Hasura Introduces First DDN with Instant Auto-scaling, Accelerated Testing, and Sub-second API Iterations

Data API and GraphQL leader Hasura announced from the fourth annual HasuraCon – a number of new innovations that transform how data producers deliver data to data consumers, and make creating, running, and scaling data APIs even faster and easier for application developers. They include the Hasura Data Delivery Network (DDN), Hasura Schema Registry, Hasura Native Data Connector for MongoDB, Native Queries and Logical Models, Open Domain Data Specification (Open DDS) and Native Data Connector (NDC) SDK.

“Data and APIs are the backbone of the modern enterprise,” said Paul Nashawaty, principal analyst at TechTarget Enterprise Strategy Group. “APIs are the predominant mechanism for delivering data from the producers to the edge consumers in a fast, secure, and reliable way. And yet the task of writing, running, scaling, and maintaining APIs on data remains slow, tedious, and a major productivity and time sink for developers.The announcements made by Hasura today make GraphQL and REST APIs easier to author and operate, enhancing the developer experience and bringing the power of APIs to more data.”

Hasura Data Delivery Network (DDN) enables unparalleled API performance at global scale

Hasura DDN is an edge network that enables developers to run low-latency/high-performance data APIs at global scale, with no additional effort and no additional fees. All projects deployed on Hasura Cloud will be automatically deployed to an edge network of 100+ global regions, with 99.99% uptime guarantees, delivering the best experience to the end consumer. It’s like a CDN for real-time, streaming and analytical data that automatically routes and executes client requests on the Hasura instance closest to the client, minimizing the round-trip time (RTT) and latency.

Hasura DDN integrates seamlessly with distributed databases (e.g. CockroachDB, Amazon Aurora, Yugabyte, etc.), automatically routing data queries from the user to the nearest database replica/shard. This minimizes the network latency between not just the client and Hasura, but also Hasura and the underlying data source.

Hasura DDN is made possible by a major architecture change of the Hasura engine that reduced its cold start time to under 1 millisecond. As a result, the Hasura runtime can be instantiated on the edge region when the API is invoked, enabling instant auto-scaling to handle any spike in traffic – globally. DDN is a much more cost-efficient way to get value-based API pricing, instead of infrastructure-based pricing, than alternative approaches like always-on or warmed-up instances.

AiThority: How to Get Started with Prompt Engineering in Generative AI Projects

DDN enables sub-second API iteration and rapid testing

Hasura DDN allows developers to iterate on their APIs – to go from code to global-scale production (including building, validation, and testing) – in less than one second. This rapid CI/CD is true regardless of the number of data models connected, across all changes and scenarios.

Building APIs traditionally requires build/validate/test on the code. This is difficult to do at scale, and previously took anywhere from minutes to hours depending on the type of change and the complexity of the metadata. Hasura’s disruptive innovations improve the speed and predictability of changes, enabling developers to rapidly test multiple deploys in a day – or even an hour. Metadata changes will now create builds that can be independently tested and applied to projects, enabling superior API version control and more seamless collaboration between developers.

These changes will go live as a part of the Hasura DDN launch. Sign up here (link) to be notified and/or be a part of the early access group.

Hasura Schema Registry streamlines governance and evolution of federated GraphQL

Hasura Federation is one of the most efficient ways to compose multiple domain APIs into a single, unified API, aka the supergraph. It makes it simple for developers to specify relationships on models across services (GraphQL and REST) and data stores. Building on these capabilities, the new Hasura Schema Registry makes managing, governing, and collaborating on a federated API even simpler. Hasura Schema Registry allows developers across multiple teams to easily control and audit GraphQL schema changes across diverse data sources, which helps them more reliably and confidently push changes in mission-critical production apps.

Instant GraphQL on NoSQL, starting with MongoDB

The launch of Hasura Data Connectors for Snowflake, MySQL, MariaDB, and Oracle in the last six months have rapidly moved forward Hasura’s mission to bring instant GraphQL/API to all data.Today’s launch of the Hasura MongoDB connector brings the Hasura magic to a whole new data store category – NoSQL. As with all previous data connectors, users can now point Hasura to their MongoDB instances and auto-generate a fully-featured GraphQL API from their collections and documents – in minutes. For any developer looking to build modern GraphQL-powered apps on MongoDB, this connector compresses the backend development effort by orders of magnitude and accelerates the time to market.

