From Concept to Reality: Solving Enterprise Challenges
summary
Overview of a deployed Erlang-based enterprise data platform that actually uses Riak, Redis, Spark and shows concrete numbers: a recommended five-node production cluster, Dev on three nodes, Spark adoption at ~9% of clients, and hot data often 5–7% of total. Explains design decisions for availability, low latency, multi-datacenter replication, caching patterns, testing-first development, performance and staging workflows, and what is automated versus what teams must build and operate.
AI-written summary · about
chapters (16)
- 0:00Introduction
- 1:26Agenda
- 4:50Why Big Data
- 7:26Evolution of NoSQL
- 9:14Why Erlan
- 10:45Data volumes
- 17:39What does Erlan plug into
- 19:43Redis vs Memcache
- 23:11Spark Analytics
- 34:31Installing Erlan
- 35:58Clients for Connections
- 38:04Use Cases
- 39:44Testing Staging
- 42:46Testing in Performance
- 44:26Problems
- 46:45Automation
related talks
45mConcurrency + Distribution = Scalability + Availability (...)
53mSpark: from interactivity to production and back
32mPromoting a data driven culture in a world of microservices
24mApps, Stacks, and Frameworks
23mDesigning Robust and Scalable Distributed Applications
59mMastering the System Design Methodology
33mExponential Growth Without Sacrificing Platform Speed
37mDeep Dive - Architecting for Million Operations Per Second!