Bill howe 2_databases

DATA MODELS
AND DATABASES
BILL HOWE, PHD
DIRECTOR OF RESEARCH, SCALABLE DATA ANALYTICS
UNIVERSITY OF WASHINGTON ESCIENCE INSTITUTE
1

HOW DOWE STORE DATA?
3
###query length COG hit #1 e-value #1 identity #1 score #1 hit length #1 description #1
chr_4[480001-580000].287 4500
chr_4[560001-660000].1 3556
chr_9[400001-500000].503 4211 COG4547 2.00E-04 19 44.6 620 Cobalamin biosynthesis protein CobT (nicot
chr_9[320001-420000].548 2833 COG5406 2.00E-04 38 43.9 1001 Nucleosome binding factor SPN, SPT16 subu
chr_27[320001-404298].20 3991 COG4547 5.00E-05 18 46.2 620 Cobalamin biosynthesis protein CobT (nicot
chr_26[320001-420000].378 3963 COG5099 5.00E-05 17 46.2 777 RNA-binding protein of the Puf family, trans
chr_24[160001-260000].65 3542
chr_9[160001-260000].243 3002 COG5077 1.00E-25 26 114 1089 Ubiquitin carboxyl-terminal hydrolase
chr_12[720001-820000].86 2895 COG5032 2.00E-09 30 60.5 2105 Phosphatidylinositol kinase and protein kina
chr_12[800001-900000].109 1463 COG5032 1.00E-09 30 60.1 2105 Phosphatidylinositol kinase and protein kina
chr_11[1-100000].70 2886
chr_11[80001-180000].100 1523
ANNOTATIONSUMMARY-COMBINEDORFANNOTATION16_Phaeo_genome
What is the data model?

WHAT IS A DATA MODEL?
4
Three components:
1.  Structures
2.  Constraints
3.  Operations

EXAMPLES
5
1.  Structures
•  rows and columns?
•  nodes and edges?
•  key-value pairs?
•  a sequence of bytes?
2.  Constraints
•  all rows must have the same number of columns
•  all values in one column must have the same type
•  a child cannot have two parents
3.  Operations
•  find the value of key x
•  find the rows where column “lastname” is “Jordan”
•  get the next N bytes

WHAT IS A DATABASE?
6
A collection of information
organized to afford efficient
retrieval
http://www.usg.edu/galileo/skills/unit04/primer04_01.phtml

ANOTHERVIEW
7
“When people use the word database, fundamentally what
they are saying is that the data should be self-describing and
it should have a schema.That’s really all the word database
means.”
-- Jim Gray,“The Fourth Paradigm”

WHYWOULD IWANT A DATABASE?
8
1.  Sharing
Support concurrent access by multiple readers and writers
2.  Data Model Enforcement
Make sure all applications see clean, organized data
3.  Scale
Work with datasets too large to fit in memory
4.  Flexibility
Use the data in new, unanticipated ways

QUESTIONS TO CONSIDER
How is the data physically organized on disk?
What kinds of queries are efficiently supported by
this organization, and what kinds are not?
How hard is it update the data, or add new data?
What happens when I encounter new queries that I
didn’t anticipate? Do I reorganize the data? How hard
is that?
9

MOTIVATING
RELATIONAL
DATABASES
BILL HOWE, PHD
10

QUESTIONS TO CONSIDER
How is the data physically organized on disk?
What kinds of queries are efficiently supported by this
organization, and what kinds are not?
How hard is it update the data, or add new data?
What happens when I encounter new queries that I didn’t
anticipate? Do I reorganize the data? How hard is that?
11

HISTORICAL EXAMPLE:NETWORK DATABASES
12
Database:A collection of information
organized to afford efficient retrieval
Orderer
Customer
Screw
Nut
Washer
Contact Rep

HISTORICAL EXAMPLE:HIERARCHICAL DATABASES
13
Orderer
Customer
Screw
Nut
Nail
Contact Rep
Orderer Screw
Nut
Washer
master
detail
detail
Works great if you want to find all
orders for a particular customer.
But what if you want to find all
Customers who ordered a Nail?

