Yandex.Cloud: The Hard Choice Between Time-to-Market and Product Manageability
summary
Yandex.Cloud's engineering choices for building a public cloud platform: which existing Yandex components were reused (for example, their internal object storage and a proprietary NewSQL database) and which parts were built from scratch to achieve a truly scalable, high-load service. The talk also covers how the team was organized to operate and evolve such a complex cloud product.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 1:52Clouds in Russia feasibility
- 4:42Building the team
- 8:47Cloud architecture principles
- 13:17Technology choices
- 17:40Ecosystem and services
- 20:19Implementation challenges
- 23:53Q&A
related talks
52mWe Cool with Air: Yandex's Data Center — What Is It Like?
59mThe Road of a Monolith to the Cloud and Back
41mClouds of Cocaine
01h 10mHow Yandex Built a Superapp
38mScaling an application without microfrontends
41mPerformer Product Platform in Yandex Go
27mScalable by Design: How to Build an Elastic Cloud Infrastructure...
26mHow we built from scratch a product analyzing terabytes of data
