How Xignite Delivers Real Time Financial Data at Scale
summary
Xignite moves from Amazon RDS SQL Server to YugabyteDB to serve time series and reference market data, replacing sharded RDS pain with hash-based vnodes, automatic rebalancing and a single endpoint. Production handles 140 billion data points, about 30,000 requests per second, and bursts of 200k writes/sec for bars, uses a scan‑resistant two‑tier cache to protect hot data, and reports roughly 50% cost savings while maintaining sub‑millisecond read latency and no downtime scaling.
AI-written summary · about
chapters (11)
- 0:00Intro
- 3:24Delivery Methods
- 4:41Shapes of Data
- 7:51Time Series Data Engine Requirements
- 10:16Legacy Time Series Data Engine Implementation
- 12:50Time Series Data Engine with Yugabyte
- 16:19Yugabyte Performance - Write Heavy Use Case
- 16:47Legacy Security Master Data Engine Implementation
- 18:00Security Master Data Engine with Yugabyte
- 19:45Yugabyte Performance - Read Heavy Use Case
- 20:28Xignite's Partnership with Yugabyte
related talks
29mDelivering Always-On Financial Services with
25mHow Sharekhan Scaled Trading Platform with Yugabyte's Distributed SQL
18mBuilding a Scalable, Resilient System for Real-Time Fraud Detection
57mNo Downtime, No Limits
29mEnterprise Scale: How a Distributed Architecture Can Deliver Scale and Resilience
22mYugabyteDB: Discover What's New
25mYugabyteDB Technical Overview
24mScaling to Billions of Internet Observations per Day