Uoit3

No S&L Doabae No SÙL daalbases re designed fo hand,

unshcttet and Sehmi sthidreddaa d tayn hahdl
mnage

charaderistics of No SoL Dabaoae:

Cah or hanqes in he data stist e and ie

Feble Data Hodelu: Fleteatâoradea No Laase
Soerots i s data modei eshich'alloc a dsta meodel thot foeft
its theis eayisenmenty
optioizedfsr Some fa
Dse cases that ead tofastex gueies
ec: No sL doabales
ptanizedfor
wse es oiteg high persoman ad los laleray
Cases

aceeas tho daas

Horeonta Sqalasidyi NosgL sotabaes are deuzne to salk
heinortly by adding more alwtea to the Sesnier: aahich
Gan hande large data
) ost-efechiveness No SL da aba are mast cast efeve
sgL databaes espealay r lage dasa.
 Qescdoçes tnendly Nog dbabases are edguelop'ey friendli
lexie" dba 'midela:
Viawing Simple API and
are, cormmony
denormalizcd, doenent osente
used tonepresen data in a

dahonal'dsbaloaseiohee daba is tpiolly nonalie

UNje Databsei ten
into ultiple idsal
ta in a singledociwhenb
pea of aggregate modeli
ase~
Some ommon
typeot
Ooouwmet-Based Nodel:This. is the, most coon
d a rnedel'sed in, No SQL
aggvegate Stoxed as docmends,ty pioaly in sON or BSoN na.
xData is elawet data fra poot'eher
posticula
al the
* Each doounent contins
entty ora reord
this, odel dote e stored as key alue
2 key Value modeli dn
*Eaoh ey is uniue ond mapc to a
a omespondin Vadues
Bimple vedievad and
kLey- alue Stores are highy efficiendfoy selatonshizs between
|Storigebt t Soppost complexqenies or
data enties
n this madel, Ured oy dtebasei ke ofachie
3) ounn-Fomily Model: coloumoamilies.
organiea dota int oims groupedioho
*Each tolumnamiy tontain multi ple hoas and eachmoes. onains
coluurmns th coespordng Vales
 oe optionj zed o ead and

aoyoss l enodes.
datalbasei reppsert daa as nodes,al,.
WGragh' Nodel: 6vçh
and proge rapresent elationhip bol
entites eoges
*Nodeseprelert
prspertes'vaprerent athibutasef anttesdrelaansli
enditea and
Graph dbabae are
nselationshigs betogenj data enh biea'
Tequire toweNing complex
Nae pairs: Aagregqte ey ae paia are essent
Asgiegate eg
tletns
imay focus
*ile ey Nalue paie strea hertsees: eomesponding Uabue
posst aysond theix
Stosng indiviuas
Impernetahon key adepais in NoSQi databades:
Using Cornpound,beu NosL dataased allow you to ompose
heys These keys coan include mutiple tomponent
tompourd ndra siogle
eaing gou to gp elated data togeter
paaert bey

as TtoeK, and Shore it ad Vaue cores poniana to a key ina

key Uale stoye
aUoeuOu fo eoe m
his alouss Cpmplex data t t e s as the
you
Valiue'associated uoita a Single key.
 ACaiog daa profires ey Sufre Nosg dedaaa oten
setsose keys
yell to retsese
bied com mon prefea or eufives .
o h consitet nal ne cowenhon,
aa consistent
keys
i 2ing gousselated dlota efficete
*By
go4 tonetieve d

gt or ists iwhioh allas you to agge gate

Qucles e
related begs togeltter
dndeY: Nos dtoases often econdan
)Uskng Sexonda index data ba ed on
7

indlexe, ohih alos qou to

to tndex and queiy
atibesothey than the primay key
R. oreatog Seuonday inexes on Specfe attsbetesos Can
etyetieve elated alata ithot needing to knos the
east hey'
madeli Documet datamodel h NosL databa
Dooumert Dosta
-ojented omat, ohese eac, docue
regresent data fo a do eument Caicha
ati badea and posì bly nested sutued
Cortains dada
doeumeta
Ouenies o Document ata Hode:
DIsoN|BsoN Foaroli Doaumeit are Shosed in Tolitobiect

