Table of Contents

Open Schema and No Vendor Lock-In for Security Data

What is vendor lock-in for security data?

Vendor lock-in for security data happens when your logs, detections and searches are tied so closely to one product that leaving it means losing history, rewriting content or paying heavily to move. It usually builds up quietly: proprietary storage formats, a vendor-specific query language, detections written against vendor field names and pricing that makes exporting data expensive.

The cost shows up at renewal time. If switching would mean abandoning two years of searchable history and rewriting hundreds of detections, the vendor knows you are unlikely to leave.

What is an open schema?

An open schema is a publicly defined, vendor-neutral structure for security data: agreed names and types for fields like source IP, user and event time. When every source is mapped to the same open schema, any tool that understands the schema can read the data.

The most widely adopted open schema for security events is the Open Cybersecurity Schema Framework (OCSF), a Linux Foundation project launched in 2022 by AWS, Splunk and other security vendors. Anomali explains it in OCSF, explained.

The three layers of lock-in, and how to avoid each

  • Schema. How lock-in happens: Data and detections use vendor-specific field names. How to avoid it: Normalize to an open schema such as OCSF at ingest.
  • Storage. How lock-in happens: Data sits in proprietary formats inside the vendor's platform. How to avoid it: Store data in open formats such as Parquet with open table formats such as Iceberg, ideally in storage you control.
  • Content. How lock-in happens: Searches and detections use a proprietary query language. How to avoid it: Write detections against the open schema; keep content in version control; prefer portable formats where possible.

Does avoiding lock-in mean giving up a unified platform?

No. Avoiding lock-in is about where your data lives and what shape it is in, not about using many tools. A unified platform that normalizes to an open schema and can work with data in your own cloud storage, or alongside platforms like Databricks and Snowflake, gives you integration without trapping the data. The test is simple: if you stopped using the platform tomorrow, could you still read and search your data?

Questions to ask any security data vendor

  • What schema is my data stored in, and is it open?
  • Can I store data in my own cloud account?
  • Can I export all of my data, in an open format, at no extra charge?
  • Can other tools query the data directly?
  • Are detections written in a portable format, or in a proprietary language?
  • What happens to my historical data if I leave?

Frequently asked questions

Does using OCSF prevent all vendor lock-in?

It removes schema lock-in, which is one of the hardest kinds to unwind. Storage formats, data ownership and detection content also matter, so check all three.

Is open schema data slower to search?

No. Schema choice does not determine search speed. Storage layout, file format and the query engine matter far more.

Can I adopt an open schema without replacing my SIEM?

Yes. Many teams normalize data to OCSF in the pipeline and send it to their existing SIEM, a data lake or both.

Anomali and Open Schema and No Vendor Lock-In for Security Data

Anomali normalizes telemetry to OCSF and is vendor-neutral by design, working with existing SIEMs and data platforms, including Databricks and Snowflake, rather than requiring them to be replaced.

Learn about the Anomali Unified Security Data Lake or request a demo.

Read next

‍