Mike Carey - University of California, Irvine...Data Models •A data model is a collection of concepts for describing data •Aschemais a description of a particular collection of

[email protected]

Michael Carey/Padhraic Smyth, UC Irvine: Stats 170A/B, Winter 2018 0

Announcements

• Read(andliveby!)thecoursewikipage:• https://grape.ics.uci.edu/wiki/asterix/wiki/stats170ab-2018

• Alsofollow(andliveby)thePiazzapage:• http://piazza.com/uci/winter2018/stats170a/home

• ThefirstHWassignmentisdueinoneweek:• https://grape.ics.uci.edu/wiki/asterix/attachment/wiki/stats170ab-2018/HW1.pdf

• (Wecantakealookatittogetherattheendoflecturetime.)

• Today:CS122AinaNutshell…..


WhatisaDatabaseSystem?• What’sadatabase?

• Averylarge,integratedcollectionofdata

• Usuallyamodelofareal-world enterprise• Entities (e.g.,students,courses,Facebookusers,…)withattributes(e.g.,name,birthdate,GPA,…)

• Relationships (e.g.,Susanistaking CS234,SusanisafriendofLynn,…)

• What’sadatabasemanagementsystem(DBMS)?• Asoftwaresystemdesignedtostore,manage,andprovideaccesstooneormoredatabases


FileSystemsvs.DBMS• Applicationprogramsmustsometimesstagelargedatasets betweenmainmemoryandsecondarystorage(forbufferinghugedatasets,gettingpage-orientedaccess,etc.)

• Specialcodeneededfordifferentqueries,andthatcodemustbe(stay)correctandefficient

• Mustprotectdatafrominconsistency duetomultipleconcurrentusers

• Crashrecoveryisimportantsincedataisnowthecurrencyoftheday(corporatejewels)

• Securityandaccesscontrolarealsoimportant(!)Michael Carey/Padhraic Smyth, UC Irvine: Stats 170A/B, Winter 2018 3

WhyUseaDBMS?• Dataindependence.• Efficientdataaccess.• Reducedapplicationdevelopmenttime.• Dataintegrityandsecurity.• Uniformdataadministration.• Concurrentaccess,recoveryfromcrashes.


• Canmake“BigData”muchlessunwieldy.• Makesiteasytoexploredata“declaratively”,includingcombinationsofmultipledatasets

DataModels

• Adatamodel isacollectionofconceptsfordescribingdata

• A schema isadescriptionofaparticularcollectionofdata,usingagivendatamodel

• Therelationalmodelis(still)themostwidelyuseddatamodeltoday

• Relation – basicallyatablewithrowsand(named)columns• Schema– describesthetablesandtheircolumns


Example:UniversityDB• Conceptualschema:

• Students(sid: string, name: string, login: string, age: integer, gpa: real)

• Courses(cid: string, cname: string, credits: integer) • Enrolled(sid: string, cid: string, grade: string)

• Physicalschema:• Relationsstoredasunorderedfiles• IndexonfirstandthirdcolumnsofStudents

• Externalschema(a.k.a.view):• CourseInfo(cid: string, cname: string, enrollment: integer)


UniversityDBExample(cont.)• Userquery(inSQL,againsttheexternalschema):

• SELECT c.cid, c.enrollmentFROM CourseInfo cWHERE c.cname = ‘Computer Game Design’

• Equivalentquery(inSQL,againsttheconceptualschema):• SELECT e.cid, count(e.*)

FROM Enrolled e, Courses cWHERE e.cid = c.cid AND c.cname = ‘Computer Game Design’GROUP BY c.cid

• Underthehood(againstthephysicalschema)• AccessCourses – useindexoncname tofindassociatedcid• AccessEnrolled – useindexoncid tocounttheenrollments


ABriefHistoryofDatabases• Pre-relationalera:1960’s,early1970’s• Codd’s seminalrelationalmodelpaper:1970• BasicRDBMSR&D:1970-80• RDBMSimprovements:1980-85• Relationalgoesmainstream:1985-90• ParallelDBMSresearch:1985-95• OLAPandwarehouseresearch:1990-2000• StreamDBandXMLDBresearch:2000-2010• “BigData”R&D(alsoincluding“NoSQL”):2005-present


OverviewofDatabaseDesign• Conceptualdesign:(ER Modelusedatthisstage.)

