-DH (roud-robi) layout 1-DV data layout 1-DH or -D data layout -D data layout After makig deciio o the layout, we tore data o file erver uig the ew layout. The 1-DH layout trategy, or the imple roud-robi layout, with differet tripe ize ad tripig factor ca be et with MPI-IO hit, uch a tripig_factor ad tripig_uit. A more complex ditributio, uch a 1-DV or -D data layout, eed to be modified at the file ytem level to provide geeral upport, but ca be emulated with differet tripig_factor ad tripig_uit cofiguratio. I additio, it i commo for parallel file ytem, uch a PVFS, to provide flexible ad extedable data ditributio [9]. PVFS iclude a modular ytem for addig ew data ditributio to the ytem ad uig thee for ew file ad optimized layout. Sice our curret implemetatio focue o prototypig the idea ad verifyig the potetial performace gai, we employ a relatively quick prototypig trategy by uig parallel file ytem cofiguratio to provide upport for variou layout trategie. The curret prototypig ytem ha demotrated a igificat performace improvemet over exitig trategie a the followig ectio how. A geeral full-fledged data layout trategy upport at parallel fileytem level i uder developmet a well.. PRELIMINARY EXPERIMENTAL RESULTS We have carried out a prototype implemetatio of applicatiopecific data layout o PVFS parallel file ytem baed o the previouly dicued model ad optimizatio trategy. We curretly upport three trategie, 1-DH, 1-DV ad -D layout. The followig ubectio preet the iitial experimetal reult of thee applicatio-pecific trategie uder differet ceario..1 Experimetal Setup Our experimet were coducted o a 17-ode Dell PowerEdge Liux-baed cluter ad a 5-ode Su Fire Liux-baed cluter. The Dell cluter i compoed of oe Dell PowerEdge 5 head ode, with dual. GHz Xeo proceor ad GB memory, ad 1 Dell PowerEdge 5 compute ode with dual 3. GHz Xeo proceor ad 1 GB memory. The head ode ha two 73 GB U3 1K-RPM SCSI drive. Each compute ode ha a GB 7.K-RPM SATA hard drive. The Su cluter i compoed of oe Su Fire X head ode, with dual.7 GHz Optero quad-core proceor ad GB memory, ad Su Fire X compute ode with dual.3ghz Optero quad-core proceor ad GB memory. The head ode ha 5GB 7.K-RPM SATA-II drive cofigured a RAID-5 ytem. Each compute ode ha a 5GB 7.K-RPM SATA hard drive. The experimet were teted o PVFS file ytem. For the Dell cluter, PVFS wa cofigured with oe metadata erver ode, the head ode, ad I/O erver ode. All ode are ued a compute ode. For the Su Fire cluter, PVFS wa cofigured with 3 I/O erver ode. The ret ode are ued a compute ode.. Experimetal Reult ad Aalye..1 Sythetic Bechmark We have coded a ythetic bechmark which doe equetial read over the file tored with differet layout. We have performed a erie of tet o the Dell cluter. The firt et of experimet coducted i to compare the performace of differet layout trategie with four compute procee. I thi ceario, four procee retrieve data from MB, 1MB, 3MB, MB ad MB file repectively. Thee file are tored o eight file erver with three layout, 1-DH, 1-DV ad -D. We meaured the performace of retrievig data i each cae ad the reult are how i Figure 3. The reported reult are the average of three ru. We fluhed the ytem buffer cache betwee each ru. 1 MB 1MB 3MB MB MB File Size Figure 3. I/O performace with differet layout trategie. Figure 3 clearly how that differet layout trategie do have a coiderable impact to the performace of parallel I/O ytem. Amog three layout, the -D layout achieved the bet performace i all cae. Thi i coitet with our model ad aalyi that the -D layout i deired whe the umber of compute procee i le tha that of I/O erver ode. I the meatime, the 1-DH layout, or the default roud-robi layout, performed wore tha both 1-DV ad -D layout, ad the performace diparity wa up to.%.
4 We have alo performed a detailed aalyi to verify the propoed model. We compute the theoretical value with the model ad the meaured dik trafer time ad tartup time. The theoretical ad experimetal reult are how i Figure (1-DH layout i omitted here due to the pace limit). A ca be ee from the reult, there i a cloe match betwee the experimetal reult ad theoretical reult, which how the model ca etimate the performace of thee layout trategie well Fileize (MB) Figure. Radom read/write with KB tripe ize. MB 1MB 3MB MB MB MB 1MB 3MB MB MB 3 File ize File ize Experimetal Theoretical Experimetal Theoretical Figure. Experimetal ad theoretical reult. (Left: 1-DV layout; Right: -D layout) The other et of experimet we have coducted i to compare the impact of layout trategie with 1 compute procee. Thi et of tet i imilar with the previou tet, but the file ize are doubled i order to compare the performace with variou file ize. The reult how that 1-DV layout outperformed the other two trategie i all cae, which i coitet with the model ad aalyi preeted i Sectio. The reult are how i Figure Fileize (MB) Figure 7. Sequetial read/write with KB tripe ize MB 3MB MB MB MB File Size File Size (MB) Figure. Radom read/write with 1MB tripe ize. 3 Figure 5. I/O performace with differet layout trategie... IOR Bechmark I additio to the ythetic bechmark meauremet, we have performed a erie of tetig o the Su cluter with the IOR-.1. bechmark from Lawrece Livermore Natioal Laboratory []. I thee experimet, we performed a larger cale of tetig. We cofigured PVFS with 3 I/O erver ode ad ru tetig with procee o 3 cliet ode (cliet ode are eparate from I/O erver ode). We performed both equetial read/write ad radom read/write tet, ad varied the tripe ize ad the file ize. Figure ad Figure 7 report the badwidth reult of acceig file with differet layout i a radom or equetial maer, repectively, with KB tripe ize for 1-DH ad -D layout. Figure ad Figure 9 report the reult i a imilar ceario, but with 1MB tripe ize for 1-DH ad -D layout File Size (MB) Figure 9. Sequetial read/write with 1MB tripe ize. A ca be ee from thee reult, differet layout trategie ca affect the IOR bechmark tetig performace coiderably. Amog the three trategie we pecifically aalyze, the 1-DV trategy geerally perform better tha the other two, while the - D trategy perform better tha the 1-DH trategy.
