James Markarian, SnapLogic | BigDataNYC 2016
summary
Claims that moving integration to a cloud‑first IPaaS cuts operational overhead while keeping data where customers want it; SnapLogic runs its design UI on AWS, offers on‑prem or cloud runtimes, and connects JSON REST SaaS endpoints to Hadoop, Redshift, Vertica and mainframes. Notes surprising practical numbers: typical clusters of 10 nodes with occasional bursts to 100–1000 nodes, and analytics teams finding 50 duplicated datasets in a raw lake. Recommends organizing into data lake zones (raw, purified, bottled) and using integration glue plus metadata to make schema‑on‑read usable.
AI-written summary · about
chapters (10)
- 0:00Start
- 0:1901. James Markarian, SnapLogic, visits #theCUBE!.
- 0:4002. How Data Management is Evolving.
- 2:2103. The Hadoop Equation.
- 4:2604. Cloud Deployment Technologies at SnapLogic.
- 7:2905. The Data Lake: Benefits and Challenges.
- 11:0306. Schema on Read.
- 14:5107. SnapLogic Origins.
- 16:1908. The SnapLogic Integrated Solution.
- 17:5009. What Do We Want to Do with Data?.
