January 2020 Summaries

The text outlines a comprehensive guide for setting up a periodic backup system for Couchbase clusters using the cbbackupmgr utility within a Kubernetes environment. Emphasizing the importance of regular data backups for disaster recovery and data consistency, it discusses best practices such as provisioning a separate backup node to avoid resource competition with active database nodes and allocating sufficient storage to accommodate multiple snapshots. The guide details a Periodic Merge strategy to minimize database overhead and storage requirements by executing incremental backups, compaction, and merging of old backups. It provides step-by-step instructions to configure storage classes, persistent volumes, and backup repositories, and includes a script to automate these processes using Kubernetes cron jobs. Moreover, it covers the restoration process, allowing for data recovery to a new Couchbase cluster through Kubernetes Jobs, while also recommending regular testing of backup restoration to ensure reliability.

This article guides users on how to extend their Couchbase Analytics setup by connecting it with Power BI Desktop using the CData Power BI driver for Couchbase. To start, a user needs to have a Couchbase Server Enterprise Edition 6.x cluster with Data and Analytic services enabled, and set up a single-node local install of Couchbase Server EE. The user then creates shadow datasets in Analytics and initializes them by connecting to their Couchbase server. Next, the user installs Power BI Desktop and sets up the CData Power BI driver for Couchbase, specifying parameters such as data source name, user credentials, and verbosity level. Once set up, users can create new reports in Power BI Desktop sourced from Couchbase Analytics, exploring insights and leveraging other tools with the many available Couchbase CData drivers.

As Valentine's Day approaches, the author reflects on childhood memories of exchanging valentines and connects this nostalgic warmth to their role as Head of Talent at Couchbase. They emphasize the importance of choosing to work with creative and dedicated individuals and aim to infuse this sentiment into recruitment efforts through a social media campaign titled "Why I Chose Couchbase." The campaign, launched with posts from new recruits like Senior Developer Advocate Eric Bishard, aims to showcase why individuals choose to work for Couchbase, inviting others to explore career opportunities at the company. The initiative underscores Couchbase's commitment to fostering a collaborative and innovative work environment while actively recruiting new talent.

The blog post serves as a tutorial for deploying an ASP.NET Core microservice on a Kubernetes cluster, specifically focusing on Azure's AKS but applicable to other Kubernetes services. It follows a previous post on creating microservices with Couchbase using docker-compose. The tutorial guides readers through building a Docker image, creating an AKS cluster, and deploying both an ASP.NET Core microservice and Couchbase Server using Kubernetes operators. It covers advanced topics like DNS SRV for database location, setting up port forwarding to access AKS pods, and configuring a Load Balancer to expose the service to the internet. The tutorial requires basic knowledge of Docker, Visual Studio, and Kubernetes tools like kubectl, and provides additional resources for further exploration.

The text details the enhancements introduced in Couchbase Server version 6.5 regarding N1QL's ANSI JOIN and MERGE functionalities. Initially, N1QL's ANSI JOIN required the right-hand side to be a keyspace, a restriction lifted in version 6.5, allowing expressions and subqueries to be used instead, thus facilitating more flexible query operations. These modifications improve N1QL's ANSI compliance and usability, particularly in migrating from relational databases. Additionally, Couchbase 6.5 introduces support for ANSI MERGE, which allows for more sophisticated merging operations based on arbitrary expressions rather than solely on document keys, expanding the functionality beyond the previous ON KEYS clause. The text also explains the introduction of hash joins as the default method for expressions or subqueries in ANSI JOIN, the handling of multiple matches in ANSI MERGE, and the requirement for specifying document keys in INSERT actions within ANSI MERGE. These updates aim to make N1QL more user-friendly and versatile, aligning its capabilities more closely with those of traditional SQL in relational databases.

Couchbase Server 6.5 GA has been released with several key features, including distributed ACID transactions, advanced query capabilities like window functions, index advisor, cost-based optimizer, eventing enhancements, and improved HA and observability. The new release also introduces collections to allow applications to structure their data by keeping similar documents together, just like tables do in relational databases. Additionally, the release includes advanced filtering in XDCR, Quality of Service for XDCR, robust rebalance operation, node-to-node encryption, LDAP group support, TLS cipher configuration, and significant improvements to backup and recovery. The new version also provides unparalleled performance at any scale, with the ability to create up to 30 buckets in a single cluster and improved storage compression.

The new Couchbase 6.5 release introduces extensive Enterprise Grade Database Query capabilities, including transactional support, Analytical Window functions, user-defined JavaScript functions, and Index Advisor to improve query performance. The release also includes a set of new enhancements to further strengthen the Couchbase Index Service, improving operational efficiency with features such as dynamic settings changes without indexer restarts, improved DCP rollback handling, optimized in-memory snapshotting, and projector memory usage improvements. Additionally, the release introduces a new `BUILD INDEX` command to rebuild all unbuilt indexes at once, allows scans during indexer warm-up, and provides stat information for each index, including its last known scan time and key size distribution. The Index UI has also been improved with important summary information about each index, including requests/sec, resident ratio, items, data size, and status. These enhancements are designed to address customer requests and improve the overall performance and manageability of the Couchbase platform.

The new feature in Full Text Search (FTS) allows for geo bounded polygon query capabilities, enabling more precise results by specifying region boundaries in a granular way, rather than relying on simple rectangular or square regions. This feature adds the capability to perform random bounding polygon queries to select documents within a polygon area, using an array of coordinate pairs representing vertices that define the polygon area. The search process involves indexing documents with geolocation fields, identifying a polygon area, and trying the query in any geoJSON format, such as specifying coordinates in an anticlockwise direction or using a closed format. The feature uses ray casting technique for finding point inclusivity within the polygon and does not support holes in polygons, but users can perform such queries using boolean queries with a must_not clause.

The text explores various methods for visualizing data using the Couchbase Query Workbench, emphasizing its capability to easily transfer query results into formats compatible with tools like Google Sheets and Excel for chart creation. It provides examples of queries on the travel-sample dataset to produce visualizations such as pie charts, histograms, doughnut pie charts, and geo charts, illustrating how to analyze data on document types, hotel counts in cities, and review contents. Additionally, it highlights the use of the RATIO_TO_REPORT() function to calculate percentage values for bubble charts and facilitates the creation of geo charts with marker bubbles. The overall aim is to demonstrate the ease of presenting data visually using Couchbase's features, thus aiding in data analysis without the need for advanced data visualization tools.

Couchbase's Index Planner is a critical component of its indexing service, designed to optimize the placement of Global Secondary Indexes (GSI) for high availability and load balancing within a cluster. It uses probabilistic simulated annealing to determine the best distribution of indexes across indexer nodes, considering various load parameters such as index size and scan rate. The planner supports two main scenarios: new index creation and index service rebalance, adapting to hardware differences among nodes. It ensures high availability by placing index replicas on different nodes and respects server group configurations for rack zone awareness, distributing replicas across failover zones to maintain data resilience. Additionally, the Index Planner manages partitioned indexes by distributing their partitions across multiple nodes, enhancing load distribution and scan efficiency. It emphasizes flexibility in load distribution, ensuring that both partitioned and non-partitioned indexes benefit from optimal resource use and availability across the cluster.