Petr Zapletal Interview
summary
Focuses on cost and state at scale: uses the SMACK stack (Spark, Mesos, Akka, Cassandra, Kafka), targets systems with millions of concurrent users, and struggles with state distribution, consistency and classic cache invalidation pain. Claims cloud bills reach tens of thousands of dollars per day if full telemetry is stored, so work prioritizes cost-efficiency and right data retention. Notes functional programming helps correctness but sometimes yields performance and hiring tradeoffs, prompting pragmatic compromises.
AI-written summary · about
chapters (5)
- 0:00Introduction
- 0:50Whats it like right now
- 2:07Problems
- 3:27Functional Programming
- 4:31Favorite Conference
related talks
19mKeynote Panel Discussion: SMACK in the Enterprise
02h 18mDistributed Systems: In the Daily Life of the Data Engineer
57mMulti Tenant NoSQL and NewSQL Cloud data Design Patterns Deep dive
29mEnhancing Flexibility and Productivity with Access Patterns and Storage-Agnostic Abstractions
25mSPARK AND AKKA FOR BIG DATA SYSTEMS, IN PRACTICE
44mThe Art and Benefits of Computing Eventfully
31mFUNCTIONAL EVENT-DRIVEN ARCHITECTURE
20mDataflow Engines for Executing Cloud Applications: a Maslow's Hammer or Natural Outcome?
