June 2022 Summaries

YugabyteDB 2.15 delivers several key new features to support greater developer productivity, including support for READ_COMMITTED isolation with pessimistic locking, new Webhook and Cloud Storage Sinks for Change Data Capture (CDC), industry-leading PostgreSQL support, and ecosystem expansion with Prisma, Django, and GORM support. The release also introduces a new REST API for YugabyteDB Managed, easy multi-region cluster deployment, single sign-on using social logins, and several other enhancements to improve developer experience and productivity.

YugabyteDB 2.15 introduces several key innovations to support dynamic workload optimization, enabling applications to adapt to changing requirements with greater ease. The new features include Quality of Service (QoS) and multi-tenancy, admission controls for heavy cluster utilization, prioritize transaction execution, rate limiting connections per node/tenant/database, admission controls for write-heavy workloads, stop writes triggers, slow down writes triggers, and dynamic app-aware sharding. These enhancements allow YugabyteDB to manage diverse sets of applications and their inherent unpredictability with greater flexibility and application support, making it easier for organizations to thrive in the face of changing workload demands.

YugabyteDB 2.15 and YugabyteDB Voyager offer numerous updates that enhance the database's performance, developer productivity, and support for diverse workloads, enabling organizations to deploy and run cloud-native transactional applications with exceptional performance in any cloud at any scale. The new features include dynamic workload optimization, improved isolation levels, enhanced Quality of Service, multi-tenancy, and dynamic app-aware sharding, as well as a powerful migration engine and automation capabilities through the YugabyteDB Voyager tool. Additionally, YugabyteDB 2.15 delivers numerous performance enhancements, including a Performance Advisor, fast bulk load of data, query optimizations, and support for new PostgreSQL features, making it an ideal choice for modern organizations with diverse transactional applications.

Narvar, an e-commerce platform trusted by over 800 retailers, recognized the need for a flexible data layer to support their growing business and evolving compliance requirements. By switching to YugabyteDB, they achieved significant cost savings (4x lower TCO) while gaining scale, performance, and zero downtime. The native geo-partitioning capabilities enabled them to easily satisfy changing compliance landscape and support multiple clouds based on customer requests. This modern data layer powers Narvar's omnichannel success by optimizing operations, diversifying channels, and elevating customer experiences.

YugabyteDB Managed – The Monthly Scoop (June 2022)` Summer is finally here, bringing new features to YugabyteDB Managed that enhance user experience and cluster management. Most clusters can now be created in about 5 minutes, making the process faster and more efficient. Social logins have also been introduced, allowing users to sign up and connect their accounts with ease. Furthermore, multi-region cluster creation has become possible, providing better availability for data and peace of mind. These updates are designed to improve user experience and provide a smoother workflow, making it easier to manage clusters and access features. The team is always listening to feedback and welcomes users to join their Slack community to share thoughts and ask questions.

The Debezium Server is a ready-to-use application that streams change events from a source database to messaging infrastructure like Amazon Kinesis or Google Cloud. To run Debezium Server with Kafka as a sink, one needs to download the Confluent Platform, compile Debezium Server, create config files, start Zookeeper and Kafka, start YugabyteDB using yugabyted, create a table and create a stream ID, start Debezium server with some configurations, start Kafka UI, insert data into the created table, and finally see the change events being populated in the Kafka topic. The setup involves several steps including downloading packages, extracting them, creating config files, starting services, and executing scripts to get the Debezium Server running with Kafka as a sink.

Distributed SQL sharding in YugabyteDB involves splitting tables into smaller units called tablets, which are then distributed across multiple servers to scale out horizontally. The ideal number and size of tablets depend on various factors such as data growth patterns, query activity, and storage constraints. Small static tables, frequently accessed with joins, can be broadcasted to a single tablet with duplicate covering indexes, while medium and large tables can be split automatically using hash or range sharding methods. Very large tables growing with business activity may require partitioning to optimize future archiving or purging operations, and huge tables should not grow indefinitely beyond the default splitting thresholds to avoid operational challenges. By understanding how sharding works and tuning the auto-splitting knobs, developers can ensure optimal performance and scalability for their distributed SQL databases.

