+ All Categories
Home > Data & Analytics > Optimising eCommerce with Machine Learning & Game Theory — Cassandra, Elasticsearch and Spark

Optimising eCommerce with Machine Learning & Game Theory — Cassandra, Elasticsearch and Spark

Date post: 16-Apr-2017
Category:
Upload: pca-predict-formerly-postcode-anywhere
View: 1,190 times
Download: 1 times
Share this document with a friend
17
Optimising Experiences Machine learning, Cassandra, Elasticsearch and Spark Joe Chittenden- Veal Jamie Turner
Transcript
Page 1: Optimising eCommerce with Machine Learning & Game Theory — Cassandra, Elasticsearch and Spark

Optimising ExperiencesMachine learning, Cassandra,

Elasticsearch and Spark

Joe Chittenden-VealJamie Turner

Page 2: Optimising eCommerce with Machine Learning & Game Theory — Cassandra, Elasticsearch and Spark

You won’t have heard of us but you will have used us!

5.5bn, 1500tps, 15m

Page 3: Optimising eCommerce with Machine Learning & Game Theory — Cassandra, Elasticsearch and Spark

ExperienceUltimate differentiator

Hard to scaleHard to maintain

Expensive

Page 4: Optimising eCommerce with Machine Learning & Game Theory — Cassandra, Elasticsearch and Spark

Tin v SkinCost

CapacityConsistency

ContextCoverageCompassion

Page 5: Optimising eCommerce with Machine Learning & Game Theory — Cassandra, Elasticsearch and Spark

TriggarInternal projectTraditional stack

Small dataExternal potential

Page 6: Optimising eCommerce with Machine Learning & Game Theory — Cassandra, Elasticsearch and Spark

Sensors

Page 7: Optimising eCommerce with Machine Learning & Game Theory — Cassandra, Elasticsearch and Spark

Games

Page 8: Optimising eCommerce with Machine Learning & Game Theory — Cassandra, Elasticsearch and Spark

Interventions

Page 9: Optimising eCommerce with Machine Learning & Game Theory — Cassandra, Elasticsearch and Spark

Problem

Volume Velocity Variety

Page 10: Optimising eCommerce with Machine Learning & Game Theory — Cassandra, Elasticsearch and Spark

Options

CouchDB, Riak, Redis, Hbase, CouchBase, Neo4j, Dynamo, XAP, Aerospike, BigTable,

Keyspace, LevelDB, Accumulo…

MySQLMongoDBCassandra

Page 11: Optimising eCommerce with Machine Learning & Game Theory — Cassandra, Elasticsearch and Spark

Research.NET friendly?Test Test TestAsk Ask Ask

Page 12: Optimising eCommerce with Machine Learning & Game Theory — Cassandra, Elasticsearch and Spark

We help organisations become more data driven through data science and the adoption of new generation big data

technologies rapidly and at low risk.

Discover, Develop, Deliver, Train, Support

Page 13: Optimising eCommerce with Machine Learning & Game Theory — Cassandra, Elasticsearch and Spark

PotentialWhy build the

games manually when we can use

ML?

Why not blend Cassandra’s speed with Elasticsearch?

Why not use Spark and Spark

Streaming over Hadoop?

Page 14: Optimising eCommerce with Machine Learning & Game Theory — Cassandra, Elasticsearch and Spark

SolutionCassandra

ElasticsearchSpark.NET

Page 15: Optimising eCommerce with Machine Learning & Game Theory — Cassandra, Elasticsearch and Spark
Page 16: Optimising eCommerce with Machine Learning & Game Theory — Cassandra, Elasticsearch and Spark

Final thoughtsWe’ve been spoilt

Real engineering choicesBig impact

Page 17: Optimising eCommerce with Machine Learning & Game Theory — Cassandra, Elasticsearch and Spark

Thanks for listening!


Recommended