Bloor Research & DataStax: How graph databases solve previously unsolvable business and technology challenges

Webinar | Bloor Research & DataStax : How graph databases solve previously unsolvable business and technology challenges

December 8th 2016

Philip Howard, Research Director, Bloor Research InternationalJonathan Lacefield, Product Manager, DataStax

1 Welcome & Introductions, Andrea Cross, DataStax

2 Graph Databases: Why & Which? Philip Howard, Bloor Research International

3 Fireside Chat, Philip Howard & Jonathan Lacefield

4 DataStax Enterprise Graph, Jonathan Lacefield, DataStax

5 Q & A Wrap Up, Andrea Cross

2© DataStax, All Rights Reserved.

telling the right storyConfidential © Bloor Research 2016

Graph databases: why and which?

Philip HowardResearch Director

telling the right storyConfidential © Bloor Research 2016

• Do you have a graph problem?• Use cases• Types of graph database

Agenda

telling the right storyConfidential © Bloor Research 2016

What is a graph?

Not this This

telling the right storyConfidential © Bloor Research 2016

Why not relational?

When it’s less about the data and more about the relationships

telling the right storyConfidential © Bloor Research 2016

Problems with complexity

BoM: Trunnion mounted ball valves

telling the right storyConfidential © Bloor Research 2016

=> poor performance/scale

Multiple self joins are often an issue

telling the right storyConfidential © Bloor Research 2016

Use case 1: multiple self joins

A fraud ring: an example of network analysis –

there are lots of other such use cases

telling the right storyConfidential © Bloor Research 2016

Use case 2: MDM and Customer 360º

telling the right storyConfidential © Bloor Research 2016

Use case 3: recommendations

Products, dates, hotels, movies.

telling the right storyConfidential © Bloor Research 2016

Use case 4: research

telling the right storyConfidential © Bloor Research 2016

Use case 5: PageRank

Many other graph algorithms

telling the right storyConfidential © Bloor Research 2016

Use case 6: text analysis

telling the right storyConfidential © Bloor Research 2016

Use case 7: data unification

telling the right storyConfidential © Bloor Research 2016

• Two fundamental types:• Property graphs• RDF (resource description framework) Graphs

• Distinction is often not clear-cut

• Native versus non-native databases• Implementation specific

• Multi-model databases• Implementation specific

Types of graph database

telling the right storyConfidential © Bloor Research 2016

• Text processing/semantics• Search engines, sometimes

ontologies• Sometimes text mining• Reasoning functions (OWL)

• Typically query the graph• Oblivious to data (un)structure• SPARQL query language• Tend to be used for complex analytics

(but all graph databases support analytics)

RDF databases

telling the right storyConfidential © Bloor Research 2016

• Typically as replacements for rdbms where relational technology cannot perform

• Good for MDM, Recommendations, Network analytics, Customer 360º…

• Starting point is structured data but unstructured also

• Traverse the graph rather than query it• Use Gremlin, Apache Tinkerpop and/or

proprietary versions of (extended) SQL

Property graphs

telling the right storyConfidential © Bloor Research 2016

• Two specialised variants:• Graph “compute” engines for complex

analytics and cognitive computing

• Unification graph databases

• Specialised niches• E.g. Cybersecurity

Other options

telling the right storyConfidential © Bloor Research 2016

Further resources

See: www.bloorresearch.com/technology/graph-databases/

Fireside Chat

Jonathan Lacefield, DataStax

Philip Howard, Bloor Research

DataStax Enterprise Graph

Founded in April 2010

500+

Santa Clara, San Francisco, Austin, London, Paris, Berlin, Tokyo, Sydney

400+Employees Customers

30+Percent

Overview

Ranked #1 in multiple operational database categories

24

© DataStax, All Rights Reserved.

DataStax Enterprise (DSE)The database for cloud applications

• Make sense of complex, seemingly unconnected data

• Scale-out graph database• Built on Apache TinkerPopTM

• Build quickly, with a consistent skill set• Graph support within DSE• Engineered for Apache CassandraTM

• Part of DSE’s multi-model platform

• Achieve instant insight• Real-time search and analytic queries

© DataStax, All Rights Reserved.

DSE Graph – Scale Out Power for Graph

© DataStax, All Rights Reserved.

Expero uses DSE Graph to manage role-based asset access to improve efficiency and compliance.

Q&A

Contacts & Resources● Download DSE Graph http://www.datastax.com/products/datastax-enterprise-graph● Read the Documentationhttp://docs.datastax.com/en/latest-dse/datastax_enterprise/graph/graphTOC.html● Take DataStax Academy TrainingDS330 – DataStax Graph https://academy.datastax.com/courses/ds330-datastax-enterprise-graph● Join the DataStax Professionals Community Conversationhttps://academy.datastax.com/slack

linkedin.com/company/datastaxlinkedin.com/in/jlacefieldlinkedin.com/in/philip-howard-08a763

@BloorResearch

@jlacefie

@DataStax© 2015 DataStax, All Rights Reserved. 28

THANK YOU!

© 2015 DataStax, All Rights Reserved. 29