ONEVIEW
14
“Relational Database Management Systems were invented
to let you use one set of data in multiple ways, including
ways that are unforeseen at the time the database is built
and the 1st applications are written.”
- Curt Monash, analyst/blogger

RELATIONAL DATABASES (CODD 1970)
Everything is a table
Every row in a table has the same columns
Relationships are implicit: no pointers
15
Course Student Id
CSE 344 223…
CSE 344 244…
CSE 514 255..
CSE 514 244…
Student Id Student
Name
223… Jane
244… Joe
255.. Susan

DATABASE PHILOSOPHY
16
God made the integers;
all else is the work of
man.
- Leopold Kronecker, 19th Century
Mathematician
slide src: Mike Franklin
Codd made relations;
all else is the work of man.
- Raghu Ramakrishnan, DB text book author

RELATIONAL DATABASE HISTORY
17
Pre-Relational: if your data changed, your application broke.
Early RDBMS were buggy and slow (and often reviled), but required only 5% of
the application code.
“Activities of users at terminals and most application programs
should remain unaffected when the internal representation of data
is changed and even when some aspects of the external
representation are changed.”
Key Ideas: Programs that manipulate tabular data exhibit an
algebraic structure allowing reasoning and manipulation
independently of physical data representation
- Codd 1979

KEY IDEA:“PHYSICAL DATA INDEPENDENCE”
18
physical data independence
files and
pointers
relations
SELECT seq
FROM ncbi_sequences
WHERE seq =
‘GATTACGATATTA’;
f = fopen(‘table_file’);
fseek(10030440);
while (True) {
fread(&buf, 1, 8192, f);
if (buf == GATTACGATATTA) {
. . .

KEY IDEA:AN ALGEBRA OF TABLES
19
select
project
join join
Other operators: aggregate,union,difference,cross product

KEY IDEA:ALGEBRAIC OPTIMIZATION
20
N = ((z*2)+((z*3)+0))/1
Algebraic Laws:
1. (+) identity: x+0 = x
2. (/) identity: x/1 = x
3. (*) distributes: (n*x+n*y) = n*(x+y)
4. (*) commutes: x*y = y*x
Apply rules 1, 3, 4, 2:
N = (2+3)*z
two operations instead of five, no division operator
Same idea works with the Relational Algebra!

EQUIVALENT LOGICAL EXPRESSIONS;
DIFFERENT COSTS
21
σp=knows(R) o=s (σp=holdsAccount(R) o=s σp=accountHomepage(R))
(σp=knows(R) o=s σp=holdsAccount(R)) o=s σp=accountHomepage(R)
σp1=knows & p2=holdsAccount & p3=accountHomepage (R x R x R)
right associative
left associative
cross product

SAME LOGICAL EXPRESSION,
DIFFERENT PHYSICAL ALGORITHMS
22
A = select(p=knows)
B = select(p=holdsAccount)
C = select(p=accountWebpage)
hA = hash(A)
AB = probe hA with B
hC = hash(C)
ABC = probe hC with AB
A = select(p=knows)
B = select(p=holdsAccount)
C = select(p=accountWebpage)
hB = hash(B)
AB = probe hB with A
hC = hash(C)
ABC = probe hC with AB
Which is faster?

ALGEBRAIC OPTIMIZATION MATTERS
23
BTC 2010 Dataset
3B quads
623 GB processed

PICASSO QUERY PLAN DIAGRAMS
24
Haritsa,VLDB 2010

KEY IDEA:DECLARATIVE LANGUAGES
25
SELECT *
FROM Order o, Item i
WHERE o.item = i.item
AND o.date = today()
join
select
scan scan
date = today()
o.item = i.item
Order oItem i
Find all orders from today,along with the items ordered

SQL IS THE“WHAT”NOT THE“HOW”
26
SELECT DISTINCT x.name, z.name
FROM Product x, Purchase y, Customer z
WHERE x.pid = y.pid and y.cid = y.cid and
x.price > 100 and z.city = ‘Seattle’
It’s clear WHAT we want, unclear HOW to get it
Product(pid, name, price)
Purchase(pid, cid, store)
Customer(cid, name, city)

RELATIONAL ALGEBRA
27
Product Purchase
pid=pid
price>100 and city=‘Seattle’
x.name,z.name
δ
cid=cid
Customer
Π
σ
T1(pid,name,price,pid,cid,store)
T2( . . . .)
T4(name,name)
Final answer
T3(. . . )
Execution order is
now clearly
specified
Product(pid, name, price)
Purchase(pid, cid, store)
But a lot of
physical details
are still left open!

ANOTHER EXAMPLE
28
left.object = right.subject
σ
predicate=knows
SELECT r1.subject
FROM R r1, R r2, R r3
WHERE r1.predicate = ‘knows’
AND r2.predicate = ‘holdsAccount’
AND r3.predicate = ‘accountHomepage’
AND r1.object = r2.subject
AND r2.object = r3.subject
σ
predicate=holdsAccount
left.object = right.subject
σ
predicate=
accountHomepage
R(subject, predicate,
object)
π
r1.subject

KEY IDEA:“LOGICAL DATA INDEPENDENCE”
29
physical data independence
logical data independence
files and
pointers
relations
views
SELECT *
FROM my_sequences
SELECT seq
FROM ncbi_sequences
WHERE seq =
‘GATTACGATATTA’;
f = fopen(‘table_file’);
fseek(10030440);
while (True) {
fread(&buf, 1, 8192, f);
if (buf == GATTACGATATTA) {
. . .

