August 2018 Summaries

Couchbase has launched its Autonomous Operator for Kubernetes, marking the first production-ready Kubernetes operator for a NoSQL database, compatible with open-source Kubernetes and Red Hat OpenShift. This innovation enables enterprises to leverage the Couchbase Engagement Database in cloud environments, enhancing app development with features like automated cluster provisioning, failure recovery, dynamic scaling, and centralized configuration management. The operator supports enterprise-grade high availability and cross-datacenter replication, facilitating hybrid and multi-cloud strategies and improving DevOps efficiency. Developed in collaboration with customers and partners through a series of beta releases, it has been praised for significantly reducing IT administration overhead and simplifying database management on Kubernetes platforms.

Couchbase Mobile 2.1 has been released, building on the advancements of version 2.0 with enhanced security, performance, and availability for mobile applications. Key features include peer-to-peer data synchronization, allowing devices to sync data directly without a central server, and on-device data encryption using AES-256, enhancing data security for industries like healthcare and finance. The Sync Gateway now guarantees high availability by using N1QL/GSI for internal functions, eliminating downtime during server rebalances or failovers, and introduces continuous logging for improved support capabilities. Additional enhancements include log redaction for privacy and support for X.509 certificate authentication, ensuring comprehensive security and operational efficiency. These updates are designed to meet the needs of business-critical applications, offering a robust platform for edge data management and syncing.

Couchbase has introduced significant updates in its 6.0 beta release, enhancing its Analytics, Eventing, and Full-Text Search services. The Couchbase Analytics Service, now in beta, supports HTAP workloads by enabling ad-hoc analytical queries on NoSQL data with an MPP query engine, allowing fast data ingestion and processing without impacting performance. This feature eliminates the need for ETL processes and enables agile, cost-effective analytics by running queries on dedicated nodes. Enhancements in the Eventing Service include the addition of timers for triggering routines at specific times, while the Full-Text Search service has improved indexing capabilities, offering faster indexing and reduced data size. These advancements aim to provide unmatched agility, flexibility, and performance at scale, making the platform easier to use and manage. Users are encouraged to explore the new functionalities and provide feedback on the Couchbase forums.

Couchbase Server version 6.0 Beta introduces the Couchbase Analytics Service (CBAS), designed to perform complex, ad hoc queries on large datasets without affecting operational throughput by functioning as a separate, scalable service. The service was showcased in a technical demonstration using over 100 million synthesized medical data documents based on the FHIR standard, highlighting the ease of configuration and the absence of ETL processes. The setup involves creating datasets and executing queries with SQL++, a language optimized for JSON formats. The analytics queries, primarily executed via REST endpoints in a Node.js environment, utilize sophisticated strategies like index nested loop joins for efficient data retrieval and visualization through Chart.js. The blog post emphasizes query optimization and provides insights into the query planning and execution process, illustrating the benefits of hints in improving query performance. Additionally, the post offers resources for further exploration of Couchbase and encourages community engagement through various platforms.

Couchbase 6.0 Beta introduces Timers in Couchbase Functions, allowing asynchronous computations linked to specific wall-clock events, which can be used for various applications like document archiving, sending notifications, and risk assessments. Developers can create timers using a specific syntax that includes a callback function, an expiry timestamp, a unique reference, and a context payload, with the requirement that each timer's combination of function name, callback, and reference must be unique. Timer features include distributed processing for scalability, ensuring execution despite node failures, and maintaining the same timeout semantics as the parent functions. However, timers cannot be debugged visually, and failures aren't automatically retried, necessitating exception handling. Timers rely on metadata storage, requiring careful management of the metadata bucket to avoid data loss or interference, particularly in high-use scenarios requiring increased memory allocation.

Azure Functions enable the use of a serverless architecture on Microsoft Azure, allowing developers to focus solely on writing code while Azure handles server management. The text highlights a video tutorial that demonstrates how to create Azure Functions that interact with Couchbase Server, with the full source code available on GitHub for those who wish to follow along. Additionally, it encourages readers to explore further blog posts about Azure and Couchbase and to join the Couchbase Community Directory to learn more about how others are utilizing Azure within the community.

Couchbase Server 5.5 and 6.0 introduce significant enhancements to the Full-Text Search (FTS) service, offering improved performance and manageability with a new indexing engine, Scorch, which reduces indexing time by 50% and decreases data size on disk by up to 70%. These releases also finalize the geospatial search feature, making it production-ready, and expand language support by adding 11 new language analyzers to the FTS, broadening Couchbase's global market reach. The updates are designed to better meet enterprise use cases, and users can explore the new features through the index creation UI or JSON-adjusted index definitions, with further guidance and examples provided in the updated documentation and training resources.

