Backend Architecture of sputnik.ru Maps
summary
A talk about Sputnik.ru’s backend that generates raster map tiles from OpenStreetMap data using a horizontally scalable distributed rendering cluster. It covers the cluster architecture, reasons for choosing the Go programming language, their testing practices, and their contributions to open-source projects.
AI-written summary · about
chapters (30)
- 0:00Introduction
- 1:55OpenStreetMap architecture
- 3:00Web map technologies
- 5:48Map style creation
- 7:55Classic approach problems
- 9:25Solution: Gopnik system
- 10:32Gopnik workflow
- 11:35Tile generation
- 14:32Caching and updates
- 15:39Cache fill utility
- 16:42Gopnik internals
- 17:51System reliability
- 20:00Go language choice
- 22:29Go advantages
- 24:56Testing and DevOps
- 26:44End-to-end testing
- 27:47Network issue emulation
- 28:51Load testing
- 29:54Future plans
- 31:39Open-sourcing Gopnik
- 32:02Q&A: Gopnik architecture
- 33:44Q&A: tile updates
- 35:10Q&A: vector maps
- 36:31Q&A: style and data
- 37:58Q&A: Retina displays
- 38:40Q&A: Leaflet and nationality
- 39:21Q&A: search engine
- 40:05Q&A: address search
- 41:10Q&A: rendering tests
- 41:51Q&A: map positioning
related talks
46mA Look Behind the Scenes of Yandex.Maps: From Raw Satellite Data to Clear Images
05mAlex Kudriashova Interview
14mMosaic: A Framework for Geospatial Analytics at Scale
23mData Journey with Golang, GraphQL and Microservices
27mScalable by Design: How to Build an Elastic Cloud Infrastructure...
28mMicroservices – Boyan Dimitrov
34mBuilding a WebScale Search Engine with a Data Driven Architecture
24mSKA Observatory: Reaching to the Stars and Beyond