Introducing
Your new presentation assistant.
Refine, enhance, and tailor your content, source relevant images, and edit visuals quicker than ever before.
Trending searches
Importing Data into Neo4j
2
Scenario 1
Small, quick and dirty import of a couple hundred / thousand nodes and relationships
Using SPREADSHEETS and cypher statements to do so
Importing Data
into
Scenario 2
Mid-sized import of up to a couple hundred thousand nodes and relationships
Using Cypher's LOAD CSV capability to do so
Scenario 3
10000000s
1000000s
Neo4j
Dataset Size
DOES matter
1000s
Existing Service
Web Service, REST API, .NET, ...
Existing File
.CSV, GraphML, Gephi, Geoff, Spreadsheet, ...
Large import of millions of nodes and relationships
Using the Batch Import capability to do so
Existing database
Relational, NoSQL, Mainframe, ...
Dataset
Format
Graph data model is richer
Test data!
Source
Data
Things to keep in mind!
you don’t just dump data
you create a (complex) structure
you want to fill the right model
no 1:1 copy of your existing database-model
Your Specific Import Requirement
YOUR IMPORT SCENARIO
YOUR DATA MODEL
YOUR
DATABASE
Initial Bulk Load
Incremental Load
Initial Bulk + Incremen-tal Load
Your Unique
Import Process
Why is
Data Import
even a topic
for
Neo4j users
Potential issues
Different Import Mechanisms
transactional - live updates
batch - high volume initial load
Cypher - all APIs
Java - embedded database
“expensive” index reads
duplicate checking
huge transaction sizes
high volumes via Cypher
transactional Mode
Spread- sheet
YOUR IMPORT TOOLSET
YOUR
PROCESS
Neo4j
Shell
Cypher's
Load CSV
DEMO TIME
Custom
Software
Batch- Importer
Tweak the tooling
Map Import to model to tooling
Java Heap Size
Caching config
3rd Party
Tools
Memory mapping
of store files
http://www.wired.com/2013/07/mapping-stereotypes/