August 2023 Summaries

Hasura has announced the general availability of its MongoDB database connector. The Hasura MongoDB database connector enables users to create a GraphQL API on MongoDB and efficiently query between collections, among other features. It also allows for joining a MongoDB connection to another data source. The new offering aims to simplify the complex task of managing and querying data while providing an efficient way to author GraphQL schema over NoSQL (with MongoDB specifically).

Scaling frontend applications and teams can be achieved by factoring the frontend into multiple components owned by independent specialized teams. This reduces coupling between components and ensures clearly specified interfaces between them. Modern technologies like Relay and backend-for-frontend (BFF) can help scale out frontend applications. By using GraphQL, a centralized cache store, and batching network requests, developers can improve performance and reduce redundancy in data fetching and state management. Relay is designed to work with any existing GraphQL API and can be incrementally adopted for an easy migration path.

Hasura introduces Notebooks, a tool designed to facilitate the swift prototyping of cutting-edge GenAI applications. The new feature is a remote Jupyter Notebook with ready-to-run examples and Jupyter Kernel Gateway baked-in to turn your code cells into endpoints. Hasura Notebooks are great if you want to quickly prototype GenAI applications with Hasura or learn GenAI from existing templated projects. The use case discussed in the text is a contextual product search powered by OpenAI, which can be built under 10 minutes using Hasura Notebook.

DuckDB is an open-source, embedded database management system optimized for data analytics and online analytical processing (OLAP). It supports standard SQL queries, aggregations, joins, and other SQL functions on imported data from JSON, Parquet, and CSV files. DuckDB's columnar structure and vectorized data processing make it efficient in handling OLAP workloads. DuckDB is gaining popularity among data scientists and analysts due to its ease of setup and usage, as well as cost savings by running locally. It can mash up data from different sources like CSV, Parquet, and PostgreSQL, making it suitable for various use cases. However, when it comes to team collaboration on large datasets or building real-world dashboards, an API layer becomes necessary. A data API platform like Hasura can bridge the gap by providing built-in declarative authorization, role-based access control (RBAC) rules, and support for various data sources. Additionally, it enables running DuckDB workloads on edge environments like Vercel Edge Functions and Cloudflare Workers with TypeScript. In conclusion, DuckDB's integration with a data API platform like Hasura can provide an efficient solution for OLAP queries, data mashup workloads, and real-world dashboard use cases while addressing authentication, authorization, and other production-ready concerns.

Hasura Cloud introduces an Alerting system to proactively notify users about important events and incidents in their Hasura infrastructure. The alerting system helps detect issues related to remote data sources, external webhooks, performance of database query and webhook invocations, and logic errors. Key benefits include faster incident resolution, improved reliability, performance issue detection, and root cause identification. Alert criteria include abnormal error rate, high latency, data source status, and schema changes. The alert message contains general information about the triggered status, time, and description of the alert rule. Currently, Hasura Alerts support email and Slack notifications. Users can enable or disable the feature, specify alert rules, and add email addresses or integrate a Slack application to receive alert messages in the Project Details -> Alerting tab.

Single Sign-on (SSO) has been introduced for Hasura Console to enhance security and provide flexibility in accessing the tool. This feature is available on Hasura Enterprise Edition and already exists for Hasura Cloud customers via SAML. The integration of existing SSO identity providers with Hasura GraphQL Engine enables authentication, while full support for OAuth 2.0 as a Single Sign-On (SSO) solution is provided without the need for additional tooling or services. Configuration involves setting up the sso-providers argument or HASURA_GRAPHQL_SSO_PROVIDERS environment variable with details such as client ID, display name, scope, authorization URL, request token URL, admin roles, and JWT secret. For other protocols like LDAP or SAML, an OpenID Connect compliant middleware service can be used. Hasura Cloud Enterprise supports SSO integration through the SAML 2.0 protocol.