• Whataretheentities andrelationships intheenterprise?• Whatinformationabouttheseentitiesandrelationshipsshouldwestoreinthedatabase?

• Whataretheintegrityconstraintsorbusinessrulesthathold?

• AdatabaseschemaintheERModelcanberepresentedpictorially(usinganERdiagram).

• CanmapanERdiagramintoarelationalschema(manuallyorusingadesigntool’sautomation).


ERModelBasics

• Entity:Real-worldobject,distinguishablefromallotherobjects.Anentityisdescribed(inDB-land)usingasetofattributes.

• EntitySet:Acollectionofsimilarentities.E.g.,allemployees.

• Allentitiesinanentitysethavethesame setofattributes.(UntilwegettoISAhierarchies…)

• Eachentitysethasakey (auniqueidentifier);thiscanbeeitheroneattribute(an“atomic”key)orseveralattributes(calleda“composite”key)

• Eachattributehasadomain (similartoadatatype).

Employees

ssnname

lot


ERModelBasics(Contd.)

• Relationship:Associationamongtwoormoreentities.E.g.,SantaClausworksintheToydepartment.

• RelationshipSet:Collectionofsimilarrelationships.• Ann-ary relationshipsetRrelatesnentitysetsE1...En;eachrelationshipinRinvolvesentitiese1:E1,...,en:En

• Oneentitysetcanparticipateindifferentrelationshipsets –orindifferent“roles” inthesameset.

• Participationconstraintsindicatewhetheranentityisrequiredtoparticipateinagivenrelationship

Reports_To

lot

name

Employees

subor-dinate

super-visor

ssn

lotdname

budgetdid

sincename

Works_In DepartmentsEmployees

ssn


CardinalityConstraints• ConsiderWorksIn:Anemployeecanworkinmanydepartments;adeptcanhavemanyemployees.

• Incontrast,eachdepthasatmostonemanager,accordingtothecardinalityconstraint onManages above. Many-to-Many

(M:N)1-to-1(1:1)

1-to Many(1:N)

Many-to-1(N:1)

dnamebudgetdid

since

lot

name

ssn

ManagesEmployees Departments1 N

(Note: A given employee can manage several departments)

ERBasics:DoTryThisatHome(J)

• Let’sseeifyoucanread/interprettheERdiagramabove…!(J)• Whatattributesareunique(i.e.,identifytheirassociatedentityinstances)?• Whataretherulesabout(themuchcoveted)parkingpasses?• Whataretherules(constraints)aboutprofessorsbeingindepartments?• And,whataretherulesaboutprofessorsheadingdepartments?

rankname dname

dno main_office

In DeptProfessor

fac_id

Head

M N

1 NAssigned

Parking Space

pidlot_num

space_num

1

1


AnswerstotheSelfTest• Uniqueattributes:

• Professor.fac_id,Dept.dno,ParkingSpace.pid

• Facultyparking:• 1space/faculty,onefaculty/space• Somefacultycanbikeorwalk(J)• Someparkingspacesmaybeunused

• Facultyindepartments:• Facultymayhaveappointmentsinmultipledepartments• Departmentscanhavemultiplefacultyinthem• Noemptydepartments,andnounaffiliatedfaculty

• Departmentmanagement:• Oneheadperdepartment(exactly)• Notallfacultyaredepartmentheads

NOTE: These things are all “rules of the universe” that are just being modeled here!

Q: Can a faculty member head a department that he or she isn’t actually in?


LogicalDBDesign:ERtoRelational

• Entitysetstotables:

CREATE TABLE Employees (ssn CHAR(11),name CHAR(20),lot INTEGER,PRIMARY KEY (ssn))Employees

ssnname

lot

RelationshipSetstoTables• Intranslatingarelationshipsettoarelation,theattributesoftherelationmustinclude:

• Keysforeachparticipatingentityset(asforeignkeys).

• Thissetofattributesformsasuperkey fortherelation.

• Alldescriptiveattributes.

CREATE TABLE Works_In(ssn CHAR(11),did INTEGER,since DATE,PRIMARY KEY (ssn, did),FOREIGN KEY (ssn)

REFERENCES Employees,FOREIGN KEY (did)

REFERENCES Departments)

dname budgetdidsince

lotnamessn

WorksInEmployees Departments

KeyConstraints(Review)

• Eachdept hasatmostonemanager,accordingtothekeyconstraint onManages.

