Unit = By Data (Bcso6!) Digit Dato Digited clolg B information Stored on a computer aystem SO Series O's and 3! ° 's Pn q bin language « DigP hal data Jumps fom one vata | fe tre Yneet ae Skp by Step Sequence, Example: Wheneve, we Send an ema! , sead a soefat medPa post, or fake pfctunes uth oust digit val cameta, we ate Workdng- af th oligPrel olala . “Dig tal data con be clanfhed ‘vo thetee forms: Q) Unstructuresl Delay “The data wihfeh clo not vonpom te a data model arte net fh a qorm that ean ke used eoasly by a tompuley prion fs cotege sized an unshuckuteal data: Absut go-do7, ata of an argantralien & fn Huo gormat- Example ; Memes, chak st00mo, Powe Point presentahons , Srnager jvides, Jetlars, seseaehes ,» urhPte paper + of the body a an emof| , ete. b) Semi — Shrvetusied ‘Dakat “The date ushih does net con — ~gorm toa data model bub hap seme Shuctuxe Fs calego~ -Bzed 00 semi — Strectuted data, poworet , tk B mst fo aq Youn drab Can be used eas? by 2 compulel proguam. it example + Emoilo, XML, masikue Langu: Uke HINL, ete Metadata te avallable, bub net suppfefent-¢) Shuttuned Balas The data which %% Pn an organPeec) frm + (ver, 96 sows and twtumns) and Con he cantly wed by a tempuley Peqnom Yo Cakegor? 20 an Grime Shructusied olatas Relahi- renchips exist between enlvKer 7 and ynetn ojeuts . Example: Data Stored fh 4 data, Svdh as clan- datalranen. History a Big Data ee .. Bo chastactesdd zed tne sapPd advance - cvmenk fy the field 4 Information Hebnolo gy - G IT hap pueme an Faleysal pout af dally tie ao well as Vasifou, the, induobies Like health, eoweahien, entectatn - coment, guente and Heehnotogy 1 Fore sr busines operah - en) and there Pnduohton ginwiale a tot | data, thin can he catlecl Bg data. Bry data toms aj lage dataceks that tonnet be managed qPeentey by the tammen clakabaso manageme - ak Systema. There, datarets Hange puem terabytes Ho Qrabyteo. Mob?les, poh eredft castd ; Raol*o Frequen Polen? Pcabion CRETd) devices, and Sodal netursykiag plakwems ovecte aumbuithy — of dota that rou Heside unulf Weel ae aroun Stuvers form many years, find wlth the evolubion a BFgdaka jis dake tan be accorged ond analyzed on a Hegulat boasts to geneale useful Pryormation +CBP, } Big data’ in a selene Ram depending en why 9s old - muuning Ep, ° "4 : For example , 8fg dota to Amazoy or Groogle oy ypocen te quem Big. daka ty medium ~s?edl Prawuonce org an? zahen. Pa Antrodverion 4 BFy Data Plalterm 8 big data plakorm 4s a type 4 DI Setuken thak eembi-~ 7Meo the Jeakuies and copabh Pier Several b% dot appircoaulisns and al Keo ulHSn a single setuhen , His b then weed fusethan de managing ad well ar analy ang Bg Dake. TP focuser providing Pbs woos with ejfscient analy hs boob fer mapiva datarels . The upew |] Seth piayasm Com ouotom bulid appleakens Ga acearding 40 thot woe eape Ike t caleulate — curinmay logatty. (E-w- mm pnmet(e Urex case) , and so on- Otoal: The main goal 9 Big Date Platter 8 to achieve: Scalabfifty ; fvabl abterty , Penformance and SeeusPly « 7 example * Seme x the mos ceroments unvedl ca Data Platjorme axes « Hadoop Delta lake MPgrab'en Platporm » Pata Catalog Platform + Data Tngesten Plaijarm ¢ ToT Anata Heo PlotpermDive fe By bate 8F9 Baty har quitkly HPsen to become one o tre mosh dlesined bopfa th the Pndushy. TRe main brine dbivets yor Such wPsinq dermanal fer BF Bata Analaitien cite | Ie Tre. ob Fttzation 4 Soefetes 2. The. atop Yn