WHAT AREVIEWS?
A view is just a query with a name
We can use the view just like a real table
30
Why can we do
this?
Because we know that every
query returns a relation:
We say that the language is
“algebraically closed”

VIEW EXAMPLE
A view is a relation defined by a query
31
CREATE VIEW StorePrice AS
SELECT x.store, y.price
FROM Purchase x, Product y
WHERE x.pid = y.pid
This is like a new table
StorePrice(store,price)
Purchase(customer, product, store)
Product(pname, price)
StorePrice(store, price)

HOW TO USE AVIEW?
32
A "high end" store is a store that sold some product
over 1000. For each customer, find all the high end
stores that they visit. Return a set of (customer-
name, high-end-store) pairs.
SELECT DISTINCT z.name, u.store
FROM Customer z, Purchase u, StorePrice v
WHERE z.cid = u.customer
AND u.store = v.store
AND v.price > 1000
StorePrice(store, price)
Purchase(customer, product, store)
Product(pname, price)

KEY IDEA:INDEXES
33
Databases are especially, but not exclusively, effective
at“Needle in Haystack” problems:
•  Extracting small results from big datasets
•  Transparently provide “old style” scalability
•  Your query will always* finish, regardless of dataset size.
•  Indexes are easily built and automatically used when
appropriate
CREATE INDEX seq_idx ON sequence(seq);
SELECT seq
FROM sequence
WHERE seq = ‘GATTACGATATTA’;
*almost

IN-DATABASE
ANALYTICS
BILL HOWE, PHD
34

35
“There is no point in bringing data ... into the data warehouse
environment without integrating it. If the data arrives at the data
warehouse in an unintegrated state, it cannot be used to support a
corporate view of data. And a corporate view of data is one of the
essences of the architected environment.”
- Inmon, 2005,“Building the Data Warehouse”
Not a good fit for analytics!

MATRIX ADDITION,VECTOR REPRESENTATION
36
SELECT A.row_number, A.vector + B.vector
FROM A, B
WHERE A.row_number = B.row_number;
A(row_number,row float[])
B(row_number,row float[])

MATRIX MULTIPLICATION,VECTOR
REPRESENTATION
38
SELECT 1, array_accum(row_number, vector*v) FROM A;

MATRIX TRANSPOSE,VECTOR
REPRESENTATION
39
SELECT S.col_number, array_accum(A.row_number, A.vector[S.col_number])
FROM A, generate_series(1,3) AS S(col_number)
GROUP BY S.col_number;
generate_series is a table function

MATRIX MULTIPLICATION,SPARSE
REPRESENTATION
40
SELECT A.row_number, B.column_number, SUM(A.value * B.value)
FROM A, B
WHERE A.column_number = B.row_number
GROUP BY A.row_number, B.column_number
A(row_number,column_number,value)
B(row_number,column_number,value)

ASIDE:USER-DEFINED FUNCTIONS AND
TYPES
41
Scalar functions
SELECT myfunc(r.a, r.b)…
Aggregate functions
SELECT concat(r.s) …
Table functions
SELECT … FROM tablefunc(a,b)

EXPERIMENT DESIGN
42
CREATE VIEW trials AS
SELECT d.trial_id, AVG(a.values) AS avg_value
FROM design d,T
WHERE d.row_id = T.row_id
GROUP BY d.trial_id
CREATE VIEW design AS
SELECT a.trial_id,
floor (100 * random()) AS row_id
FROM generate_series(1,10000) AS a (trial_id),
generate_series(1,3) AS b (subsample_id)
src: Cohen et al.,VLDB 2009

43
Prior to the implementation of this functionality within the
DBMS, one Greenplum customer was accustomed to
calculating the OLS by exporting data and importing the
data into R for calculation, a process that took several
hours to complete.They reported signicant performance
improvement when they moved to running the regression
within the DBMS. Most of the benet derived from running
the analysis in parallel close to the data with minimal
data movement.
src: Cohen et al.,VLDB 2009

SLIDES CAN BE FOUND AT:
TEACHINGDATASCIENCE.ORG

Bill howe 2_databases

More Related Content

Similar to Bill howe 2_databases (20)

More from Mahammad Valiyev (6)

Recently uploaded (20)

Bill howe 2_databases