Microservices are an architectural pattern that has gained popularity in recent years. However, deploying microservices can come with several challenges. This article outlines the top five pitfalls to watch out for when implementing microservices: 1) complex access control of data; 2) complicated distributed transactions; 3) challenging aggregation of data; 4) elusive observability; and 5) increased infrastructure complexity. These issues can make it difficult to maintain and manage microservices, making them less suitable for all problems. It is essential to carefully consider the use of microservices and weigh their benefits against these potential challenges.

Modernizing legacy applications with Hasura significantly enhances observability, leading to improved performance, increased agility, and better user experiences. By transforming a legacy monolith into microservices with Hasura, organizations can overcome many observability challenges, including granular observability, real-time monitoring, centralized data access, dynamic tracing, and automated debugging. Hasura offers an array of default observability options and integrates seamlessly with popular observability platforms like Prometheus, Grafana, and Jaeger. Enhanced database observability is also possible through strategic steps such as enabling observability agents and leveraging query tags. Proactive alerting and alert propagation are crucial for maintaining optimal performance.

At Google Cloud Next 2023, Hasura is excited about five sessions featuring data APIs. These include a guide on choosing between GraphQL, gRPC, AsyncAPI, or REST API; integrating apps and workflows with Google Meet APIs and SDKs; supercharging startups' market entry and monetization with APIs; staying ahead of API security threats with ML detection; and participating in an Integrate with Machine Learning APIs Challenge. Hasura will be at booth #543, offering data API solutions and a chance to win prizes through raffle and demo bookings.

In the banking industry, attribute-based access control (ABAC) is a more flexible approach to managing access to sensitive financial data compared to traditional role-based access control mechanisms. This technical blog post explores how to implement a dynamic ABAC authorization system for the banking industry using Hasura's GraphQL engine. The process involves defining the data model, integrating Hasura's GraphQL Engine, implementing dynamic ABAC authorization, leveraging external data sources, and testing and auditing the system. By defining fine-grained access control policies based on dynamic attributes and roles, a robust and efficient authorization system can be built to ensure the security and privacy of sensitive financial data in the banking industry.

The Hasura Snowflake connector allows for generating APIs (GraphQL and REST) on Snowflake data with minimal coding. By using the Hasura Snowflake connector, one can replace multiple components in an architecture with just one, simplifying the process. This article demonstrates how to implement APIs over Snowflake using Hasura's query engine and user-defined functions (UDFs). The APIs are protected with API keys, and role-based authorization can be added for enhanced security. By using Hasura, developers can focus on building the end product without getting entangled in details.

The blog discusses the improvements made by Hasura to enhance the observability of subscriptions and event triggers in its GraphQL engine. Drawing inspiration from Google's Golden Signals monitoring approach, Hasura incorporates latency, traffic, errors, and saturation into subscription observability. This allows developers to monitor and debug their Hasura-based applications effectively. Event triggers in Hasura enable reacting to database events and invoking webhooks. Monitoring metrics helps identify saturation and take necessary actions such as increasing the number of HTTP workers or scaling the Hasura instance. The blog also provides instructions on how to try out the observability setup with self-hosted Enterprise Edition version and export observability data from the Hasura cloud.

The blog discusses the benefits of integrating SQL Server stored procedures with Hasura GraphQL Engine to leverage their combined power and performance for application development. SQL Server stored procedures encapsulate business logic and complex data operations, while Hasura's GraphQL engine provides efficient data retrieval capabilities. By supporting read-only stored procedures in SQL Server, developers can utilize existing logic, improve performance, control access to data, and simplify business logic. The integration allows for the creation of robust applications that meet the demands of today's digital landscape.

Hasura now supports Remote Joins across all supported data sources, including PostgreSQL, Snowflake, MySQL, SQL Server, BigQuery, Oracle, Athena, and Remote Schemas. This feature enables developers to join data across tables and remote data sources, empowering them to build sophisticated and data-rich applications. By providing a unified view of data, Hasura's remote joins can significantly reduce the time it takes to develop data intensive applications, allowing enterprises to bring applications and data products to market faster.