Today’s CIOs and CTOs don’t just need to manage larger volumes of data – they need to generate insight from their existing data. In this case, the relationships between data points matter more than the individual points themselves. In order to leverage data relationships, organizations need a database technology that stores relationship information as a first-class entity. That technology is a graph database.
Attend this webinar to hear about:
1. Why graph technologies are essential for the future of increasingly connected data
2. How enterprises such Walmart, eBay, and UBS are using Neo4j’s native-graph platform for a diverse set of use cases, including security & fraud detection, real-time recommendation engines, master data and many more
3. And how Neo4j on IBM POWER8 can scale your massive graph data with real-time graph processing that’s entirely in-memory.
[2024]Digital Global Overview Report 2024 Meltwater.pdf
Beyond Big Data: Leverage Large-Scale Connections
1. Beyond Big Data: Leverage
Large-Scale Connections
and IBM Power Systems
October 4th, 2017
2. 1. Why graph technologies?
2. How enterprises are using
native-graph
3. Common Neo4j
reference architecture
4. Neo4j on IBM POWER8 to
better leverage your big data
Agenda
Nav Mathur
Sr. Director Global Solutions @ Neo4j
@nav_mathur, in/navmathur
Richard Sheppard
Director of Sales @ Blair Technology Solutions
in/rmjsheppard
Amy Hodler
Sr. Marketing Mgr @ Neo4j
@amyhodler, in/amyhodler
6. On Stage
Behind the Scene
Linear Supply Chain InformationOrganizations Multi-related Knowledge
Business
Processes
Data
Structure
7. ”Graph analysis is possibly the single most effective
competitive differentiator for organizations pursuing
data-driven operations and decisions“
The Impact of Graphs
8. Connected Data is Transforming Industries
Social Graph
Knows
Knows
Knows
Knows
People & Products
Bought
Bought
Viewed
Returned
Bought
People & Content
adidas
Plays
Lives_in
In_sport
Likes
Fan_of
Plays_for
10. Native Property Graph
The Whiteboard Model is
the Physical Model
A unified view for
ultimate agility
• Easily understood
• Easily evolved
• Easy collaboration
between business and IT
11. Property Graph Model Components
Nodes
• Can have name-value properties
• Can have Labels to classify nodes
• Relate nodes by type and direction
• can have name-value properties
name:”Dan”
born: May 29, 1970
twitter:”@dan”
name:”Ann”
born: Dec 5, 1975
Since:
Jan 10, 2011
brand: “Volvo”
model: “V70”
CAR
Married_to
Lives_with
D
rives
PERSON
O
wns
Relationships
PERSON
21. • Neo4j powers the next generation of
applications and analytics
• Prominent use cases are found in areas
like machine learning, personalized
recommendations, fraud detection, data
governance and more.
Neo4j: The #1 Platform for Connected Data
23. Ecosystem
Neo4j Professional Services
300+ partners
47,000 group members
61,000 trained engineers
3.5M downloads
Mindset
“Graph Thinking” is all about
considering connections in
data as important as the data
itself.
Native Graph Platform
Neo4j is an internet-scale,
native graph database which
executes connected workloads
faster than any other database
management system.
Neo4j
24. Digital native companies like Medium, Ebay, and
LinkedIn, as well as companies in transformation
like Walmart, Adidas and Airbus, have all chosen
to adopt Neo4j.
25. Hundreds of successful deployments
— from Fortune 500 companies to exciting startups
Examples of enterprise adoption:
26. Adoption Highlights
Retail
7 out of top 10
retailers in the world
Finance
12 out of 25 top
financial services firms
8 out of top 10
software vendors
Software
(As per 2017)
29. 29
Ebay powers its machine
learning based ‘shopbot’
with Neo4j knowledge
graph
"Feels like talking to a friend"
”
Online Shopping
30. Ebay powers its machine
learning based ‘shopbot’
with Neo4j knowledge
graph
30
Online Shopping
• 3 developers, 8M nodes, 20M relationships
• Needed high-performance traversals to respond to live
customer requests
• Easy to train new algorithms and grow model
• Generating revenue since launch
Solutions and benefits
"Feels like talking to a friend"
”
31. UBS is using Neo4j to manage their complex data
infrastructure of over 400 integration points across 18 data-
domains to improve access for data consumers.
UBS uses Neo4j for
trustworthy data
Financial Services | Master Data Management
• Dramatic improvement to data distribution flow
• Knowledge Base improves Ad-hoc analytics
• Data governance, lineage and trust improved across
entire company
• Better service level from IT to data consumers
Solutions and benefits
32. “Neo4j’s high performance engine provides flexibility of
data representation along with features that go beyond
traditional relational databases.”
