Hadoop Distribution (Hortonworks) - Scaling and Performance

Agenda

1. Technical overview of Hortonworks Framework

2. Vendor use cases of large scale Hadoop deployments

3. Key factors impacting deployment

• Storage and Bandwidth vs Latency
• block size, replication factor
• map, copy/shuffle, and reduce phase tuning parameters
• compression, etc.,

4. Horizontal scaling vs Data center foot print