Translation to relational model?

Many-to-Many1-to-1 1-to Many Many-to-1

dname

budgetdid

since

lot

name

ssn

ManagesEmployees Departments1 N

ERTranslationwithKeyConstraints• Maptherelationshiptoatable(Manages):

• Notethatdid(alone)isthekey!

• StillseparatetablesforEmployeesandDepartments.

• But,sinceeachdepartmenthasauniquemanager,wecouldchoosetofoldManagesrightintoDepartments.

CREATE TABLE Manages (ssn CHAR(11),did INTEGER,since DATE,PRIMARY KEY (did),FOREIGN KEY (ssn) REFERENCES Employees,FOREIGN KEY (did) REFERENCES Departments)

CREATE TABLE Departments2 (did INTEGER,dname CHAR(20),budget REAL,mgr_ssn CHAR(11),mgr_since DATE,PRIMARY KEY (did),FOREIGN KEY (mgr_ssn) REFERENCES Employees)

vs.

(Q: Why do that...?)

Note: The relationshipinfo has been pushed tothe N-side’s entity table

MappingAdvancedERFeatures• Multi-valued(vs.single-valued)attributes

Employees

phonename

ssn

Employees

namessn address

snum

street

city

zip

v Composite(vs.atomic)attributes

Employees_phones(ssn, phone)• ssn is an FK in this table• (ssn, phone) is its PK

Employees(ssn, name, address_snum, address_street, address_city, address_zip)

Employees(ssn, name)• ssn is the PK in this table

So,GivenaRelationalSchema...• HowdoIknowifmyrelationalschemaisa“good”logicaldatabasedesignornot?

• Whatmightmakeit“notgood”?• HowcanIfixit,ifindeedit’s“notgood”?• How“good”isit,afterI’vefixedit?

• Notethatyourrelationalschemamighthavecomefromoneofseveralplaces

• YoustartedfromanE-Rmodel(butmaybethatmodelwas“wrong”insomeway?)

• Youwentstraighttorelationalinthefirstplace• It’snotyourschema– youinheritedit!J


Ex:WisconsinSailingClubProposed schema design #1:

Q: Do you think this is a “good” design? (Why or why not?)

sid sname rating age date bid bname color

22 Dustin 7 45.0 10/10/98 101 Interlake blue22 Dustin 7 45.0 10/10/98 102 Interlake red22 Dustin 7 45.0 10/8/98 103 Clipper green22 Dustin 7 45.0 10/7/98 104 Marine red31 Lubber 8 55.5 11/10/98 102 Interlake red31 Lubber 8 55.5 11/6/98 103 Clipper green31 Lubber 8 55.5 11/12/98 104 Marine red... ... ... ... ... ... ... ...

Ex:WisconsinSailingClubProposed schema design #2:

Q: What about this design?• Is #2 “better than #1...?

Explain!• Is it a “best” design?• How can we go from

design #1 to this one?

sid sname rating age

22 Dustin 7 45.031 Lubber 8 55.5... ... ... ...

sid

bid date

22 101 10/10/9822 102 10/10/9822 103 10/8/9822 104 10/7/9831 102 11/10/9831 103 11/6/9831 104 11/12/98... ... ...

bid bname color

101 Interlake blue102 Interlake red103 Clipper green104 Marine red

NormalForms

All “relations”

1NF

2NF

3NF

BCNF. . .


FirstNormalForm(1NF)• Rel’n Risin1NF ifallofitsattributesareatomic.

• Noset-valuedattributes!(1NF=“flat”J)• Usuallygoesw/osayingforrelationalmodel(butnotforNoSQLsystems,aswe’llseeattheendofthequarterJ).

• Ex:

bname color

Interlake blue, redClipper greenMarine red

bname color

Interlake blueInterlake redClipper greenMarine red


OntoSQL...!

• relation-list Alistofrelationnames(possiblywitharange-variable aftereachname).

• target-list Alistofattributesofrelationsinrelation-list• qualification Comparisons(Attr op const orAttr1op Attr2,whereop isoneof<,<=,=,>,>=,<>)combinedusingAND,ORandNOT.