” — Sebastian Verheughe, Telenor
Telenor uses Neo4j to
provide complex
self-service
32
Telecom | Identity and Access Management
Telenor uses Neo4j to provide businesses and residential
customers with a self-service portal that brings together
information about corporate structures, subscription
information, price plan and owner/payer/user data, billing
accounts and any discount agreements.
• Shifted authentication from Sybase to Neo4j
• Moved resource graph to Neo4j
• Replaced batch process with real-time login responses
• Mitigated customer retention risks
33. Marriott is using Neo4j to allow hotel managers to
control room rate price optimization across 1.5 million
rooms on a daily basis.
Marriott reinvents
room rate pricing with
Neo4j
Travel Services | Pricing Recommendations
• Created a graph per hotel for 4500 properties in 3 clusters
• Enabled a 1000% increase in volume over 4 years while
cutting infrastructure cost in half.
• "Use Neo4j Support!"
“We couldn’t have done this without Neo4j commercial
support”
” Scott Grimes
Senior Director of Revenue Management, Marriott International
“With Neo4j, we’ve been able to take our average
processing time [for pricing operations] from over
four minutes to about 13 seconds…and reduce our
overall infrastructure cost by about 50%.”
– Scott Grimes, Marriott
35. ı
Neo4j: #1 Database for Connected Data
Neo4j is an enterprise-grade native graph database that enables you to:
• Store and access data and relationships
• Traverse data at any levels of depth in real-time
• Add and connect new data on the fly
Designed, built and tested natively for
graphs from the start to ensure:
• Performance
• ACID Transactions
• Agility
• Developer Productivity
• Hardware Efficiency
36. Q R
Q R
Using Other NoSQL to Join Data
Using Neo4j
Slow queries due to
index lookups &
network hops
Lightning-fast queries
due to replicated in-
memory architecture and
index-free adjacency
Relationship Queries on non-native Graph Architectures
MACHINE 1 MACHINE 2 MACHINE 3
UNIFIED, IN MEMORY MAP
37. Real-Time Query Performance
Relational and
Other NoSQL
Databases
ResponseTime
Connectedness and Size of Data Set
0 to 2 hops
0 to 3 degrees
Few connections
5+ hops
3+ degrees
Thousands of connections
1000x
Advantage
“Minutes to milliseconds”
Neo4j
38. How Fast is Fast?
*6 machines, each with 48 VCPUs, 256 GB disk and 256 GB of RAM; ~10M node, ~100M relationship graph
Workload Non-native graph DB* Neo4j: single thread
Count nodes
Count outgoing rels
Count outgoing rels at depth 2
Count outgoing rels at depth 3
Group nodes by property val
Group rels by type
Count depth 2 knows-likes
Page Rank
201s
202s
276s
511s
212s
198s
324s
2571s
< 1ms
< 1ms
23s
423s
8s
54s
149s
27s
39. Keeping Your Graph Intact Essential for Graph Operations
Atomic Causal Consistency
The graph
transaction
moves together
as one ACID
transaction with
built-in safety
Guarantees Graph Consistency
Graph Writes: Neo4j vs. Non-Native Distributed
40. Non-Native
Graph DB
Keeping Your Graph Intact Essential for Graph Operations
Atomic Causal Consistency Non-Atomic Eventual Consistency
The graph
transaction
moves together
as one ACID
transaction with
built-in safety
Without atomic,
ACID graph
transactions the
view of the graph
& its property
values is
necessarily
inconsistent
Guarantees Graph Consistency Not Good Enough for Graphs
Graph Writes: Neo4j vs. Non-Native Distributed
42. Native Graph Storage
Designed, built, and tested for graphs
Native Graph Query Processing
For real-time, relationship-based apps
Evaluate millions of relationships in a blink
Whiteboard-Friendly Data Modeling
Faster projects compared to RDBMS
Data Integrity and Security
Fully ACID transactions, causal consistency
and enterprise security
Powerful, Expressive Query Language
Improved productivity, with 10x to 100x less
code than SQL
Scalability and High Availability
Architecture provides ideal balance of
performance, availability, scale for graphs
Built-in ETL
Seamless import from other databases
Integration
Fits easily into your IT environment, with
drivers and APIs for popular languages
Neo4j: Built for the Enterprise
47. Money
Transferring
Purchases Bank
Services Relational
database
Data Lake
+ Good for Map Reduce
+ Good for Analytical Workloads
– No holistic view
– Non-operational workloads
– Weeks-to-months processes Develop Patterns
Data Science-team
Merchant
Data
Credit
Score
Data
Other 3rd
Party
Data
48. Neo4j powers
360° view of
transactions in
real-time
Neo4j
Cluster
SENSE
Transaction
stream
RESPOND
Alerts &
notification
LOAD RELEVANT DATA
Relational
database
Data Lake
Visualization UI
Fine Tune Patterns
Develop Patterns
Data Science-team
Known
Threats
Known
Vulnerabilities
Data-set used
to explore
new insights
IBM Identity
& Access
Mgmt.
