Taming the data demon using HPCC systems
summary
HPCC Systems is an open‑source big‑data engine with a data‑centric language ECL and separate analytic and delivery clusters Thor/Roxie. Real examples: LexisNexis runs 40+ petabytes, a credit‑default job drops from 26 days to six minutes, profiling five million rows in ~4 seconds, and one‑line dedupe or spray operations ingest CSVs into distributed slaves. Data is append‑only with rollups (no updates/deletes), built‑in ML libraries, and a cloud‑native 8.0 option; proves fast, auditable, and production‑ready for fintech workloads.
AI-written summary · about
chapters (8)
- 0:00Start
- 5:04Anatomy of How Hpcc Systems Works
- 22:43Key Value Lookups
- 52:44Hpcc Cheat Sheet
- 55:16Hpcc Can Virtualize within a Single Box or across Multiple Boxes
- 55:30Network Latency
- 1:07:26Join on a Table
- 1:25:08Download the Vm
related talks
28mHerding Data to the Cloud
47mSolving Real & Big (Data) Problems Using Hadoop
18mOur Analytical Data Journey – from Paper to Petabytes
06mEnterprise Big Data Hub & the Federal Government
49mData Platform Architecture, Evolution, and Philosophy
02h 05mBig Data and the Data Lake from the Analyst Perspective
45mModern Data Architecture
37mIs There Really Such a Thing as Turnkey Data Analytics Solution?