*Toueion deurnenti toriet of eg-Value prux cheke key

ase Sthrge andvae Can be Voñaedatypes.eg
 erd,al shudres bcuenla in NoSgL datobae can bo

thenspvepseledaons oteom odex and rest.d

*This eabdes
didadakionshis ithin asingedacment
taditonal and elahiona da
hera eii hy: Unlike defin,doeimen
dta henos are giel and pre-
wohere
oerted ddabale offex herna Heyibiyhhh
*0ousmest itin a tolecon hoNe diffeent featres,
and sthitves that tan e added o temosed dynamial
withod efeig other docurne in the eleoho
Colection Cbenetta are yplcalyoganizadita colleshbeg
twhich ae onalaqo ttaleu in elationaldatlojel.
groptagees douent tht hae imilar fadsu
) DogisnetID: fach documet ieagped a nlae islentfa
thd distrqieres foom ther doumeni
yetieva s doeurenta
Indekea nle&eLdatabale alla forhe Ureaton inene

9guengt Nasguddabotes pruide qeylang uaga forqegg

doeurentsbased oneir athibtes and aue
 Databases: Groofn aatabales is a tipe af nbSgL
Gaph
emphasize tie
emphasize tie vepruentaton
cbdalase ht cbod
of nodes
9raphs consisting getyng tolathiship
These database exel at madelingad
data entitie making thenm Litable Seoife
for
Laueen
Rcenanos

Chabradoristica of raph Dadabase:

bales oranizes data as gaph,
Cota odel: Grafh daa
dge and propeties Nodes reelet
ohidh coníst of nodes nedes
edgs represet relationship batueen
end tes
Native mph Quon anguagei Grrph daabase tyriaypide andquening
qyuen
rothue geg language optinmised foY traresing
gaph Bthcure
optiniaton: aph datooadel emply Voias
3) Inadesing and gueny trawete
opimiztion hthniqua to eifcenty
irdesing ond quey
and queg lage graphe Aco hanodi
Many dtab asea provide Suppot
) AcaD Tropaachions: prexence of
data istegihy and conciefeny e¯an in the
to encuse
Concurent upd es and queie madeling
mdeling ond qeg8
)Use Cosel: Gruph databes excel tit
ralationships,in vaioo dornainc. to Soale
dataoaes oge delignecl
Scala>ildh arol Pefrmance: Graph Hhem
hoizortally acoudistäb§cd dustese of ooustng
maahioes,
data and complex quey coot loads.
handle lage veuma o
 s)
araysia to 3> 2) Oses *These Bdhemarleas
Signifeant
ing.and aradComlex Velu )
mes
) saalmalingthen inget
St oyed
dathtthe
Seni-sthudued
Fexibi te ntraida
dhnolagie
Real-Tirme
faciitata Captxe Sclaloility:
ta Agile of
tine and lity
Shemaes
Dada Deselopnent:
apabitities datateinqestion o oithot in dalabasel,
agle and Doda ole
Dadatngesn: data woell Shema prekt Prouide Doabasei:
Anaies: pr oce inmana
deseoprent commonylted predefned hodels: oten
The less aa' Databaes: flexi
ehema Shema-less the of Sehema rg
gtghema
ng amig Mony endtabaiey s kh
asociated
texrt les prahces handling
couleedin ng
Vonisehema, in data and
data Schenna eac
anaysisdotabas
alooing St ae
ruci u re hand Proceisin g es
natere
and fro nm the ehabtidot
n g obade Cing th
les cormnot
Bi desiahed datablesplay
NolgL
paaéing
qgraph e Seurcei
iteaie oAatabase, Deda andfemat unstucd Wast
enaloles ojanizatonsto
to l o u
analysis these
gaizahons rmass taase yolumes
haslae le do
actane wefdatabases Suppost ve and a
be af
 Map Rede is ta Progaoning nodel and
MagRedue: Soogle t appst
asoiated processirg framedk develg ped g
dtaet in ja dst, but eloompo,
the proces o lotge Scle
enwirones
af Map Redute:
wotiog
inpt data is divitedi
)Map Phase: dn map phauedetned map funch'on ic ofplied
Sonalex ehunts and a wset
Smalles
to eaah chen independenty dta ond gehekole a
functionpceses the inpt
*The map
gt of intenediate key value pai. complbe the islemnedie
)Shufte and Sost: once mapphaje biaed on theiy keyn
pas are shuffed and osted
key Walae alues asioiated otthe Rame
*This proceSensresthat al preping wedete phose
key are groped togtr Value pois
phase, the Sosted bey
Redae Phase: In Reducedefinedredued funcon The realute
procesied biy a wes axsociated lit o
ase
uniue key alerg oith it
function hes tach Compuath pesideat
Values and pexfonm tapiealhy Stoseolin a
of rredute fonchion is Hapßeduce ob.
he oytpt as the fral oattt
eagate or etne d designed to
Bhorag Redee rame o are
) faultToleyante: Mopénsaxefltolerane in distibtna
hardieailuse and
enuonme
s) 'rogrornming Madel : Mapledaue ame soodke prouide dd.
mdelthat enable desedope to expres
programig
prouesing tass uing funitix progyanming langgs
ange of
Cases: Mapeduce /s ocl Beited for coile
6 Use data deansing, teut
Big DadaAnalya indading
wse Cose