With this connector, Hasura now supports the top five databases as ranked by DB-Engines. Bringing API automation to NoSQL stores required Hasura to rethink how the Hasura JIT compiler automatically translates an arbitrary GraphQL query into a performant, native MongoDB query. It takes into account pushing down permission rules and projecting fields as the first cut on performance, with an eye to several other areas where it is possible to make fewer requests to optimize end-user performance. Connectors for other NoSQL stores, including Cassandra/Datastax and Elasticsearch, will be available soon, as will support for existing NoSQL capabilities including PostgreSQL JSONB columns.

Native Queries and Logical Models boost database interactivity

Native Queries and Logical Models allow developers to seamlessly incorporate the full capacity of a database’s query language, such as SQL, into the Hasura auto-generated API. It offers users an enriched and extended range of query possibilities and operations, enhancing the developer experience and broadening the capabilities of the Hasura-generated API.

Read More: ChatGPT Won’t Replace Your Marketing Job, But it’s Critical to Leverage for Success

Native Queries and Logical Models come with an array of compelling benefits:

• Enhanced Query Expressivity: By enabling users to employ their database’s full query language capabilities, Native Queries opens up a broader set of operations that can be performed, bringing the expressivity of SQL into Hasura auto-generated API.
• Lower Privilege Requirements: Native Queries does away with the need for DDL privileges to build custom mapping, making it far more accessible for users in different settings, particularly within enterprises.
• Expanded Function Return Types: This feature is designed to support a more comprehensive range of function return types. It allows Hasura to handle functions that do not necessarily return a table, thus accommodating a more diverse set of databases and function types.

Hasura Cloud now available on Microsoft Azure

Hasura Cloud is now available on Microsoft Azure, in addition to its existing availability on AWS and GCP. This newly introduced capability will enhance and expedite API development within the Azure ecosystem leveraging Azure data sources such as Azure SQL. Developers can now build low-latency, high-performance and secure applications by co-hosting their APIs and databases within the same Azure regions.

Open sourcing the Hasura approach to model-driven API design and data connectivity

Hasura introduced two major open source innovations last year: GraphQL Data Specification (GDS), which enables API developers to create enterprise-grade APIs, and the GraphQL Data Connector (GDC) SDK, which enables developers to author custom data agents. Both of these open source projects serve as foundations for true community collaboration on the Hasura platform.

We are excited to announce major updates to these foundational blocks and integrate them more tightly into the Hasura platform.

• Open Data Domain Specification (Open DDS): Previously known as GDS, this specification has seen a major revision and been entirely formalized, making it ready to be used as a reference implementation. With this change, Hasura has implemented a domain model-driven approach to API design and development. The new Hasura metadata is modeled on the Open DDS, allowing developers to now confidently author high performant, secure, and standardized APIs in a declarative manner.
• Native Data Connector (NDC): Previously known as the GDC, we have reimagined the Hasura NDC to support enterprise use cases and allow developers to build high-quality integrations for Hasura. With the new NDC SDK, we have made it simpler for anyone to create an NDC agent for a new data source. Creators can now leverage state in NDC to build advanced features such as elastic connection pooling into their agents. The spec is now open, giving creators better support for creating non-database agents such as those for REST-based services and leveraging new features, such as parametrized models and commands.

“This is the largest and most significant set of innovations that we have created to date in our journey to make data APIs available and beneficial to all developers,” said Tanmai Gopal, co-founder and CEO of Hasura. “Hasura DDN introduces a number of industry firsts, and does more to reduce the time needed for data to make it from provider to consumer than any other Hasura capability. We are extremely proud of what we have accomplished since the last HasuraCon, and look forward to showing the community what else we have in store.”

Latest Insights: Is Customer Experience Strategy Making or Breaking Your ‘Shopping Festival’ Sales?

[To share your insights with us, please write to sghosh@martechseries.com]