Fechnolog costs S Connectivity though “cloud computing 4 Ineveareal Knowledge about claka gence S+ Socal med appttahions. & The sso of Prtesnet— a - Hrings ( ToT) Example: 4 number tom panies thab have 8% Date at He wre oy thety Stra keg tikes Apple, Amazen | Facebook and Nekp ry have bere. vera Succery fu} ok He beptaning oy we ait centory : Arch? fectuste 4 ‘Bfy Decte [ Oachesbrakon }s The b?q data cubflerruste incluole the fotleutng Compon - tue, Data sowed » Ay big dala soluhons Stat uth one be moye dala Sowice, Lrample t- + Applicaton data Stove , such a solahbnal olata- ~baves . © Stabe ple Produced by appleations , such as Web Server log fies: ° Real -rlime dala Sowies , Such ao Iof devices. Bro Bate Ingeshen La Lower This lager of Big Data quiftecwie B the frst Skep qe the dala eng, cor vaxPable sewites #0 Shu fs Journey 1 geotion meano the claba bb prov ted ay chee making late fume ema thy Ww fecthey Clea in the bala Ingeohion pre fo Data collechy Lavery Pn thio loge more foun Fs en the framportation Data yuem the Engonen layer to tre Het a the data pipeline. Tt 4ne toe daha OHtelP pure where — tompenelo axe detsupled So Haat Onalyte Capabl Ufeo moss begin.Dale Proce? ng Loe Th thin pHimawy layer ef Big Bala auehilectone , the fous fs to specfalfoe fn the data ffpeline Preeving system - We can soy the dale we hava wlleted &h the previous legen b proceped th +nty l ' Here we olo some m eae wlth Hre date gp feute them to a difperent deonatien and choo? tq the lata fer, and itr -the (Pst pet wheve the analyte may occu. Bate (a Bince the data seb ane So lange 1 thedere aq bFg late, Seluvien musk Pree data le uss > Hunning Bete, Sebs & flier, agyuegale and prepasie the daky qr analysic. Real- time Meese _Ingeohion 1 a Setubien tndludey seal- time sewteos, the asichPfeebaye musk include a way I> capture and shrre Heal- time menegio Jor stream pecering . Shream feeceeing After Coed Heal time Mepager, the seluben Mush prowe them by dP lief aggeregationy 7 and PrpePrg the clota fe analysis « “the. percewed Styeam data Py then usPHen fo an output sinks We can wwe pen ~souree Apache Streanthng technologies ike Sfopme Spark Shreaming _Dale Shorage. bs ey Stovage bersmes @ challenge urhen tre sae gy tne data | you one deating with — busmen lavge- Sevexal possPhle solut— -fom, like Data ingestion Paleuny , tan Heocue fuen Sveh pwrblems Finding a shvage Solution 4 meeh Tenportant ushen He size qf your dala beswmes lange. Thi loyer 4 BFg Deu PreinitebuHe fous on "wher a qo sire Such Lasige data cyprPenty Date Query toyed Th & Fhe auch? beckueal layed wee aebPre an Prrceming ¥ Big Bala take place. Hese, the pm qouw is bh gather the data valua 4 be more helpful do We nest leper Aino Heed Data She. Many bra dala Selubern prepare olaka yur Analysts anol then seive the prcened date th a stevckwied format Hho can be qyrerted using analytfeal Wols. Example + f2wie Synapse Analy Heo provides a managecl gervice fe. lasige — scale , cleud —based data wane housing «S V's yf Big Date * Velume * Veroe? 