• DISTINCT isanoptionalkeywordindicatingthattheanswershouldnotcontainduplicates.Defaultisthatduplicatesarenot eliminated!(Bags,notsets.)

SELECT [DISTINCT] target-listFROM relation-listWHERE qualification

SQL “SPJ” Query:


ManySQL-BasedDBMSs• CommercialRDBMSchoicesinclude

• DB2(IBM)• Oracle• SQLServer(Microsoft)• Teradata

• OpensourceRDBMSoptionsinclude• MySQL• PostgreSQL

• Andforso-called“BigData”,wealsohave• ApacheHive(onHadoop)+newerwannabees


ExampleInstances

• We’llusetheseinstancesofourusualSailorsandReservesrelationsinourexamples.

sid sname rating age22 dustin 7 45.031 lubber 8 55.558 rusty 10 35.0sid sname rating age28 yuppy 9 35.031 lubber 8 55.544 guppy 5 35.058 rusty 10 35.0

sid bid day22 101 10/10/9658 103 11/12/96

R1

S1

S2

ConceptualEvaluationStrategy• SemanticsofanSQLquerydefinedintermsofthefollowingconceptualevaluationstrategy:

• Computethecross-productofrelation-list.(✕)• Discardresultingtuplesiftheyfailqualifications.(σ)• Projectoutattributesthatarenotintarget-list.(π)• IfDISTINCT isspecified,eliminateduplicaterows. (δ)

• Thisstrategyisprobablytheleast efficientwaytocomputeaquery!Anoptimizerwillfindmoreefficientstrategiestocomputethesameanswers.


ExampleofConceptualEvaluationSELECT S.snameFROM Sailors S, Reserves R ß using table S1WHERE S.sid=R.sid AND R.bid=103

(sid) sname rating age (sid) bid day 22 dustin 7 45.0 22 101 10/10/96 22 dustin 7 45.0 58 103 11/12/96 31 lubber 8 55.5 22 101 10/10/96 31 lubber 8 55.5 58 103 11/12/96 58 rusty 10 35.0 22 101 10/10/96

58 rusty 10 35.0 58 103 11/12/96

Findsailorswhoʼve reservedatleastoneboat

• WouldaddingDISTINCTtothisquerymakeadifference?(Withourdata?Withpossibledata?)

• WhatistheeffectofreplacingS.sid byS.sname intheSELECT clause?WouldaddingDISTINCT tothis variantofthequerymakeadifference?

SELECT S.sidFROM Sailors S, Reserves RWHERE S.sid=R.sid

Sailors(sid,sname,rating,age)Reserves(sid,bid,day)Boats(bid,bname,color)


ExpressionsandStrings

• Illustratesuseofarithmeticexpressionsandstringpatternmatching:FindnamesandagesandafielddefinedbyanexpressionforsailorswhosenamesbeginandendwithBandcontainatleastthreecharacters.

• ASprovidesawayto(re)namefieldsinresult.• LIKE isusedforstringmatching.`_ʼ standsforanyonecharacterand`%ʼ standsfor0ormorearbitrarycharacters.(SeeSQLdocsformoreinfo...)

SELECT S.sname, S.age, (7 * S.age) AS dogyearsFROM Sailors SWHERE S.sname LIKE ‘B_%B’

Sailors(sid,sname,rating,age)


SomeBetterExampleData

SailorsReserves

Boats

Findsidʼsofsailorswhoʼve reservedaredor agreenboat

• IfwereplaceORbyANDinthisfirstversion,whatdoweget?

• UNION:Canbeusedtocomputetheunionofanytwounion-compatible setsoftuples(whicharethemselvestheresultofSQLqueries).

• Alsoavailable:EXCEPT (WhatwouldwegetifwereplacedUNIONbyEXCEPT?)

SELECT DISTINCT S.sidFROM Sailors S, Boats B, Reserves RWHERE S.sid=R.sid AND R.bid=B.bid

AND (B.color=‘red’ OR B.color=‘green’)

(SELECT S.sidFROM Sailors S, Boats B, Reserves RWHERE S.sid=R.sid AND R.bid=B.bid

AND B.color=‘red’)UNION(SELECT S.sidFROM Sailors S, Boats B, Reserves RWHERE S.sid=R.sid AND R.bid=B.bid

AND B.color=‘green’)


Findsidʼsofsailorswhoʼve reservedaredand agreenboat

• INTERSECT:Canbeusedtocomputetheintersectionoftwounion-compatiblesetsoftuples.