5 Although the curret experimetal reult are prelimiary, they have demotrated that data layout trategie have a coiderable impact o parallel I/O ytem. The propoed model ad applicatio-pecific data layout optimizatio are deired to dyamically adapt the layout to achieve a better performace uder differet ceario. 5. ONGOING WORK We have reported ome of iitial reult, while everal tudie are ogoig ad are ot ready to report at thi time. For itace, we are workig o a compreheive data layout model to characterize the performace impact of layout trategy i geeral cae baed o probability ad queuig theory. The baic idea of the geeral model i that each I/O ode ca be modeled a a idepedet queue. I/O requet come ito thee queue ad are erviced for either torig or retrievig data. Whe cotetio occur, the requet ha to wait i the queue to be erviced. Multiple queue are idepedet from each other, ad data layout optimizatio o parallel file erver are derived accordigly. Thi model characterize cocurrecy (parallelim) ad cotetio, two major role that data layout trategy play i affectig the ytem performace, to guide a optimal layout electio. We have developed a theoretical model ad are workig o the experimetal part to verify the model. We are alo movig the experimetal tetig to a much larger computer cluter tha what we have ued.. CONCLUSION Parallel I/O middleware ad parallel file ytem are fudametal ad critical compoet for petacale torage. While both of the techologie have made their ucce, little ha bee doe to applicatio-pecific data layout. I mot exitig file ytem, data i ditributed amog multiple erver primarily with a imple roud-robi trategy. Thi imple data layout trategy doe ot alway work well for parallel I/O ytem, where I/O requet are geerated cocurretly. I thi tudy, we have propoed a applicatio-pecific data layout trategy to optimize the performace of acceig data accordig to ditict applicatio feature. Thi data layout trategy optimizatio i built upo a imple but effective data layout model, ad ha bee prototyped with the cofiguratio facility of the uderlyig PVFS parallel file ytem. Parallel file ytem have bee deiged a oe-et-for-all ad have bee tatic. There i a great eed for reearch ito extgeeratio I/O architecture to upport acce awaree, itelligece, ad applicatio-pecific adaptive data ditributio ad reditributio. Although our curret reult are very limited, our prototypig ytem ha demotrated the great potetial i improvig parallel I/O acce performace via data layout optimizatio whe acce characteritic are take ito coideratio. We believe that the applicatio-pecific data layout optimizatio approach eed a commuity attetio. Thi approach appear to be a feaible olutio to mitigatig the I/O wall problem, epecially for petacale data torage. 7. ACKNOWLEDGMENTS The author are thakful to Dr. Rajeev Thakur, Dr. Rob Ro ad Sam Lag of Argoe Natioal Laboratory for their cotructive ad thoughtful uggetio toward thi tudy. Thi reearch wa upported i part uder NSF grat CCF-35 ad CCF REFERENCES [1] J. Bet, G. Gibo, G. Grider, B. McClellad, P. Nowoczyki, J. Nuez, M. Polte, M. Wigate, PLFS: A Checkpoit Fileytem for Parallel Applicatio, i Proc. of ACM/IEEE SuperComputig'9. [] R. E. Bryat ad D. O'Hallaro, Computer Sytem: A Programmer' Perpective, Pretice-Hall, 3. [3] P. H. Car, W. B. Ligo III, R. B. Ro, ad R. Thakur, PVFS: A Parallel File Sytem For Liux Cluter, i Proceedig of the th Aual Liux Showcae ad Coferece,. [] P. E. Cradall, R. A. Aydt, A. A. Chie, ad D. A. Reed, Iput/Output Characteritic of Scalable Parallel Applicatio, i Proceedig of the ACM/IEEE Coferece o Supercomputig, [5] Cluter File Sytem Ic., Lutre: A Scalable, High Performace File Sytem, Whitepaper, [] Iterleaved or Radom (IOR) Bechmark, [7] J. F. Loftead, S. Klaky, K. Schwa, N. Podhorzki ad C. Ji, Flexible IO ad Itegratio for Scietific Code Through the Adaptable IO Sytem (ADIOS), i Proc. of the th Iteratioal Workhop o Challege of Large Applicatio i Ditributed Eviromet,. [] J. May, Parallel I/O for High Performace Computig, Morga Kaufma Publihig, 1. [9] PVFS Developmet Team, PVFS Developer' Guide, guide.pdf. [1] A. Phaihayee, E. Krevat, V. Vaudeva, D. Adere, G. Gager, G. Gibo ad S. Seha, Meauremet ad Aalyi of TCP Throughput Collape i Cluter-Baed Storage Sytem, i Proceedig of File ad Storage Techologie (FAST),. [11] F. Schmuck ad R. Haki, GPFS: A Shared-Dik File Sytem for Large Computig Cluter, i 1 t USENIX Coferece o File ad Storage Techologie, USENIX,. [] The HDF5 Project, HDF5 - A New Geeratio of HDF, NCSA, Uiv. of Illioi at Urbaa Champaig. Available at [13] R. Thakur, W. Gropp ad E. Luk, Optimizig Nocotiguou Accee i MPI-IO, Parallel Computig, ()1:3-15,. [] R. Thakur, W. Gropp ad E. Luk, Data Sievig ad Collective I/O i ROMIO, i Proceedig of the 7th Sympoium o the Frotier of Maively Parallel Computatio, [15] Top 5 Supercomputig Webite.
