SberAuto uses SberCloud.Advanced for stable service operation and business scaling

How the company has stabilized the operation of the service in the context of many-fold business growth.

Logo
Arrow
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
SberCloud.Advanced
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari
Christofari

About the company

SberAuto, part of the Sber ecosystem, is a convenient service for selecting and buying a car online. As part of the service, a personal assistant will pick up the car, credit, and insurance. The car will be delivered to the buyer all over Russia or to pickup centers in major cities.

Project goal

The SberAuto technology platform was originally built on cloud-native principles as a greenfield project1 as part of the classified industry for the Sber ecosystem2. The product team focused on a rapid pilot launch with flexible requirements for infrastructure reliability and SLA. In this regard, the services of several Russian and foreign cloud providers were used as a sandbox platform3 to accelerate development. Later, the team set clear goals:

  1. Achieve higher development speed through the use of cloud solutions;
  2. Exchange expertise on the administration of containerization systems;
  3. Achieve infrastructure reliability without significant costs;
  4. Scale processes with SLA compliance.

Why SberCloud?

In the early stages of service development, requirements were flexible, and the SberAuto team tested the services of different vendors. Bearing in mind the scaling of the business, the main criteria for choosing a cloud provider were:

  • maturity of platform services;
  • infrastructure reliability;
  • compliance with SLA.
Based on the results of the pilot launch of the Cloud Container Engine service, the decision was made to migrate to the SberCloud.Advanced cloud platform. It is recognized by the customer as the most mature platform solution on the market.

Solution

Technically, the migration to the platform took a few days. We were migrating in the context of growing traffic, and we had to do this in stages:

  1. Move the object storage;
  2. Transfer data and applications;
  3. Complete the migration.
As a result, we connected SberCloud cloud services smoothly and without failures.

SberAuto uses:

  • SberCloud Managed Kubernetes for automatic deployment, scaling, and management of applications;
  • Distributed Message Service for Kafka and Distributed Message Service for RabbitMQ — software message brokers to increase throughput;
  • S3 object storage for storing information of any type and volume;
  • Elastic Cloud Server for creating virtual machines for any use cases.

Result

The SberAuto team primarily wanted to ensure the stability of the service and infrastructure in the context of scaling and multiple growth of the business. This goal has been successfully achieved.

Further plans

SberAuto continues to dive into machine learning. In particular, the team plans to take advantage of the ML Space platform, using the power of the Christofari supercomputer in its work.

"The operation of cloud services in normal mode allows more time to devote to strategic issues of SberAuto development. We are constantly working to improve the service, adapt it to market requirements, offer new products and solutions to customers. Close collaboration between our team and SberCloud specialists will help us implement our plans," comments Yury Buylov, SberAuto CTO.

"Working together with colleagues in the ecosystem enriches our experience and opens up opportunities for experimentation. SberAuto uses the services of the SberCloud.Advanced platform and plans to connect to the ML Space cloud platform to create and deploy machine learning models. Judging by the evaluation of the work done, this approach will make it possible for the customer service to develop even faster," said Denis Sokolov, Head of the area for collaboration with subsidiary and dependent organizations of PJSC SBER ecosystem, SberCloud.


1Greenfield project — a project created from scratch 2The Sber ecosystem includes services with various advertisements for the purchase or sale of goods or services 3Sandbox platform — a secure testing environment
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud
SberCloud