anslysis nrachine leanigand graph processirg

Parttioning and Combining
vejen tothe proes ofdiuidina
Patfianing* Paritioning manageableBosts called
largedalet iato Smale, mexe
pastiions
* Rotiioring is otben pexrmed based on data loality
pinaples to misiize data mevemert and nookoehead.
Types of Potitioning The tpes of gathorigae
tash Past;toning: Lo, hash patistioring ushexe bash funton is
apolied t the input data to ddenined Lohieh pa heo eoh
ha elemert shobelogt .
Qange Hdhing Ranbtioring Range Pantitioniog iousleas
en-da baed peredefined ada
potitioinga

shexe eah pastition Containg a Set ool colum h do te

enhve dalet
 ) Nesto Pastitionig: 4 oiyides data basedon colemns
each pattion contains all roasfor a, Saoset f
uohexe
toumns
involwes perfor ing Partial og3vgahin
Combining: omlbinirg befoye shuffing the
of data othin each compude node
the dlustes!
data a Chosr
ombining is
k to goup togehter bey
* The pimam qoal ot&ame ey ens§og that al vaues
Vodues pais siththe processed by the Boe Teluca
assoclaled otth a Spehie ave
Benift'
) Reduced Data
Data Pranafo:
Tronso tombining aows for patialaggg
omenaeeae
mafper Thic reduces the amount f
of data locally on eachchufted
chuffled acoss the nehoors:
to be
data that needs to
pexfoinga Poelininay
) Redared Reducexleod: By oal aggregan
onthe reduesr.
combinina ightens he oor
alalations: Composing map reduce lalaulat
Cançosiog Map-veduclg processing tasks ioto
Complex data
invower byeaking douon untt that ean be ekeeutedin pablel
Srnaldes,more manageable
bting Compuhsng custes
acose disti the rature of-thedata
Ientithe Poblem: Undesstandtying to oe.
yoy
problemyou te
and he Specifie
Problem Decomposethe roblerm into Smales
2 Sreakdounthe
independet units,of ook h an be eueaed inpel
 daaa
3) Map phase: Implerment the map phaiejohereindependa
poxttioned into Smales chune and processed
by individal map tas ks phasethe inte
8huAeand Bot Qing Bhase and Sot
Walue pais generaeod by he maps
-medite bey
aye Soted and grouped loy
5)Reduce Phoue: Trnplemet he reduiced phase,sbere the
Soted iteonedide key-Walue pais are processed and
aqgegatd based on ther tes
* Oeñae a nreduce funchion thot kes cash wntqe teg
as eqired
cutat of the weduced fuch on is atned a the finas
* The
cutput o the MapRedue'plo
Combine MapReduceobgi Canpose multiple Map Reduc Jobs to
- tefomcenle dada procesig asky thct reqre multple
Soges omptotion:
ahophmjze Pesformanea: Op tienize the perfomanee b4 Mop Reduue
(alalaions ky tuing parome e &uehas no