Hasura, the enterprise helping build globally relevant, data-driven applications and APIs, is announcing a series of innovations—including the Hasura Data Delivery Network (DDN), the Hasura Schema Registry, Native Queries and Logical Models, and more—that optimize data consumption as well as API creation and management.
Hasura DDN—an edge network for running low latency, high performance data APIs at global scale—was born from an architectural change within the Hasura engine, where its cold start time was reduced to under a millisecond. Easily integrating into distributed databases, Hasura DDN automatically routes data queries from the user to the nearest database replica or shard, reducing latency between the client, Hasura, and the data source.
Additionally, Hasura DDN employs value-based API pricing, as opposed to infrastructure-based pricing, boosting cost efficiency.
“[Cold start time reduction] allows us to spin up Hasura almost instantly, versus having warm Hasura instances deployed and running on this edge network,” said Asawari Samant, VP of marketing at Hasura. “When the API call comes through, we just route it to the closest Hasura instance…because we were able to reduce the cold start times, we don't have to have these warmed up instances running, we can spin them up right as the API call comes through.”
Furthermore, Hasura DDN provides developers with the ability to iterate on their APIs in less than a second, significantly accelerating CI/CD regardless of data models connected. Lengthy time drains caused by building, validation, and testing of the code during API development is no more; Hasura drastically improves the speed and predictability of changes, where developers can rapidly test multiple deploys in as little as an hour, at scale, according to the company.
“Previously, what used to happen with the older historic architecture, is those [build/test/validate] loops were very variable, depending on how many models like database tables or models you have connected to in the backend,” explained Samant. “Now it's consistent; it doesn't matter how complex your metadata is or how many tables you connect it to in the back-end, you get sub-second loops going from making a change to testing a change, and that means [developers] can build faster, ship faster, and iterate faster.”
The Hasura DDN will soon be available on Hasura Cloud and as a Private DDN for Hasura self-hosted.
Hasura is also announcing the launch of the Hasura Schema Registry, which builds on Hasura Federation, a solution that composes numerous APIs into a single, unified API. Hasura Schema Registry simplifies the process of managing, governing, and collaboration on federated APIs, offering streamlined control and audit of GraphQL schema changes throughout a variety of data sources. This empowers developers to make changes in production apps, without the fear of compliance issues.
“Hasura has the world’s most advanced authorization engine and an intuitive control plane to apply rich authorization rules across data sources owned by different teams in one single place,” said Tanmai Gopal, co-founder and CEO of Hasura. “The Hasura Schema Registry allows developers across multiple teams to easily audit and get proactively alerted on GraphQL schema changes across diverse data sources, which helps them more reliably and confidently push changes in mission-critical production apps.”
Hasura Data Connectors have continued to support databases by compressing backend development efforts, demonstrating its usability in regard to bringing instant GraphQL and API to all data, according to the company. Now, Hasura is introducing the latest connector—a Hasura MongoDB connector—marking its support for the top five databases, as ranked by DB-Engines.
The Hasura MongoDB connector propels the usability of Hasura to a new data store category, NoSQL. This enables users to point Hasura to MongoDB instances for an auto-generated, fully featured GraphQL API sourced from enterprise collections and documents, in mere minutes.
After making its splash into NoSQL, Hasura intends to release connectors for other NoSQL stores—such as Cassandra and Elasticsearch—soon.
“All of the foundational work that has gone into figuring out API data to API for Mongo will allow us to build such connectors for other NoSQL databases in a much faster manner,” said Samant.
Hasura has also made innovations on database query language, introducing Native Queries and Logical Models, which allows developers to leverage the full capacity of a database query language within the Hasura auto-generated API.
Two open source innovations made by Hasura—GraphQL Data Specification (GDS) and the GraphQL Data Connector (GDC) SDK—are undergoing a reimaging, bringing major updates and more tight integration with Hasura’s platform.
Open Data Domain Specification (OpenDDS), previously GDS, has undergone significant modification that implements a domain model-driven approach to API design and development.
Native Data Connector (NDC), previously GDC, now supports enterprise use cases, allowing developers to create high-quality integrations for Hasura.
“Both the specifications [OpenDDs and NDC] have been re-imagined to double down on two key product values—ease of use and enablement of the ecosystem,” said Gopal. “Our aim has been to make the two specifications intuitive for users to author and reason about while opening up the framework for wider consumption and longevity.”
Additionally, Hasura Cloud is now available on Microsoft Azure, enhancing and accelerating API development within the Azure ecosystem.
“Enterprises have so much data, and everybody wants to be data-centric and data-driven…but the reality is that a lot of people who are building on top of data oftentimes don't have the tools, or the existing tools are not great enough,” explained Samant. “With a lot of the announcements we’re making this week, as well as the core foundation of Hasura, we just want to make that data delivery seamless and fast.”
To learn more about Hasura’s announcements, please visit https://hasura.io/.