Analyze: Introduction
You have now ingested, structured, and cataloged your data and created all of your accompanying data assets. Now what do you do with your data?
TileDB provides a range of capabilities to perform impactful data analysis:
Slicing data: TileDB’s serverless architecture and the underlying storage format allow for elastic and scalable data access and a completely lock-free multi-reader and multi-writer. Learn about these in more detail in this section.
Multi-region redirection: TileDB Cloud supports automatic region redirection to reduce latency and egress costs. This section explains how it works and how to access a region directly.
Notebooks: Learn how to create Jupyter notebooks, mixing both Markdown and code, to perform analysis on your data. Learn the basics of notebooks, how to launch notebook servers in TileDB, and how to create Jupyter/IPython widgets.
Dashboards: Dashboards are based on notebooks and provide a beautiful and interactive means of analyzing your data in a no-code manner.
Preview Widgets: For specific asset types, TileDB offers a way for you to preview them directly in the web browser.
User-defined functions (UDFs): Run Python and R code on a full-managed node in TileDB Cloud. UDFs can be arbitrary or array-based, and this section provides you with examples of each.
Task graphs: Orchestrate multiple UDFs and scale your code in a cost-optimized manner using task graphs in TileDB’s serverless infrastructure.
Serverless SQL: SQL is the universal standard for interfacing with tabular data. TileDB’s serverless SQL capabilities allow you to interact with your arrays using the same, familiar language with which you speak to tabular databases.
Monitoring: View details about the tasks and task graphs executed by you and your colleagues, so you can monitor costs and optimize your code assets.
Read on to explore the many ways you can analyze your data.