Confidence Intervals for Linear Regression Slope
Chapter 856 Cofidece Iterval for Liear Regreio Slope Itroductio Thi routie calculate the ample ize eceary to achieve a pecified ditace from the lope to the cofidece limit at a tated cofidece level for
More informationTopic 5: Confidence Intervals (Chapter 9)
Topic 5: Cofidece Iterval (Chapter 9) 1. Itroductio The two geeral area of tatitical iferece are: 1) etimatio of parameter(), ch. 9 ) hypothei tetig of parameter(), ch. 10 Let X be ome radom variable with
More informationDomain 1: Designing a SQL Server Instance and a Database Solution
Maual SQL Server 2008 Desig, Optimize ad Maitai (70-450) 1-800-418-6789 Domai 1: Desigig a SQL Server Istace ad a Database Solutio Desigig for CPU, Memory ad Storage Capacity Requiremets Whe desigig a
More informationModified Line Search Method for Global Optimization
Modified Lie Search Method for Global Optimizatio Cria Grosa ad Ajith Abraham Ceter of Excellece for Quatifiable Quality of Service Norwegia Uiversity of Sciece ad Techology Trodheim, Norway {cria, ajith}@q2s.tu.o
More information(VCP-310) 1-800-418-6789
Maual VMware Lesso 1: Uderstadig the VMware Product Lie I this lesso, you will first lear what virtualizatio is. Next, you ll explore the products offered by VMware that provide virtualizatio services.
More informationDetermining the sample size
Determiig the sample size Oe of the most commo questios ay statisticia gets asked is How large a sample size do I eed? Researchers are ofte surprised to fid out that the aswer depeds o a umber of factors
More informationCluster-Aware Cache for Network Attached Storage *
Cluter-Aware Cache for Network Attached Storage * Bin Cai, Changheng Xie, and Qiang Cao National Storage Sytem Laboratory, Department of Computer Science, Huazhong Univerity of Science and Technology,
More informationUnicenter TCPaccess FTP Server
Uiceter TCPaccess FTP Server Release Summary r6.1 SP2 K02213-2E This documetatio ad related computer software program (hereiafter referred to as the Documetatio ) is for the ed user s iformatioal purposes
More informationDomain 1 - Describe Cisco VoIP Implementations
Maual ONT (642-8) 1-800-418-6789 Domai 1 - Describe Cisco VoIP Implemetatios Advatages of VoIP Over Traditioal Switches Voice over IP etworks have may advatages over traditioal circuit switched voice etworks.
More information1 Computing the Standard Deviation of Sample Means
Computig the Stadard Deviatio of Sample Meas Quality cotrol charts are based o sample meas ot o idividual values withi a sample. A sample is a group of items, which are cosidered all together for our aalysis.
More informationReliability Analysis in HPC clusters
Reliability Aalysis i HPC clusters Narasimha Raju, Gottumukkala, Yuda Liu, Chokchai Box Leagsuksu 1, Raja Nassar, Stephe Scott 2 College of Egieerig & Sciece, Louisiaa ech Uiversity Oak Ridge Natioal Lab
More informationA Spam Message Filtering Method: focus on run time
, pp.29-33 http://dx.doi.org/10.14257/atl.2014.76.08 A Spam Meage Filtering Method: focu on run time Sin-Eon Kim 1, Jung-Tae Jo 2, Sang-Hyun Choi 3 1 Department of Information Security Management 2 Department
More informationOn k-connectivity and Minimum Vertex Degree in Random s-intersection Graphs
O k-coectivity ad Miimum Vertex Degree i Radom -Iterectio Graph Ju Zhao Oma Yağa Virgil Gligor CyLab ad Dept. of ECE Caregie Mello Uiverity Email: {juzhao, oyaga, virgil}@adrew.cmu.edu Abtract Radom -iterectio
More information.04. This means $1000 is multiplied by 1.02 five times, once for each of the remaining sixmonth
Questio 1: What is a ordiary auity? Let s look at a ordiary auity that is certai ad simple. By this, we mea a auity over a fixed term whose paymet period matches the iterest coversio period. Additioally,
More informationA Software System for Optimal Virtualization of a Server Farm 1
БЪЛГАРСКА АКАДЕМИЯ НА НАУКИТЕ BULGARIAN ACADEMY OF SCIENCES ПРОБЛЕМИ НА ТЕХНИЧЕСКАТА КИБЕРНЕТИКА И РОБОТИКАТА, 60 PROBLEMS OF ENGINEERING CYBERNETICS AND ROBOTICS, 60 София 2009 Sofia A Software System
More informationQuantitative Computer Architecture
Performace Measuremet ad Aalysis i Computer Quatitative Computer Measuremet Model Iovatio Proposed How to measure, aalyze, ad specify computer system performace or My computer is faster tha your computer!
