1. New
Opportuni0es
for
Connected
Data
@ianSrobinson
ian.robinson@neotechnology.com
#neo4j

2. Agenda
• Data
complexity
• Graph
databases
–
features
and
benefits
• Querying
graph
data
• Use
cases
#neo4j

3. Data
Complexity
complexity = f(size, semi-structure, connectedness)
#neo4j

4. Data
Complexity
complexity = f(size , semi-structure, connectedness)
#neo4j

5. Semi-­‐Structure
#neo4j

6. Semi-­‐Structure
USER_ID
FIRST_NAME
LAST_NAME
EMAIL_1
EMAIL_2
FACEBOOK
TWITTER
SKYPE
315
Ian
Robinson
ian@neotechnology.com
iansrobinson@gmail.com
NULL
@iansrobinson
iansrobinson
Email:
ian@neotechnology.com
Email:
iansrobinson@gmail.com
CONTACT
Twi*er:
@iansrobinson
0..n
Skype:
iansrobinson
USER
CONTACT_TYPE
#neo4j

7. Connectedness
#neo4j

8. #neo4j

9. #neo4j

10. #neo4j

11. #neo4j

12. #neo4j

13. #neo4j

14. #neo4j

15. Graphs
Are
Everywhere
#neo4j

16. Neo4j
is
a
Graph
Database
• Store
• Manage
• Query
data
as
a
graph
#neo4j

17. Property
Graph
Data
Model
#neo4j

18. Graph
Database
Benefits
Densely-­‐connected,
semi-­‐structured
domains
• Lots
of
join
tables?
Connectedness
• Lots
of
sparse
tables?
Semi-­‐structure
Data
Model
Vola=lity
• Easy
to
evolve
Join
Complexity
and
Performance
• Millions
of
‘joins’
per
second
• Consistent
query
0mes
as
dataset
grows
#neo4j

19. Querying
Graph
Data
• A
language
for
describing
graphs
• Crea0ng
nodes,
rela0onships
and
proper0es
• Querying
data
#neo4j

20. Describing
Graphs
#neo4j

21. Cypher
(graphs)<-[:INTERESTED_IN]-(ben)-[:INTERESTED_IN]->(rest),
(ben)-[:WORKS_FOR]->(acme)
#neo4j

22. Cypher
(ben)-[:INTERESTED_IN]->(graphs),
(ben)-[:INTERESTED_IN]->(rest),
(ben)-[:WORKS_FOR]->(acme)
#neo4j

23. Create
Some
Data
CREATE ben = { name : 'Ben' },
acme = { name : 'Acme' },
rest = { label : 'REST' },
graphs = { label : 'Graphs' },
ben-[:WORKS_FOR]->acme,
ben-[:INTERESTED_IN]->rest,
ben-[:INTERESTED_IN]->graphs
RETURN ben
#neo4j

24. Create
Nodes
CREATE ben = { name : 'Ben' },
acme = { name : 'Acme' },
rest = { label : 'REST' },
graphs = { label : 'Graphs' },
ben-[:WORKS_FOR]->acme,
ben-[:INTERESTED_IN]->rest,
ben-[:INTERESTED_IN]->graphs
RETURN ben
#neo4j

25. Create
Rela0onships
CREATE ben = { name : 'Ben' },
acme = { name : 'Acme' },
rest = { label : 'REST' },
graphs = { label : 'Graphs' },
ben-[:WORKS_FOR]->acme,
ben-[:INTERESTED_IN]->rest,
ben-[:INTERESTED_IN]->graphs
RETURN ben
#neo4j

26. Return
Node
CREATE ben = { name : 'Ben' },
acme = { name : 'Acme' },
rest = { label : 'REST' },
graphs = { label : 'Graphs' },
ben-[:WORKS_FOR]->acme,
ben-[:INTERESTED_IN]->rest,
ben-[:INTERESTED_IN]->graphs
RETURN ben
#neo4j

27. Eventually…
#neo4j

28. Querying
a
Graph
Graph
local:
• One
or
more
start
nodes
• Explore
surrounding
graph
• Millions
of
joins
per
second
#neo4j

29. Queries:
Pacern
Matching
Pacern
#neo4j

30. Start
Node
Pacern
#neo4j

31. Match
Pacern
#neo4j

32. Match
Pacern
#neo4j

33. Match
Pacern
#neo4j

34. Non-­‐Match
Pacern
#neo4j

35. Non-­‐Match
Pacern
#neo4j

36. Colleagues
Who
Share
My
Interests
#neo4j

37. Cypher
Pacern
me-[:INTERESTED_IN]->topic,
me-[:WORKS_FOR]->company,
colleague-[:INTERESTED_IN]->topic,
colleague-[:WORKS_FOR]->company
#neo4j

38. Find
Colleagues
START me=node:users(name = "sarah")
MATCH me-[:INTERESTED_IN]->topic,
me-[:WORKS_FOR]->company,
colleague-[:INTERESTED_IN]->topic,
colleague-[:WORKS_FOR]->company
RETURN colleague.name AS name,
count(topic) AS score,
collect(topic.label) AS interests
ORDER BY score DESC
#neo4j

39. Find
Start
Node
START me=node:users(name = "sarah")
MATCH me-[:INTERESTED_IN]->topic,
me-[:WORKS_FOR]->company,
colleague-[:INTERESTED_IN]->topic,
colleague-[:WORKS_FOR]->company
RETURN colleague.name AS name,
count(topic) AS score,
collect(topic.label) AS interests
ORDER BY score DESC
#neo4j

40. Describe
Pacern
START me=node:users(name = "sarah")
MATCH me-[:INTERESTED_IN]->topic,
me-[:WORKS_FOR]->company,
colleague-[:INTERESTED_IN]->topic,
colleague-[:WORKS_FOR]->company
RETURN colleague.name AS name,
count(topic) AS score,
collect(topic.label) AS interests
ORDER BY score DESC
#neo4j

41. Construct
Results
START me=node:users(name = "sarah")
MATCH me-[:INTERESTED_IN]->topic,
me-[:WORKS_FOR]->company,
colleague-[:INTERESTED_IN]->topic,
colleague-[:WORKS_FOR]->company
RETURN colleague.name AS name,
count(topic) AS score,
collect(topic.label) AS interests
ORDER BY score DESC
#neo4j

42. Case
Studies
#neo4j

43. Network
Impact
Analysis
• Which
parts
of
network
does
a
customer
depend
on?
• Who
will
be
affected
if
we
replace
a
network
element?
#neo4j

44. Asset
Management
&
Access
Control
• Which
assets
can
an
admin
control?
• Who
can
change
my
subscrip0on?
#neo4j

45. Logis0cs
• What’s
the
quickest
delivery
route
for
this
parcel?
#neo4j

46. Social
Network
&
Recommenda0ons
• Which
assets
can
I
access?
• Who
shares
my
interests?
#neo4j

47. Thank
You
@iansrobinson
ian.robinson@neotechnology.com
#neo4j

48. Graph
Databases
na0ve
Graph
Processing
FlockDB AllegroGraph
non-­‐na0ve
Graph
Storage
na0ve
#neo4j

49. RDF
Triples
#neo4j

50. Hypergraph
#neo4j