• IncludedintheSQL/92standard,butnot inallsystems(e.g.,MySQL).

• ContrastsymmetryoftheUNION andINTERSECTquerieswithhowmuchtheotherversionsdiffer.

SELECT S.sidFROM Sailors S, Boats B1, Reserves R1,

Boats B2, Reserves R2WHERE S.sid=R1.sid AND R1.bid=B1.bid

AND S.sid=R2.sid AND R2.bid=B2.bidAND (B1.color=‘red’ AND B2.color=‘green’)

Key field!SELECT S.sidFROM Sailors S, Boats B, Reserves RWHERE S.sid=R.sid AND R.bid=B.bid

AND B.color=‘red’INTERSECTSELECT S.sidFROM Sailors S, Boats B, Reserves RWHERE S.sid=R.sid AND R.bid=B.bid

AND B.color=‘green’


NestedQueries

• AverypowerfulfeatureofSQL:aWHERE clausecanitselfcontainanSQLquery!(Actually,socanFROM andHAVINGclauses!!)

• Tofindsailorswho’venot reserved#103,useNOTIN.• Tounderstandsemantics(includingcardinality)ofnestedqueries,thinknestedloops evaluation:

• ForeachSailorstuple,checkqualificationbycomputingsubquery.

SELECT S.snameFROM Sailors SWHERE S.sid IN (SELECT R.sid

FROM Reserves RWHERE R.bid=103)

Find names of sailors who’ve reserved boat #103:


NestedQuerieswithCorrelation

• EXISTS isanothersetcomparisonoperator,likeIN.• Illustrateswhy,ingeneral,subquery mustbere-computedforeachSailorstuple(conceptually).NOTE:Recallthattherewasajoinwaytoexpressthisquery,too.Relationalqueryoptimizerswilltrytounnest queriesintojoinswhenpossibletoavoidnestedloopqueryevaluationplans.

SELECT S.snameFROM Sailors SWHERE EXISTS (SELECT *

FROM Reserves RWHERE R.bid=103 AND S.sid=R.sid)

Find names of sailors who’ve reserved boat #103:


MoreonSet-ComparisonOperators

• Weʼve alreadyseenIN andEXISTS..CanalsouseNOT IN andNOT EXISTS.

• Alsoavailable:op ANY,op ALL (forops: )

• FindsailorswhoseratingisgreaterthanthatofsomesailorcalledHoratio:

> < = ³ £ ¹, , , , ,

SELECT *FROM Sailors SWHERE S.rating > ANY (SELECT S2.rating

FROM Sailors S2WHERE S2.sname=‘Horatio’)


RewritingINTERSECT QueriesUsingIN

• Similarly,EXCEPT queriescanbere-writtenusingNOTIN.• Thisiswhatyou’llneedtodowhenusingsystemslikeMySQLwhosesetoperatorcollectionisincomplete

Find sid’s of sailors who’ve reserved both a red and a green boat:

SELECT S.sidFROM Sailors S, Boats B, Reserves RWHERE S.sid=R.sid AND R.bid=B.bid AND B.color=‘red’

AND S.sid IN (SELECT S2.sidFROM Sailors S2, Boats B2, Reserves R2WHERE S2.sid=R2.sid AND R2.bid=B2.bid

AND B2.color=‘green’)


Orderingand/orLimitingQueryResults

Find the ratings, ids, names, and ages of the three best sailors

SELECT S.rating, S.sid, S.sname, S.ageFROM Sailors SORDER BY S.rating DESCLIMIT 3

SELECT [DISTINCT] expressionsFROM tables[WHERE condition]....

[ORDER BY expression [ ASC | DESC ]]LIMIT number_rows [ OFFSET offset_value ];

v Thegeneralsyntaxforthis:


AggregateOperators

• Significantextensionoftherelationalalgebra.