More informationLECTURE 13: Cross-validation
LECTURE 3: Cross-validatio Resampli methods Cross Validatio Bootstrap Bias ad variace estimatio with the Bootstrap Three-way data partitioi Itroductio to Patter Aalysis Ricardo Gutierrez-Osua Texas A&M
More information5.4 Amortization. Question 1: How do you find the present value of an annuity? Question 2: How is a loan amortized?
5.4 Amortizatio Questio 1: How do you fid the preset value of a auity? Questio 2: How is a loa amortized? Questio 3: How do you make a amortizatio table? Oe of the most commo fiacial istrumets a perso
More informationBaanERP. BaanERP Windows Client Installation Guide
BaaERP A publicatio of: Baa Developmet B.V. P.O.Box 143 3770 AC Bareveld The Netherlads Prited i the Netherlads Baa Developmet B.V. 1999. All rights reserved. The iformatio i this documet is subject to
More informationComparative Analysis of Round Robin VM Load Balancing With Modified Round Robin VM Load Balancing Algorithms in Cloud Computing
Iteratioal Joural of Egieerig, Maagemet & Scieces (IJEMS) Comparative Aalysis of Roud Robi Balacig With Modified Roud Robi Balacig s i Cloud Computig Areeba Samee, D.K Budhwat Abstract Cloud computig is
More information3D BUILDING MODEL RECONSTRUCTION FROM POINT CLOUDS AND GROUND PLANS
3D BUILDING MODEL RECONSTRUCTION FROM POINT CLOUDS AND GROUND PLANS George Voelma ad Sader Dijkma Departmet of Geodey Delft Uiverity of Techology The Netherlad g.voelma@geo.tudelft.l KEY WORDS: Buildig
More informationAutomatic Tuning for FOREX Trading System Using Fuzzy Time Series
utomatic Tuig for FOREX Tradig System Usig Fuzzy Time Series Kraimo Maeesilp ad Pitihate Soorasa bstract Efficiecy of the automatic currecy tradig system is time depedet due to usig fixed parameters which
More informationDomain 1 Components of the Cisco Unified Communications Architecture
Maual CCNA Domai 1 Compoets of the Cisco Uified Commuicatios Architecture Uified Commuicatios (UC) Eviromet Cisco has itroduced what they call the Uified Commuicatios Eviromet which is used to separate
More informationA technical guide to 2014 key stage 2 to key stage 4 value added measures
A technical guide to 2014 key tage 2 to key tage 4 value added meaure CONTENTS Introduction: PAGE NO. What i value added? 2 Change to value added methodology in 2014 4 Interpretation: Interpreting chool
More informationAn Inventory Decision Support System to the Glass Manufacturing Industry
A Ivetory Deciio Suort Sytem to the Gla Maufacturig Idutry Nuo M. R. Órfão (Mechaical Egieerig Deartmet, ESTG-IPLEI) morfao@etg.ilei.t Carlo F. G. Bio (Ititute for Sytem ad Robotic, IST-UTL) cfb@ir.it.utl.t
More informationADAPTIVE NETWORKS SAFETY CONTROL ON FUZZY LOGIC
8 th Iteratioal Coferece o DEVELOPMENT AND APPLICATION SYSTEMS S u c e a v a, R o m a i a, M a y 25 27, 2 6 ADAPTIVE NETWORKS SAFETY CONTROL ON FUZZY LOGIC Vadim MUKHIN 1, Elea PAVLENKO 2 Natioal Techical
More informationEngineering Data Management
BaaERP 5.0c Maufacturig Egieerig Data Maagemet Module Procedure UP128A US Documetiformatio Documet Documet code : UP128A US Documet group : User Documetatio Documet title : Egieerig Data Maagemet Applicatio/Package
More informationElectrostatic solutions for better efficiency
Electrostatic solutios for better efficiecy idustry for egieers, professioals ad techicias i developmet, productio ad istallatio. www.kerste.de/e Electrostatic solutios kerste has bee the leadig supplier
More informationSimple Annuities Present Value.
Simple Auities Preset Value. OBJECTIVES (i) To uderstad the uderlyig priciple of a preset value auity. (ii) To use a CASIO CFX-9850GB PLUS to efficietly compute values associated with preset value auities.