To publish data from a YugabyteDB database to a Kafka topic using Debezium CDC, first configure the source connector to publish changes to a Kafka topic, then deploy the JDBC sink connector with a custom message transformer to apply the changes on the sink as needed. Start by setting up a PostgreSQL instance and connecting to it, inserting data into YugabyteDB, and then checking if the records have been replicated to the Postgres database using CDC. This approach allows for real-time replication of data between databases, enabling more efficient data processing and analysis.

YugabyteDB is a cloud native distributed SQL database designed for resilient Kubernetes workloads, providing availability and resiliency challenges when running application workloads on Kubernetes. The database functions as a single logical database deployed as a cluster of nodes, taking care of sharding, replication, load balancing, and data distribution. YugabyteDB is available on Red Hat OpenShift, the industry's leading enterprise Kubernetes platform, allowing for efficient Day 1-2 operations and geo-distributed stateful workloads. It provides zero data loss and continuous availability during a major system outage or natural disaster, making it an ideal combination to ensure Kubernetes deployments are resilient and continuously available. The database serves business-critical applications with SQL query flexibility, high performance, and cloud-native agility, allowing enterprises to focus on business growth instead of complex data infrastructure management.

Kroger is enhancing its customer experience through the adoption of a modern data layer powered by Yugabyte's distributed multi-cloud SQL database, YugabyteDB. This move aims to address scalability, reliability, and cost concerns while providing a seamless shopping experience for customers and efficient frontline operations. With YugabyteDB, Kroger can deliver accurate product information, meaningful loyalty programs, and competitive pricing while reducing IT operational costs. The company has chosen YugabyteDB for its distributed ACID transactions, scalability, multi-region active-active deployments, and automatic data sharding capabilities, allowing it to achieve single-digit millisecond latency in its multi-region configuration.

Log Aggregation in YugabyteDB with Grafana Loki` Log aggregation is a crucial aspect of distributed systems where multiple processes generate vast amounts of data. In a system like YugabyteDB, aggregating logs from various processes helps transform raw log data into insightful information that enables better product management. This blog explores how to aggregate log data from YugabyteDB using Promtail, Loki, and Grafana. It starts with installing the necessary tools and configuring them for efficient log aggregation. The process involves setting up a three-node database cluster, configuring Promtail to collect logs from each instance, forwarding the data to Loki, and then visualizing the data in Grafana. This tutorial provides a hands-on experience with log aggregation using YugabyteDB-Gitpod integration, making it accessible to developers who want to try it out for themselves.

As a Yugabyte APJ intern, Lakshmi learned the importance of supercharging her first 90 days by picking up technical knowledge and networking with the team, adopting a curious mindset by experimenting and pushing boundaries, and having fun while collaborating with the team. Despite typical internship expectations, Lakshmi's experience was a blast and she gained valuable lessons that will help her grow in her career. She also appreciated the experimental and fail-forward approach of the team, which allowed her to integrate into the company culture. The Yugabyte APJ team is currently hiring for various positions, offering opportunities for individuals to join a dynamic company in growth mode.

To secure communication between CQL clients and the YCQL query interface of YugabyteDB, organizations can enable client-to-server encryption in transit using various command line flags. This achieves Goal 1, making it cryptographically impossible to snoop or modify network communications. Verification of server certificates is also crucial, which can be achieved through strict validation of the server certificate's Common Name and Subject Alternate Names. However, verification of client certificates by the server is not currently supported by YugabyteDB. Organizations can also specify a minimum TLS version and customize cipher lists to meet their specific requirements, ensuring Goal 3, enabling the server to identify and verify the client, is achieved. By implementing these settings, organizations can secure their CQL clients and achieve resilience to snooping, while also supporting custom CAs and CA-signed certificates in YugabyteDB Anywhere.

Building a Simple Application with YugabyteDB and Prisma` is a tutorial that guides developers through the process of creating a simple application using YugabyteDB and Prisma. The tutorial begins by introducing Prisma, an ORM tool for Node.js applications, and its integration with YugabyteDB. It then demonstrates how to connect to a YugabyteDB Managed instance, initialize the database, and configure the Prisma Client. The tutorial also shows how to seed the database with users and run the sample application. Throughout the process, the authors highlight the ease of use and flexibility of using Prisma to interact with YugabyteDB, making it an attractive option for developers building modern web applications.