COUNT (*)COUNT ( [DISTINCT] A)SUM ( [DISTINCT] A)AVG ( [DISTINCT] A)MAX (A)MIN (A)

SELECT AVG (S.age)FROM Sailors SWHERE S.rating=10

SELECT COUNT (*)FROM Sailors S

SELECT AVG(DISTINCT S.age)FROM Sailors SWHERE S.rating=10

SELECT S.snameFROM Sailors SWHERE S.rating= (SELECT MAX(S2.rating)

FROM Sailors S2)

single column

SELECT COUNT (DISTINCT S.rating)FROM Sailors SWHERE S.sname=‘Bob’

Findnameandageoftheoldestsailor(s)

• Thatfirsttryisillegal!(We’llseewhyshortly,whenwedoGROUPBY.)

SELECT S.sname, MAX (S.age)FROM Sailors S

SELECT S.sname, S.ageFROM Sailors SWHERE S.age =

(SELECT MAX (age)FROM Sailors)

SELECT S.sname, MAX (S.age)FROM Sailors S

MotivationforGrouping

• Sofar,weʼve appliedaggregateoperatorstoall(qualifying)tuples.Sometimes,wewanttoapplythemtoeachofseveralgroups oftuples.

• Consider:Findtheageoftheyoungestsailorforeachratinglevel.

• Ingeneral,wedonʼtknowhowmanyratinglevelsexist,andwhattheratingvaluesfortheselevelsare!

• Supposeweknowthatratingvaluesgofrom1to10;wecanwrite10queriesthatlooklikethis(J):

SELECT MIN (S.age)FROM Sailors SWHERE S.rating = i

For i = 1, 2, ... , 10:


QueriesWithGROUPBYandHAVING

• The target-listcontains (i)attributenames and(ii)termswithaggregateoperations(e.g.,MIN(S.age)).

• Theattributelist(i) mustbeasubsetofgrouping-list.Intuitively,eachanswertuplecorrespondstoagroup,and theseattributesmusthavea singlevaluepergroup.(Agroup isasetoftuplesthathavethesamevalueforallattributesingrouping-list.)

SELECT [DISTINCT] target-listFROM relation-listWHERE qualificationGROUP BY grouping-listHAVING group-qualification


ConceptualEvaluation• Thecross-productofrelation-list iscomputed,tuplesthatfailthequalification arediscarded,`unnecessary’ fieldsaredeleted,andtheremainingtuplesarepartitionedintogroupsbythevalueofattributesingrouping-list.

• Agroup-qualification (HAVING)isthenappliedtoeliminatesomegroups.Expressionsingroup-qualificationmustalsohaveasinglevaluepergroup!

• Ineffect,anattributeingroup-qualification thatisnotanargumentofanaggregateopmustappearingrouping-list.(Note:SQLdoesn’tconsiderprimarykeysemanticshere.)

• Oneanswertupleisgeneratedperqualifyinggroup.


Findageoftheyoungestsailorwithage18foreachratingwithatleast2such sailors.

rating minage 3 25.5 7 35.0 8 25.5

SELECT S.rating, MIN (S.age) AS minage

FROM Sailors SWHERE S.age >= 18GROUP BY S.ratingHAVING COUNT (*) >= 2

sid sname rating age 22 dustin 7 45.0 29 brutus 1 33.0 31 lubber 8 55.5 32 andy 8 25.5 58 rusty 10 35.0 64 horatio 7 35.0 71 zorba 10 16.0 74 horatio 9 35.0 85 art 3 25.5 95 bob 3 63.5 96 frodo 3 25.5

Answer relation:

³

Sailors instance:


Findageoftheyoungestsailorwithage18foreachratingwithatleast2such sailors.

rating minage 3 25.5 7 35.0 8 25.5

rating age 7 45.0 1 33.0 8 55.5 8 25.5 10 35.0 7 35.0 10 16.0 9 35.0 3 25.5 3 63.5 3 25.5

³

rating age 1 33.0 3 25.5 3 63.5 3 25.5 7 45.0 7 35.0 8 55.5 8 25.5 9 35.0 10 35.0


Foreachredboat,findthenumberofreservationsforthisboat

• Notice:We’regroupingoverajoinofthreerelations

SELECT B.bid, COUNT(*) AS scountFROM Sailors S, Boats B, Reserves RWHERE S.sid=R.sid AND R.bid=B.bid AND B.color=‘red’GROUP BY B.bid


Findageoftheyoungestsailorwithage>18foreachratingwithatleast2sailors(ofany age)

• Notes:AHAVING clausecanalsocontainasubquery.

