Alexander Zevaykin, PhD in Technical Science, Technical Leader at Yandex Infrastructure. Working at Yandex, particularly in the development of YDB databases, presents many technical challenges that are fascinating to solve. One such challenge is implementing vector search on large volumes of data.
技术科学博士，在 Yandex 基础设施担任技术领导。在 Yandex 工作，特别是在 YDB 数据库的开发中，面临着许多有趣的技术挑战，其中一个便是在大量数据上实现向量搜索。
YDB is a versatile open-source Distributed SQL Database that combines high availability and scalability with strong consistency and ACID transactions. It provides services for ML-products and goes ahead of vector search.
Outline：
The database is used for industrial operations within Yandex. Among its clients, Yandex Market, Yandex Alice, and Yandex Taxi. They are some of the largest and heaviest AI-based applications.
The database provides true elastic scalability, capable of scaling up or down by several orders of magnitude.
At the same time, the database is fault-tolerant. It has been designed to work across three availability zones, ensuring that the database continues to function even if one of the zones becomes unavailable. The database automatically recovers after a disk failure, server failure, or datacenter failure, with minimal latency disruptions to the applications.
Currently, work is underway to implement accurate and approximate nearest neighbor search for machine learning purposes.
Takeaways：
京公网安备 11010502039052号 | 产品资质
微信咨询
电话咨询
微信联系我们