7 Velo hy > Value 5 Mier Veloei High Speed 4 Atiomulalion Of etka VasPety Dijporent ermal ef data ysoro vasuleus Sewites yy Volume eFg Data gousier dodly + Sokal med?a plalferm Ha vook Mt yplumes” % dota geneuateal meas th an buna pre, machines, pebutorkn , human jniswh ons and so on" $ ae can be Shrueckured _unshodtuteal anol semi-stroch— wk, nok ate befng couected quem Agere Source,3) Velodtg Veloctiy Hepour to tre Speed with ukfch lata ks qrowared 9 steal -time + Velocity Plays an Pmportant stele compattecl 0 athe. TH conteuiva Hetcing a frveming daka Sebo 1 speedla 7 Hohe. % change and achity bush . a) Venodity Venadig Meqena tb the wate ay the dlalg that Fs bein zed, d Tt Bb tne prep x being able 42 handle and manewpe. data ogficenty - s+) Vale FA Value tb an ewenKal chatatien’sHer 4 big data. TM B net the data frat we pwten 6 sore , fF valua- -ble and sietlable data that we store, proce and cece eee Big Date _Teshrel gi re 4 © Storage i One 7 FnatyHfeo 1) Spark a) Mong odb a) Kogka 2) Carrandstoy 3) Glockehain i. mPrfrng t-1) Roptdminey + VisvelP2ation y 1) Plot a) Presto 8) PableauAe Be Lomponerte Data Sowue . ‘ b caPh%el vompaenenls Data Sountre te OD 4 the mos 4 bi, daka becaue. they Seve ao te feunclation er denhing vatuoble Pnotg ho and levenagin the advan- stages of bsg dala. wthouk q dPreud and hPgh- qu- matey Bek of dala Sewiresy organPaations cannek pully tap ‘bp the porentrPal 4 big data fo make Png - sormed deePgienm. * Dake Shroge Busfnew need to she the data semeuhoe before befng pwcewecl and the Ideal locction dot UW 18 @ typhatiy a data take uthfoh bh a bg Scalable tng hru- rchuted databace Capable hotdiing a hwge fAumbey 4 difporenty gamatted fileo+ © Batch preening Batch pacomy 7s waking far a pauuslay wantity ag of Hou data 4 be obtained before Pergerming an ETL fob b Ble, aggre gale , and prEpAte mamFve yoturnes of dala yor analyte. Wt aled uthan dahe givhnen % nok a prblem . Hacloop Open - sowie fier nework aie a, C6mmory atteinatve gor Such large data pascee ta 06 Shean Procening. Thin specie tmpenent hb steopomible for the cenh'nuo- un fw a daha uthfoh Ps Necemasyy ¢* seal fime data cenaty tea Te wovedlly dow +thh by tocahing ang polling data as seen ao ik & genenaked and puch tt bb pthey tog date Feahnology tom penenio g& Heal time nae Gtike — bateh precewing , uth®h operat on StalPa ach ef data ab Ssehecluled Entewvealy 7 Skream proce ~ ~Aoing , Enabler ovgan?2 alibn s to handle and deufye ‘bn ght fiero data In meten. e Machine teaser Machina leosu! “WH coed t > an eperhiaf empenenk and leehn}- help ekprack trol hty and Pelentiey Prlley - 7M) fem 9 log @ 7 Wmpley latasety. The mor dale get have shed - the more the atgorthms — became atcastale and helpfuf Over, time + Theso agortnns Sequine a huge number |] data te be frained an. o Prorat es and Reporting fey and eporbirg asie vVikel eempoenanta a bo data becuse they dranoferm saw data into a jeonable might, ' @nab Ungs ongeni zations 4. make : a data, businewes can Fdent}y prendo ¢ 8p himiza proce. -9e2 rand enhance custom ef expedences. data -dviven decPsiens. 8y analy zing and yirsval?s;;bey Dake Qmportance. © BuoPne : Pn Ap Gfireioney - Bg dala helps buoipeper Intequan re olakey § yuorn vostiour sour eo be pphimize precener and work} fou: + DerPsion making ce Big dala helps bua nener oa" data t make dala- ‘unformed detisions + tompeh K age + , D a e ve advant ‘poe Big clata helps buninenes colle ack and analyze Heal - time dala 40 adopl fete « Pq Data — Prpplitation urosid iq data have Senoral application “Look gyme of “bre ote Lbted bales . Tracking Cuchmeny — spending hab? - 1 Shopping behav - 7 OU © Rewyrmendabieo e Smaset trey System + Education Ropall 4 wkoleoate frece + tommunteation , medfa { enfettainment « Health taste provides Qyrouttance + Pranoper