The Couchbase Analytics Service utilizes the SQL++ query language, which is designed to work with semi-structured data and the JSON data model, being backwards-compatible with SQL. The service is currently in developer preview, with occasional breaking changes between releases, so this post aims to document those changes. Key updates include automatic bucket association, removal of the SHADOW keyword for dataset definitions, and a revised syntax for connecting and disconnecting datasets, now using CONNECT/DISCONNECT LINK instead of CONNECT/DISCONNECT BUCKET. Additionally, Couchbase is open source and free to try out, with various resources available on its developer portal, including sample code, tutorials, and forums for support.

The new ANSI JOIN feature in Couchbase version 5.5 allows for more flexibility and complexity in joining data, but also introduces challenges in testing its functionality. To address this challenge, Couchbase has developed a framework called Random Query Generator (RQG) that leverages another framework called Randgen to generate SQL query templates based on user-defined parameters. These templates are then converted to N1QL formatted queries and executed against the Couchbase server and an equivalent MySQL server to ensure functional similarity. RQG is capable of generating highly complex queries, including multiple chained predicates and joins, which makes it suitable for testing ANSI JOINs. Through the use of RQG, Couchbase has found several bugs in the ANSI JOIN functionality, including issues with index selection and query hint usage. The framework's ability to generate a wide range of queries allows developers to thoroughly test the feature and ensure its correctness.

This summary provides an overview of creating a GraphQL API with PHP and Couchbase as the NoSQL database. The process involves installing necessary dependencies, setting up a connection to Couchbase, defining data models using GraphQL objects, developing queries to populate these objects, and configuring the application for use with GraphQL. The example uses Pokemon data as a model to demonstrate how to create a data relationship between different fields in a single object. The code also showcases modularizing queries on a per-field basis to manage complex relationships.

The text discusses implementing a typeahead search functionality in an application using Golang as the backend language, Couchbase as the NoSQL database, and jQuery for the frontend. The process involves creating a full-text search index, developing a RESTful API to execute queries, and designing a jQuery frontend that utilizes typeahead.js to implement the autocomplete feature. The example uses a simple dataset of song information and demonstrates how to create a native Go data structure to hold search results, configure cross-origin resource sharing for CORS, and render the typeahead functionality using Handlebars templates.

Couchbase Server 5.0 and 5.5 introduced a variety of features that enhance its functionality for developers, including the ability to manipulate sub-documents, an Eventing Service for automatic server-side functions, and the use of ANSI JOIN syntax for queries. These versions also bring Full-Text Search capabilities directly into the database, reducing the need for third-party tools, and offer performance improvements through faster queries and aggregation pushdowns. Security is bolstered with Role-Based Access Control, X.509 certification, and field encryption, while the Reactive SDK supports performance optimization through reactive programming. Developers can fine-tune performance with SDK options that balance speed and data safety, and Response Time Observability provides insights into request performance using the OpenTracing format, helping to identify and address operations with poor performance.

The new Eventing service in Couchbase Server allows developers to write JavaScript functions that respond to documents getting created, updated, or deleted, and can execute a curl function to make requests to HTTP endpoints. A proof of concept example was created using Functions (part of Eventing) to send a text message when a customer's ticket is scanned at the gate. The example uses Twilio for sending text messages and demonstrates how to create a concierge service that assigns VIP customers to specific concierges and notifies them via text message when a new ticket scan document is created or modified. The function reads documents from a bucket, gets the Twilio credentials, constructs a message, and sends it using the curl construct, which is still in development and intended for development purposes only. The service can be used to create a system that scans tickets at the gate and notifies VIP customers' concierges via text message.

The Internet of Things (IoT) introduces a new dimension to database workloads, with data generated from edge devices at gateways and servers for instant analysis. The data is heterogeneous in terms of schema and constantly evolving, requiring flexible modeling approaches such as JSON representation. IoT systems can utilize various storage formats including RDBMS, NoSQL systems, simple JSON, XML, or proprietary formats to manage the diverse data streams. An architecture for IoT data management involves aggregating, filtering, and analyzing data at the sensor level before sending it to the cloud for deeper analysis. The traditional OLTP schema is not suitable for IoT's write-heavy workload, prompting the creation of a new benchmark TPCx-IOT to measure the price-performance of IoT gateway systems. This benchmark provides a comprehensive framework for evaluating the performance of IoT databases, with Couchbase successfully implementing the workload driver and achieving impressive results on Cisco M4 hardware.

This summary provides a concise overview of building a typeahead form using Couchbase's full-text search service, Node.js, and jQuery. The application assumes that users input a query, which is then searched against the music dataset stored in Couchbase. The results are displayed as suggestions with the title and artist fields. A RESTful API endpoint is created to handle the search requests, and a frontend using jQuery and typeahead.js libraries is designed to render the suggestions. The application uses a fuzzy search with a value of 1 to accommodate misspellings of words and limits the results to 5 items displaying.