Chris Wensel | Hadoop Summit 2014
summary
Concurrent announces $10M funding and details how Cascading provides a Java API that abstracts MapReduce Tez and Spark, letting teams write business logic once and swap execution engines. Scalding and Cascal expose higher‑level DSLs, and Driven/ Cascading.doio add real‑time visibility into running jobs so engineers can spot distributional anomalies, trace slow counters to code paths, and decide whether to change properties, algorithms, or data rather than rebuild pipelines. Open source, Apache licensed, production focused.
AI-written summary · about
chapters (24)
- 0:00Introduction and sponsors
- 0:48Company funding and vision update
- 1:52Cascading technology explained
- 3:56Mama Bear Papa Bear analogy
- 4:58Supporting multiple frameworks
- 6:23LLVM analogy for Cascading
- 6:44Early adopters and use cases
- 8:09Product Driven for real-time insights
- 9:30Open source and ecosystem access
- 10:54Target developer persona
- 11:56Dev analyst and team roles
- 13:21Hadoop Summit trends since 2011
- 15:07Worse is better approach
- 16:09Data visualization role
- 17:10Talk reactions and feedback
- 18:14Breaking through B round growth
- 18:55Concurrent company vision
- 20:38Business process management value
- 21:42CFO focus and business outcomes
- 22:25Exciting tech trends today
- 23:53Reporting and debugging at scale
- 25:58DevOps complexity and insights
- 26:41Underlying Big Data challenges
- 27:02Roadmap and future plans