tectionty. / Bea Dota | Bj ee tee ° SerssPhy eee A sot o practices and Keubnotogion bb probe daly yor unauthorized acco, Hey he and other treaka . Wor authenheanian , puunfssions, and user Hole f enowre on auinovtzed ptople Con accom data. Dak mate’ nay and anon mization Prmtecr sernfive dlaba by Jreplaeing Er waft Scran - ~bled os pPctihous Lnprmation - ° tem pul ance Data usmpfiiance basen A gemma Shudure thar enowie 7 a gonP2aken getsur olote — Helaled Stanadlaridl - lau , seguta Leer , and Dora pro techen Cos las Like the “(stenenol Daha Protechen Moder Roguternon (OrDPRY. e fed? ne Seely curesis Requtan audits that Vdent?ty VulnerabiiPhios ond poten- -Bol thsreach do daba Secu « fel Seely cual? Audits Heo Pdenhdy VulnecabPPHes Spee pt & APIS: ‘Bry Date Firat too Big date analytes fo a compley proces of examiury ih dale fe uma thjamatien 7 such an hidden pot + ev) wrrelations , marke tHends and curslemey proper ~ -enter. ThPs can help ongant 2ectiens make Preferred buss - -neMr leePgions - Doky analytrer techno Lo gros and Hehniques ge agort- sTolisn oO wey “to onaly7e date geto and gothey new Anpomalion: Bfg Date Analy feo Crables —enkeepafses do analyse tier clala jn gull Contech quteky and seme agit QfeH steal time Oral: Beneyt>_Y Big Dela Anatytito Tn wrporetung wy clata analy Heo ‘into a buofneto or Orga ~ anfscclion har several advantages: © (ost gecluclion $ 8tq clala can stecduce tas, tn ShoFrg ol) busine date jin Bne place. Tracking ana- = tytieo abo helps tompanies find ways “Le work mor qrucnty fo tut cools Ucheste~ “VO! possible»+ Product evelopment ; Developing and masekehi new Prodwets , Sesivicer-, ay byande fs moch easter when baad en date elected Juorn ulema needs and wants. . Strategic bupinewo deesiens + The bitty to conohauttit analyse date helps buvineod make buen and fevtor aleerst- 61: © Rf&k Management + Bufnewres can Identity Ssks by an- ~ aysing dlata Patlexnn and dleweloping Selulionn Yo managing phose Hisky- Types J BG Date Anaty ties (1) Deseaiphve Analy Heo Clohot to happening ) DeowsfpHive analyer sefeue to daa -that can ke ean %y stead and “wkuprked + TAS data helps creak Heposk £ visvalfso. information that can detail company — preys ts and Soles © Dfagnos tes finales (lohat ofa’ “tr happen) Dfegnostto analyte helps wmpanfes understand uuhy 9 mm occewted: By data teenhnotogies and Role altow da ups te mine and steover data that helps ofsseet an gesus and prevent iF geo happening in the fertewie - PreclPatfve AnatytPeo (ushat might happen) Hye onalyties Looks at past and present data to - pred ae 5 predichions :pe i+ i) Presemelive Analytfoo (tohar acher should be Faken) Preanphivo analatfes Solvay a problem , pelying en an machine, leastning to gether and uno dala fo sus management. Challenges of tenvenenal Systeme tonventional systems have challenges wwhon ucorking uuth big data because they can't scale ,pyoren data Slously and lakk advanced toolo- © Sealab? lenvetionad seater can't handle Laxae ameunb data ater - ° Speed tenvenhional systems ane Slow ther whan preening a analyzing laxae ameundy 4 debe. » Data qpalttes lostgen ‘dabasen ase mor dikely to wonkain gn acceenPer , Fmompele seeords , wus and dey clupticates- ° SecusPhy tonvenrhonal System moy nob be able “le proteck alakg ageinat- unathortred acers, breaches and eyber~ alladks Modexn Data Array te toely. «Tobleav Public . excel * Apache Spork © Raphd Mines © KNfime.