More informationDigital Enterprise Unit. White Paper. Web Analytics Measurement for Responsive Websites
Digital Eterprise Uit White Paper Web Aalytics Measuremet for Resposive Websites About the Authors Vishal Machewad Vishal Machewad has over 13 years of experiece i sales ad marketig, havig worked as a
More informationMore examples for Hypothesis Testing
More example for Hypothei Tetig Part I: Compoet 1. Null ad alterative hypothee a. The ull hypothee (H 0 ) i a tatemet that the value of a populatio parameter (mea) i equal to ome claimed value. Ex H 0:
More informationA Novel Virtual Machine Placement in Cloud Computing
Autralia Joural of Baic ad Applied Sciece, 5(10): 1549-1555, 2011 ISSN 1991-8178 A Novel Virtual achie Placeet i Cloud Coputig 1 Ela ohaadi, 2 ohaadbager Karii, 3 Saeed aouli Heikalabad 1,2 Techical ad
More informationSoftware Engineering Guest Lecture, University of Toronto
Summary Beyod Software Egieerig Guest Lecture, Uiversity of Toroto Software egieerig is a ew ad fast growig field, which has grappled with its idetity: from usig the word egieerig to defiitio of the term,
More informationTI-83, TI-83 Plus or TI-84 for Non-Business Statistics
TI-83, TI-83 Plu or TI-84 for No-Buie Statitic Chapter 3 Eterig Data Pre [STAT] the firt optio i already highlighted (:Edit) o you ca either pre [ENTER] or. Make ure the curor i i the lit, ot o the lit
More informationhp calculators HP 12C Statistics - average and standard deviation Average and standard deviation concepts HP12C average and standard deviation
HP 1C Statistics - average ad stadard deviatio Average ad stadard deviatio cocepts HP1C average ad stadard deviatio Practice calculatig averages ad stadard deviatios with oe or two variables HP 1C Statistics
More informationChapter 6: Variance, the law of large numbers and the Monte-Carlo method
Chapter 6: Variace, the law of large umbers ad the Mote-Carlo method Expected value, variace, ad Chebyshev iequality. If X is a radom variable recall that the expected value of X, E[X] is the average value
More informationTap Into Smartphone Demand: Mobile-izing Enterprise Websites by Using Flexible, Open Source Platforms
Tap Into Smartphone Demand: Mobile-izing Enterprie Webite by Uing Flexible, Open Source Platform acquia.com 888.922.7842 1.781.238.8600 25 Corporate Drive, Burlington, MA 01803 Tap Into Smartphone Demand:
More informationResearch Article Sign Data Derivative Recovery
Iteratioal Scholarly Research Network ISRN Applied Mathematics Volume 0, Article ID 63070, 7 pages doi:0.540/0/63070 Research Article Sig Data Derivative Recovery L. M. Housto, G. A. Glass, ad A. D. Dymikov
More informationTruStore: The storage. system that grows with you. Machine Tools / Power Tools Laser Technology / Electronics Medical Technology
TruStore: The storage system that grows with you Machie Tools / Power Tools Laser Techology / Electroics Medical Techology Everythig from a sigle source. Cotets Everythig from a sigle source. 2 TruStore
More informationA Flexible Elastic Control Plane for Private Clouds
A Flexible Elastic otrol Plae for Private louds Upedra Sharma IBM Watso usharma@us.ibm.com Prashat Sheoy Dept. of omputer Sciece Amherst MA 01003 sheoy@cs.umass.edu Sambit Sahu IBM Watso sambits@us.ibm.com
More informationSECTION 1.5 : SUMMATION NOTATION + WORK WITH SEQUENCES
SECTION 1.5 : SUMMATION NOTATION + WORK WITH SEQUENCES Read Sectio 1.5 (pages 5 9) Overview I Sectio 1.5 we lear to work with summatio otatio ad formulas. We will also itroduce a brief overview of sequeces,
More informationChair for Network Architectures and Services Institute of Informatics TU München Prof. Carle. Network Security. Chapter 2 Basics
Chair for Network Architectures ad Services Istitute of Iformatics TU Müche Prof. Carle Network Security Chapter 2 Basics 2.4 Radom Number Geeratio for Cryptographic Protocols Motivatio It is crucial to
More informationCREATIVE MARKETING PROJECT 2016
CREATIVE MARKETING PROJECT 2016 The Creative Marketig Project is a chapter project that develops i chapter members a aalytical ad creative approach to the marketig process, actively egages chapter members
More informationOn Formula to Compute Primes. and the n th Prime
Applied Mathematical cieces, Vol., 0, o., 35-35 O Formula to Compute Primes ad the th Prime Issam Kaddoura Lebaese Iteratioal Uiversity Faculty of Arts ad cieces, Lebao issam.kaddoura@liu.edu.lb amih Abdul-Nabi
More informationDomain 1: Identifying Cause of and Resolving Desktop Application Issues Identifying and Resolving New Software Installation Issues
Maual Widows 7 Eterprise Desktop Support Techicia (70-685) 1-800-418-6789 Domai 1: Idetifyig Cause of ad Resolvig Desktop Applicatio Issues Idetifyig ad Resolvig New Software Istallatio Issues This sectio
More informationLicense & SW Asset Management at CES Design Services
Licene & SW Aet Management at CES Deign Service johann.poechl@iemen.com www.ces-deignservice.com 2003 Siemen AG Öterreich Overview 1. Introduction CES Deign Service 2. Objective and Motivation 3. What
More informationDISTRIBUTED DATA PARALLEL TECHNIQUES FOR CONTENT-MATCHING INTRUSION DETECTION SYSTEMS. G. Chapman J. Cleese E. Idle
DISTRIBUTED DATA PARALLEL TECHNIQUES FOR CONTENT-MATCHING INTRUSION DETECTION SYSTEMS G. Chapman J. Cleee E. Idle ABSTRACT Content matching i a neceary component of any ignature-baed network Intruion Detection
More informationAnalyzing Longitudinal Data from Complex Surveys Using SUDAAN
Aalyzig Logitudial Data from Complex Surveys Usig SUDAAN Darryl Creel Statistics ad Epidemiology, RTI Iteratioal, 312 Trotter Farm Drive, Rockville, MD, 20850 Abstract SUDAAN: Software for the Statistical
More informationQueueing systems with scheduled arrivals, i.e., appointment systems, are typical for frontal service systems,
MANAGEMENT SCIENCE Vol. 54, No. 3, March 28, pp. 565 572 in 25-199 ein 1526-551 8 543 565 inform doi 1.1287/mnc.17.82 28 INFORMS Scheduling Arrival to Queue: A Single-Server Model with No-Show INFORMS
More informationPerformance Evaluation of the MSMPS Algorithm under Different Distribution Traffic
Paper Performace Evaluatio of the MSMPS Algorithm uder Differet Distributio Traffic Grzegorz Dailewicz ad Marci Dziuba Faculty of Electroics ad Telecommuicatios, Poza Uiversity of Techology, Poza, Polad
More informationMeasures of Spread and Boxplots Discrete Math, Section 9.4
Measures of Spread ad Boxplots Discrete Math, Sectio 9.4 We start with a example: Example 1: Comparig Mea ad Media Compute the mea ad media of each data set: S 1 = {4, 6, 8, 10, 1, 14, 16} S = {4, 7, 9,
More informationINFORMATION Technology (IT) infrastructure management
IEEE TRANSACTIONS ON CLOUD COMPUTING, VOL. 2, NO. 1, MAY 214 1 Buine-Driven Long-term Capacity Planning for SaaS Application David Candeia, Ricardo Araújo Santo and Raquel Lope Abtract Capacity Planning
More informationC.Yaashuwanth Department of Electrical and Electronics Engineering, Anna University Chennai, Chennai 600 025, India..
