Hasura, the data API and GraphQL leader, is debuting a series of enhancements designed to support the process of connecting large language models (LLMs) to highly diverse enterprise data. With new features that expand on the Data Delivery Network (DDN), Hasura forwards a metadata-driven, federated data access layer capable of streamlining governance to effectively support AI initiatives.
Being metadata driven is core to Hasura, merging data and API ecosystems to build fast, secure, and composable data APIs, according to the company. By declaratively modeling domain, entities, relationships, and permissions as a semantic graph to be exposed through the API, this metadata-driven approach improves the governance of the data access layer.
Hasura’s latest features build on this capability, allowing enterprises to leverage the power of metadata to cultivate data federation and streamlined governance—which is particularly crucial for enabling AI, according to the company. Four key capabilities shape this extension of Hasura’s DDN: modular, multi-repo metadata; federated, multi team CI/CD; hierarchical, domain-centric access control; and a new schema registry and changelog.
“Despite all the advances in data storage, compute, and governance technologies, streamlined data access, especially across distributed domains, remains an elusive goal,” said Tanmai Gopal, co-founder and CEO of Hasura. “The innovations we are announcing today for Hasura DDN enable organizations of all sizes and skills to reap the benefits of federated data access patterns, powered by the Supergraph architecture. This will accelerate all use cases but will be especially beneficial for the plethora of GenAI use cases that are hindered by data access challenges.”
To help multi-domain teams cooperatively create metadata, which is monolithic for the end user, Hasura’s DDN now features modularized metadata. This modular, multi-repo metadata allows teams to manage and iterate only their team’s metadata within independent repositories, all while adhering to federation best practices. This enables teams to reap the benefits of a metadata-driven API—including that of automation, governance, and standardization—without sacrificing the ability to describe and accommodate their unique domain.
As part of modularized metadata, federated, multi-team CI/CD allows teams to independently iterate on the metadata and publish their changes as domain-level builds in a federated build and release framework. Each build is instantly validated within the Supergraph context, ensuring that any potential conflicts are flagged early. This capability empowers easy, rapid, cross-domain collaboration from within a unified access layer, according to Hasura.
“Everybody can now build on their bits of metadata that describe their domains independently, without having to work inside one repository [and] have shared CI/CD; they can all have independent CI/CD,” explained Gopal. “They can all make changes independently, and our tooling ensures that whatever changes the different teams make, they are composed together coherently.”
Hierarchical, domain-centric access control delivers a decentralized authorization system that eases the pains of access control within complex, multi-domain and multi-team environments. Now, developers can iterate their metadata at the domain level, while only domain admins can apply changes to the production Supergraph. This ensures that updates are controllable without sacrificing developer autonomy, according to Hasura.
Finally, Hasura is introducing a new schema registry and changelog, engineered to streamline schema evolution and management so that developers can evolve their API safely and confidently, according to Hasura. Included in this feature is a new schema diff tool, which allows users to identify safe, dangerous, and breaking changes between builds, improving comprehension of downstream impact.
To learn more about Hasura’s latest capabilities, please visit https://hasura.io/.