Time series analytics for Big Data and IoT with Kx
summary
Claims that kdb+ (columnar time-series DB) unifies streaming, real-time and historical data and cuts developer and TCO by using in‑memory and on‑disk compression and parallelism. Demonstrates an event engine / tickerplant that logs to durable shards, real‑time DB and streaming query engines written in q language with vector ops and temporal joins. Shows it running on a server with ~0.5 TB RAM and 22 TB NYSE data, handling 538B quotes / 253B trades (~1.2T rows) and answering queries interactively.
AI-written summary · about
chapters (14)
- 0:00Intro
- 1:24Memory
- 2:00Storage
- 2:24Hardware
- 3:15Big Data IoT
- 4:46About Kx
- 5:57What we do
- 6:20Kx
- 6:45Features
- 7:54APIs
- 8:19Programming Language
- 9:11Time series analysis
- 10:09Architecture diagram
- 11:53Demo
related talks
28mTick Tock: What the heck is time-series data?
47mBuilding a Real-Time Analytics Database
29mRealtime Analytics Unleashed: Unlocking the Potential of Data at Scale
29mWide Event Analytics
44mProcess and analyze millions of events per second with open source tools
24mStreaming analytics on edge with Kuiper and KubeEdge
26mLeveraging Data in Motion in a Cloud-first World
28mBest of Stream + Batch: When Streaming Analytics Isn’t