(IJCSIS) Iteratioal Joural of Computer Sciece ad Iformatio Security, A New Schedulig Algorithms for Real Time Tasks C.Yaashuwath Departmet of Electrical ad Electroics Egieerig, Aa Uiversity Cheai, Cheai
More informationProperties of MLE: consistency, asymptotic normality. Fisher information.
Lecture 3 Properties of MLE: cosistecy, asymptotic ormality. Fisher iformatio. I this sectio we will try to uderstad why MLEs are good. Let us recall two facts from probability that we be used ofte throughout
More informationApigee Edge: Apigee Cloud vs. Private Cloud. Evaluating deployment models for API management
Apigee Edge: Apigee Cloud v. Private Cloud Evaluating deployment model for API management Table of Content Introduction 1 Time to ucce 2 Total cot of ownerhip 2 Performance 3 Security 4 Data privacy 4
More informationRecovery time guaranteed heuristic routing for improving computation complexity in survivable WDM networks
Computer Commuicatios 30 (2007) 1331 1336 wwwelseviercom/locate/comcom Recovery time guarateed heuristic routig for improvig computatio complexity i survivable WDM etworks Lei Guo * College of Iformatio
More informationThe Forgotten Middle. research readiness results. Executive Summary
The Forgotte Middle Esurig that All Studets Are o Target for College ad Career Readiess before High School Executive Summary Today, college readiess also meas career readiess. While ot every high school
More informationOptimize your Network. In the Courier, Express and Parcel market ADDING CREDIBILITY
Optimize your Network I the Courier, Express ad Parcel market ADDING CREDIBILITY Meetig today s challeges ad tomorrow s demads Aswers to your key etwork challeges ORTEC kows the highly competitive Courier,
More information*The most important feature of MRP as compared with ordinary inventory control analysis is its time phasing feature.
Itegrated Productio ad Ivetory Cotrol System MRP ad MRP II Framework of Maufacturig System Ivetory cotrol, productio schedulig, capacity plaig ad fiacial ad busiess decisios i a productio system are iterrelated.
More informationCCH Accountants Starter Pack
CCH Accoutats Starter Pack We may be a bit smaller, but fudametally we re o differet to ay other accoutig practice. Util ow, smaller firms have faced a stark choice: Buy cheaply, kowig that the practice
More informationI. Chi-squared Distributions
1 M 358K Supplemet to Chapter 23: CHI-SQUARED DISTRIBUTIONS, T-DISTRIBUTIONS, AND DEGREES OF FREEDOM To uderstad t-distributios, we first eed to look at aother family of distributios, the chi-squared distributios.
More informationBaan Service Master Data Management
Baa Service Master Data Maagemet Module Procedure UP069A US Documetiformatio Documet Documet code : UP069A US Documet group : User Documetatio Documet title : Master Data Maagemet Applicatio/Package :
More informationDesktop Management. Desktop Management Tools
Desktop Maagemet 9 Desktop Maagemet Tools Mac OS X icludes three desktop maagemet tools that you might fid helpful to work more efficietly ad productively: u Stacks puts expadable folders i the Dock. Clickig
More informationPerformance of a Browser-Based JavaScript Bandwidth Test
Performance of a Brower-Baed JavaScript Bandwidth Tet David A. Cohen II May 7, 2013 CP SC 491/H495 Abtract An exiting brower-baed bandwidth tet written in JavaScript wa modified for the purpoe of further
More informationLecture 2: Karger s Min Cut Algorithm
priceto uiv. F 3 cos 5: Advaced Algorithm Desig Lecture : Karger s Mi Cut Algorithm Lecturer: Sajeev Arora Scribe:Sajeev Today s topic is simple but gorgeous: Karger s mi cut algorithm ad its extesio.
More informationni.com/sdr Software Defined Radio
i.com/sdr Software Defied Radio Rapid Prototypig With Software Defied Radio The Natioal Istrumets software defied radio (SDR) platform provides a itegrated hardware ad software solutio for rapidly prototypig
More informationGCSE STATISTICS. 4) How to calculate the range: The difference between the biggest number and the smallest number.
GCSE STATISTICS You should kow: 1) How to draw a frequecy diagram: e.g. NUMBER TALLY FREQUENCY 1 3 5 ) How to draw a bar chart, a pictogram, ad a pie chart. 3) How to use averages: a) Mea - add up all
More informationPatentability of Computer Software and Business Methods
WIPO-MOST Itermediate Traiig Course o Practical Itellectual Property Issues i Busiess November 10 to 14, 2003 Patetability of Computer Software ad Busiess Methods Tomoko Miyamoto Patet Law Sectio Patet
More informationRunning Time ( 3.1) Analysis of Algorithms. Experimental Studies ( 3.1.1) Limitations of Experiments. Pseudocode ( 3.1.2) Theoretical Analysis
Ruig Time ( 3.) Aalysis of Algorithms Iput Algorithm Output A algorithm is a step-by-step procedure for solvig a problem i a fiite amout of time. Most algorithms trasform iput objects ito output objects.
