There needs to be a new type of database optimized for serving simple, massive-scale data in real-time. I’d like to propose a scalable engine that’s actually optimized to drive 80% of websites, built on HBase. Given at HBase User Group, August 09
6. Let’s do this real-time
style
• Sites use 1% of the power of RDBMS
• So optimize for reality
• HBase .20 can scan through several M
rows / node quickly
• Aggregates drive almost all data
7.
8. Roadmap
• Aggregate on Regionservers
• Concurrency / node / data size is key
• May need indexing?