IBM BigFix
Endpoint
Security
IBM QRadar
Log
Management
Virus
Signatures
49. Business Value
Real-Time Fraud Prevention
• $MM recaptured in savings
• Increased revenue & customer
satisfaction due to decreased false
positives
Graph-Based Analytics
• Increased effectiveness of Data Science
Team
• Months of Analysis now done in minutes
& hours
• Graph-based visualization accelerates
complex analysis of patterns
Typical Customers
Graph-Based Sense & Respond Architecture for SIEM
Financial
Services
Government Telecom
Events & Transactions
50. Your Engines
& Algorithms
Customer
Facing Apps
Native Graph
Database
Customer
Data Sources
Complete Solution
Real-Time
Threat Management
Snapshots:
Modeling & Predictive
VisualizationImpact
Analysis
Recommendations
Anomaly Detection
Decisioning
Variance
Analysis
Rules
Engine
Complex Event
Procedures
APIs & Drivers
Data Ingest
CMDB
Assets Data
Network
Monitoring Data
Application
Monitoring Data
Bus. Process
Monitoring Data
Data
Lake
Threat Vectors
Mitigation Patterns
51. 51
ON
Neo4j On IBM Power8
Real - Time Graph Processing That’s
Entirely In-Memory
Richard Sheppard
Director of Sales @ Blair Technology Solutions
52. 4X
Threads per core*
4X
Mem. Bandwidth*
4X
More cache* @
Lower Latency
These design decisions result in best performance for data centric
workloads like:
Database, NoSQL, Big Data Analytics, OLTP
POWER8
SMT8
x86
Hyperthread
Parallel Processing
POWER8
pipe
Data flow
x86 pipe
POWER8
x86 POWER8 +
OpenPOWER
x86
ON
53. The POWER of an open ecosystem
ON
300Worldwide members of 2,500+
Linux ISVs developing on
POWER
30
Hardware and
technology providers
100,000+
Open source packages100+
Collaborative
innovations under way
54. innovations under way
POWER8 with CAPI enabled acceleration running Neo4j delivers
1.61X the price-performance versus Intel Xeon E5-2650 v4 with
NVMe
IBM Power S822LC
(20-core, 128GB)
HP
DL380 Gen9
(24-core, 128GB)
Server price*
-3-year warranty
$19,123 $16,911
Mixed graph transaction
Workload
(total operations per second)
711 390
1.61X
Price-Performance
1.82X
Performance
per Server
• Based on IBM internal testing of single system and OS image running mixed graph transaction s based on 200 GB data model internal IBM and Neo4j workload. Conducted under laboratory condition, individual result can vary
based on workload size, use of storage subsystems & other conditions. Data as of October 19, 2016
• IBM Power System S822LC; 20 cores (2 x 10c chips) / 160 threads, POWER8; 128 GB memory (16 x 8GB), 1.6 TB CAPI NVMe adapter , Neo4j 3.0.4, Ubuntu 16.04. Competitive stack: HP Proliant DL380 Gen9; 24 cores (2 x 12c chips) /
48 threads; Intel E5-2650 v4; 128 GB memory,(16 x 8GB), 1.6 TB NVMe adapter, Neo4j 3.0.4, Ubuntu 15.10.
* Pricing is based bundled pricing for S822LC with Integrated CAPI Flash card (IBM ordering system) and HP Web price https://h22174.www2.hp.com/SimplifiedConfig/Index
ON
55. Scalability
Only POWER8 can provide up to 56
terabytes of extended memory space
with up to 40 terabyte CAPI flash
architecture.
Performance
TCO
Services
Why Neo4j on Linux On POWER
Reduced downtime, HA database
monitoring / management and data
integration for mission critical
enterprise applications.
Offer ability to use Flash as
extended memory without
compromising real-time capabilities .
Master datasets connecting data inside a
single graph inventory or supply chain
management data for global enterprise
manufactures.
56. 56
➢ Trusted IBM Business Partner for 21 years
➢ Enabling business to utilize data for improved business insights
➢ Certified IBM POWER server (AIX, i, Linux) leader
➢ Contact us to discuss your challenges in adopting graph database to
meet business needs
Richard Sheppard
rsheppard@blairtechnology.com
905-474-4206
57. Next Steps
Register for a brown-bag graph talk with
your team @ https://neo4j.com/brownbag/
Attend GraphConnect use 50% off code
IBMCD50
Thanks!57
Nav Mathur
Sr. Director Global Solutions @ Neo4j
@nav_mathur, in/navmathur
Richard Sheppard
Director of Sales @ Blair Technology
Solutions
Amy Hodler
Sr. Marketing Mgr @ Neo4j
@amyhodler, in/amyhodler