More informationA Guide to the Pricing Conventions of SFE Interest Rate Products
A Guide to the Pricig Covetios of SFE Iterest Rate Products SFE 30 Day Iterbak Cash Rate Futures Physical 90 Day Bak Bills SFE 90 Day Bak Bill Futures SFE 90 Day Bak Bill Futures Tick Value Calculatios
More information1 Correlation and Regression Analysis
1 Correlatio ad Regressio Aalysis I this sectio we will be ivestigatig the relatioship betwee two cotiuous variable, such as height ad weight, the cocetratio of a ijected drug ad heart rate, or the cosumptio
More informationConfidence Intervals for One Mean
Chapter 420 Cofidece Itervals for Oe Mea Itroductio This routie calculates the sample size ecessary to achieve a specified distace from the mea to the cofidece limit(s) at a stated cofidece level for a
More informationwhere: T = number of years of cash flow in investment's life n = the year in which the cash flow X n i = IRR = the internal rate of return
EVALUATING ALTERNATIVE CAPITAL INVESTMENT PROGRAMS By Ke D. Duft, Extesio Ecoomist I the March 98 issue of this publicatio we reviewed the procedure by which a capital ivestmet project was assessed. The
More informationRO-BURST: A Robust Virtualization Cost Model for Workload Consolidation over Clouds
!111! 111!ttthhh IIIEEEEEEEEE///AAACCCMMM IIInnnttteeerrrnnnaaatttiiiooonnnaaalll SSSyyymmmpppoooiiiuuummm ooonnn CCCllluuuttteeerrr,,, CCClllooouuuddd aaannnddd GGGrrriiiddd CCCooommmpppuuutttiiinnnggg
More informationRainbow options. A rainbow is an option on a basket that pays in its most common form, a nonequally
Raibow optios INRODUCION A raibow is a optio o a basket that pays i its most commo form, a oequally weighted average of the assets of the basket accordig to their performace. he umber of assets is called
More informationProject Management Basics
Project Management Baic A Guide to undertanding the baic component of effective project management and the key to ucce 1 Content 1.0 Who hould read thi Guide... 3 1.1 Overview... 3 1.2 Project Management
More informationThe Wider Role and Benefits of Investors in People
RESEARCH The Wider Role ad Beefit of Ivetor i People Mark Cox ad Rod Spire PACEC Reearch Report RR360 Reearch Report No 360 The Wider Role ad Beefit of Ivetor i People Mark Cox ad Rod Spire PACEC The view
More informationCS100: Introduction to Computer Science
Review: History of Computers CS100: Itroductio to Computer Sciece Maiframes Miicomputers Lecture 2: Data Storage -- Bits, their storage ad mai memory Persoal Computers & Workstatios Review: The Role of
More informationBusiness Rules-Driven SOA. A Framework for Multi-Tenant Cloud Computing
Lect. Phd. Liviu Gabriel CRETU / SPRERS evet Traiig o software services, Timisoara, Romaia, 6-10 dec 2010 www.feaa.uaic.ro Busiess Rules-Drive SOA. A Framework for Multi-Teat Cloud Computig Lect. Ph.D.
More informationQueuing Systems: Lecture 1. Amedeo R. Odoni October 10, 2001
Queuig Systems: Lecture Amedeo R. Odoi October, 2 Topics i Queuig Theory 9. Itroductio to Queues; Little s Law; M/M/. Markovia Birth-ad-Death Queues. The M/G/ Queue ad Extesios 2. riority Queues; State
More informationE-Plex Enterprise Access Control System
Eterprise Access Cotrol System Egieered for Flexibility Modular Solutio The Eterprise Access Cotrol System is a modular solutio for maagig access poits. Employig a variety of hardware optios, system maagemet
More informationCASE STUDY BRIDGE. www.future-processing.com
CASE STUDY BRIDGE TABLE OF CONTENTS #1 ABOUT THE CLIENT 3 #2 ABOUT THE PROJECT 4 #3 OUR ROLE 5 #4 RESULT OF OUR COLLABORATION 6-7 #5 THE BUSINESS PROBLEM THAT WE SOLVED 8 #6 CHALLENGES 9 #7 VISUAL IDENTIFICATION
More informationINVESTMENT PERFORMANCE COUNCIL (IPC) Guidance Statement on Calculation Methodology
Adoptio Date: 4 March 2004 Effective Date: 1 Jue 2004 Retroactive Applicatio: No Public Commet Period: Aug Nov 2002 INVESTMENT PERFORMANCE COUNCIL (IPC) Preface Guidace Statemet o Calculatio Methodology
More informationAgency Relationship Optimizer
Decideware Developmet Agecy Relatioship Optimizer The Leadig Software Solutio for Cliet-Agecy Relatioship Maagemet supplier performace experts scorecards.deploymet.service decide ware Sa Fracisco Sydey
More information3G Security VoIP Wi-Fi IP Telephony Routing/Switching Unified Communications. NetVanta. Business Networking Solutions
3G Security VoIP Wi-Fi IP Telephoy Routig/Switchig Uified Commuicatios NetVata Busiess Networkig Solutios Opportuity to lower Total Cost of Owership ad improve Retur o Ivestmet The ADTRAN Advatage ADTRAN
