Microsoft Fabric Updates Blog

Announcing the General availability of time travel and 30 days of data retention in Fabric Warehouse

In the rapidly evolving world of generative artificial intelligence, historical data plays a significant role in influencing the decision-making process and shaping organizational strategies. Data retention within data warehouse refers to the practice of preserving and managing previous iterations of the data encompassing any inserts, updates or deletes made to the warehouse for a specified period. As a quick response to the valuable customer feedback, we are thrilled to extend the data retention period to 30 calendar days from 7 calendar days and announce the General availability of Time travel at the T-SQL Statement level.

Unlocking the potential of increased retention

Extending the data retention period to 30 calendar days opens new avenues for exploration, by leveraging Fabric warehouse features – Time travel at T-SQL statement level, Time travel with table clone, and restore in place.

  1. Enhanced Analytical depth: Increase in retention helps analyze trends, patterns and discrepancies with greater precision and context. This helps make strategic decisions with a greater level of confidence based on a more holistic understanding of their data landscape.
  2. Improved Predictive Capabilities: Historical data is the backbone of predictive analytics, helping organizations to anticipate future trends and proactively address challenges. Extension of retention period enables organizations to build more enhanced predictive models, foresee market trends and help them stay ahead of the curve.
  3. Regulatory Compliance and Risk Mitigation: In highly regulated industries, extended data retention enables in maintaining compliance with regulations and facilitates establishing the audit trails. This helps to mitigate risks associated with non-compliance.
  4. Fabric warehouse recovery: Extending the data retention to 30 days ensures that in the form of restore points the entire Fabric warehouse history is preserved for that duration, while also expanding the total number of system-generated restore points from 42 to 180 available at any given time. In scenarios such as failed release or data corruption, these restore points can be utilized to restore the Fabric warehouse that corresponds to the most stable state prior to the incident. In addition, users are empowered to create any number of user-defined restore points which are retained for 30 calendar days.

In the data-centric world, data retention is pivotal in modern data management enabling organizations to utilize historical data effectively for analysis, reporting, compliance, and strategic decision-making. By the utilization of data retention, businesses can extract valuable insights from the past to drive success both now and in the future. In today’s rapidly evolving digital landscape, adaptability is crucial. Our decision to extend the data retention period from 7 to 30 calendar days underscores our commitment to agility and customer-centricity. Come embrace the data retention of 30 calendar days to obtain deeper insights and stay ahead of the curve.

Related blog posts

Announcing the General availability of time travel and 30 days of data retention in Fabric Warehouse

September 26, 2024 by Jovan Popovic

We’ve improved the JSON support in Fabric Datawarehouse (Fabric DW) and have added the following features in Fabric DW: The JSON_PATH_EXIST function checks if there is a value on the given path in JSON text. The JSON_OBJECT and JSON_ARRAY functions enable you to more easily create JSON objects or arrays based on a set of … Continue reading “Announcing improved JSON support in Fabric DW”

September 26, 2024 by Jovan Popovic

Traditionally, warehouses are designed for smaller data types (numbers, dates, smaller strings) that are suitable for efficient analytics. Currently, Fabric DW has a limit that allows you to store string or binary data up to 8KB per cell. Increasing this limit has been one of the top requests for Fabric DW. We are excited to … Continue reading “Working with large data types in Fabric Warehouse”