SELECT S.rating, MIN(S.age)FROM Sailors SWHERE S.age > 18GROUP BY S.ratingHAVING 1 < (SELECT COUNT(*)

FROM Sailors S2WHERE S.rating = S2.rating)


FindthoseratingsforwhichtheaverageageistheminimumageoverallSailors

SELECT Temp.rating, Temp.avgageFROM (SELECT S.rating, AVG(S.age) AS avgage

FROM Sailors SGROUP BY S.rating) AS Temp

WHERE Temp.avgage = (SELECT MIN(age) FROM Sailors)

v Correct solution (in SQL/92): Compute theaverage age foreach rating...

Find the overallminimum age


NullValues• Fieldvaluesinatuplearesometimesunknown(e.g.,aratinghasnotbeenassigned)orinapplicable(e.g.,nospouseʼsname).

• SQLprovides specialvaluenull forsuchsituations.

• Thepresenceofnull complicatesmanyissues.E.g.:• Specialoperatorsneededtocheckifvalueis/isnotnull.• Israting>8 trueorfalsewhenrating isequaltonull?WhataboutAND,ORandNOT connectives?

• Weneeda3-valuedlogic (true,falseandunknown).• Meaningofconstructsmustbedefinedcarefully.(TheWHEREclauseeliminatesrowsthatdonʼtevaluatetotrue.)

• Newoperators(inparticular,outerjoins)possible/needed.


ExampleDatawithNullValues

SailorsReserves

Boats

NullsandSQL’s3-ValuedLogicAND true false unknown

true true false unknown

false false false false

unknown unknown false unknown

OR true false unknown

true true true true

false true false unknown

unknown true unknown unknown

NOTtrue falsefalse true

unknown unknown

Note: SQL arithmetic expressions involving nullvalues will yield null values (Ex: EMP.sal + EMP.bonus)


BasicSQLQueriesw/NullsSELECT *FROM Sailors SWHERE age > 35.0

SELECT *FROM Sailors SWHERE age <= 35.0

SELECT COUNT(*)FROM Sailors SWHERE age > 35.0

OR age <= 35.0

SELECT COUNT(*)FROM Sailors SWHERE age > 35.0

OR age <= 35.0OR age IS NULL


Nullsw/AggregatesSELECT COUNT(rating)FROM Sailors

SELECTCOUNT (DISTINCT rating)FROM Sailors

SELECT SUM(rating),COUNT(rating),AVG(rating)

FROM Sailors

(11)

(7)

(70, 11, 6.3636)

(Useful, but logically “wrong”!)Michael Carey/Padhraic Smyth, UC Irvine: Stats 170A/B, Winter 2018 54

Nullsw/Aggregates&Grouping

SELECT bid, COUNT(*)FROM ReservesGROUP BY bid

SELECT COUNT( DISTINCT bid)FROM Reserves (4)


Nullsw/Joinsà Innervs.OuterJoinsSome “dangling” tuple examples


Inner vs.OuterJoinsinSQL

SELECT DISTINCT s.sname, r.dateFROM Sailors s, Reserves rWHERE s.sid = r.sid


Inner vs.OuterJoinsinSQL(2)

SELECT DISTINCT s.sname, r.dateFROM Sailors s INNER JOIN Reserves r ON s.sid = r.sid


Innervs.Outer JoinsinSQL(3)(1) SELECT DISTINCT s.sname, r.dateFROM Sailors s LEFT OUTER JOIN Reserves r ON s.sid = r.sid

v Variationsonatheme:§ JOIN(orINNERJOIN)§ LEFTOUTERJOIN§ RIGHTOUTERJOIN§ FULLOUTERJOIN(VariesfromRDBMStoRDBMS)(Seesystem’sdocumentationforjoinsyntax)

(2) SELECT DISTINCT s.sname, r.dateFROM Reserves r RIGHT OUTER JOIN Sailors s ON s.sid = r.sid


NowLet’sPeekatHW#1

• https://grape.ics.uci.edu/wiki/asterix/attachment/wiki/stats170ab-2018/HW1.pdf


Documents

Mike Carey - University of California, Irvine...Data Models •A data model is a collection of concepts for describing data •Aschemais a description of a particular collection of