More informationFrance caters to innovative companies and offers the best research tax credit in Europe
1/5 The Frech Govermet has three objectives : > improve Frace s fiscal competitiveess > cosolidate R&D activities > make Frace a attractive coutry for iovatio Tax icetives have become a key elemet of public
More informationSCM- integration: organiational, managerial and technological iue M. Caridi 1 and A. Sianei 2 Dipartimento di Economia e Produzione, Politecnico di Milano, Italy E-mail: maria.caridi@polimi.it Itituto
More informationEnhancing Oracle Business Intelligence with cubus EV How users of Oracle BI on Essbase cubes can benefit from cubus outperform EV Analytics (cubus EV)
Ehacig Oracle Busiess Itelligece with cubus EV How users of Oracle BI o Essbase cubes ca beefit from cubus outperform EV Aalytics (cubus EV) CONTENT 01 cubus EV as a ehacemet to Oracle BI o Essbase 02
More informationMaximizing Acceptance Probability for Active Friending in Online Social Networks
Maximizing for Active Friending in Online Social Network De-Nian Yang, Hui-Ju Hung, Wang-Chien Lee, Wei Chen Academia Sinica, Taipei, Taiwan The Pennylvania State Univerity, State College, Pennylvania,
More informationIn nite Sequences. Dr. Philippe B. Laval Kennesaw State University. October 9, 2008
I ite Sequeces Dr. Philippe B. Laval Keesaw State Uiversity October 9, 2008 Abstract This had out is a itroductio to i ite sequeces. mai de itios ad presets some elemetary results. It gives the I ite Sequeces
More informationIncremental calculation of weighted mean and variance
Icremetal calculatio of weighted mea ad variace Toy Fich faf@cam.ac.uk dot@dotat.at Uiversity of Cambridge Computig Service February 009 Abstract I these otes I eplai how to derive formulae for umerically
More informationPROCEEDINGS OF THE YEREVAN STATE UNIVERSITY AN ALTERNATIVE MODEL FOR BONUS-MALUS SYSTEM
PROCEEDINGS OF THE YEREVAN STATE UNIVERSITY Physical ad Mathematical Scieces 2015, 1, p. 15 19 M a t h e m a t i c s AN ALTERNATIVE MODEL FOR BONUS-MALUS SYSTEM A. G. GULYAN Chair of Actuarial Mathematics
More informationGroup Mutual Exclusion Based on Priorities
Group Mutual Excluion Baed on Prioritie Karina M. Cenci Laboratorio de Invetigación en Sitema Ditribuido Univeridad Nacional del Sur Bahía Blanca, Argentina kmc@c.un.edu.ar and Jorge R. Ardenghi Laboratorio
More informationAgenda. Outsourcing and Globalization in Software Development. Outsourcing. Outsourcing here to stay. Outsourcing Alternatives
Outsourcig ad Globalizatio i Software Developmet Jacques Crocker UW CSE Alumi 2003 jc@cs.washigto.edu Ageda Itroductio The Outsourcig Pheomeo Leadig Offshore Projects Maagig Customers Offshore Developmet
More informationAdaLab. Adaptive Automated Scientific Laboratory (AdaLab) Adaptive Machines in Complex Environments. n Start Date: 1.4.15
AdaLab AdaLab Adaptive Automated Scietific Laboratory (AdaLab) Adaptive Machies i Complex Eviromets Start Date: 1.4.15 Scietific Backgroud The Cocept of a Robot Scietist Computer systems capable of origiatig
More informationOn the Capacity of Hybrid Wireless Networks
O the Capacity of Hybrid ireless Networks Beyua Liu,ZheLiu +,DoTowsley Departmet of Computer Sciece Uiversity of Massachusetts Amherst, MA 0002 + IBM T.J. atso Research Ceter P.O. Box 704 Yorktow Heights,
More informationVladimir N. Burkov, Dmitri A. Novikov MODELS AND METHODS OF MULTIPROJECTS MANAGEMENT
Keywords: project maagemet, resource allocatio, etwork plaig Vladimir N Burkov, Dmitri A Novikov MODELS AND METHODS OF MULTIPROJECTS MANAGEMENT The paper deals with the problems of resource allocatio betwee
More informationDISTRIBUTED DATA PARALLEL TECHNIQUES FOR CONTENT-MATCHING INTRUSION DETECTION SYSTEMS
DISTRIBUTED DATA PARALLEL TECHNIQUES FOR CONTENT-MATCHING INTRUSION DETECTION SYSTEMS Chritopher V. Kopek Department of Computer Science Wake Foret Univerity Winton-Salem, NC, 2709 Email: kopekcv@gmail.com
More informationData Center Ethernet Facilitation of Enterprise Clustering. David Flynn, Linux Networx Orlando, Florida March 16, 2004
Data Ceter Etheret Facilitatio of Eterprise Clusterig David Fly, Liux Networx Orlado, Florida March 16, 2004 1 2 Liux Networx builds COTS based clusters 3 Clusters Offer Improved Performace Scalability
More informationIdeate, Inc. Training Solutions to Give you the Leading Edge
Ideate, Ic. Traiig News 2014v1 Ideate, Ic. Traiig Solutios to Give you the Leadig Edge New Packages For All Your Traiig Needs! Bill Johso Seior MEP - Applicatio Specialist Revit MEP Fudametals Ad More!
More information