YTsaurus

lastmod 2023-03-27created 2023-03-272 edits

Open-source Exabyte-Scale storage & processing system

Features

• An open-source distributed storage & processing big data platform released by Yandex
  • Use cases: batch processing, ad hoc analytics, OLTP, machine learning, metadata storage, ETL pipelines
• Multi-tenant ecosystem
• Reliability and safety: no SPOF. Automated replication between servers. Updates without loss of progress
• Scalability
  • Scales up to 1 million CPU cores and thousands of GPUs
  • Supports more than 10,000 nodes. Automatic server scale up & down
  • Supports various storage media, up to exabytes of data
• Rich features
  • Extensive MapReduce model
  • Distributed ACID transactions
  • Provides various SDKs (C++, Python, Java, Go) and APIs
  • Secure isolation of computing resources and storage
  • User-friendly UI
• CHYT powered by ClickHouse
  • Familiar SQL Dialect and similar features
  • Fast analytical queries
  • Integrates with popular BI solutions (JDBC/ODBC)
• SPYT powered by Apache Spark
  • A tool for building ETL processes
  • Supports multiple independent clusters of various sizes
  • Easy migration of existing solutions

See also

• Big Data

Favorite site

• YTsaurus is a distributed storage and processing platform for Big Data
• Github - ytsaurus