November 2022 Summaries

Hybrid tables are a type of table that supports both transactional and analytical workloads, enabling faster performance at lower costs and simplified development for developers. They offer better performance at lower costs by consolidating multiple databases into one, reducing data migration efforts, and pushing data in and out of disk and memory without worrying about network costs. Hybrid tables also simplify development by giving developers a single solution to serve all their data workloads, and reduce ETL workloads by providing a scalable storage engine that saves time and money. With hybrid tables, businesses can improve efficiency by implementing them into their data workflow, which supports various business functions with data warehouses and analytics, and enables real-time data analytics without requiring a separate analytics data store.

The 2022 SingleStore Hackathon saw impressive submissions across five categories, with winners announced in each category. The winning projects showcased innovative applications of SingleStoreDB, including a multi-model application for animal discovery, a hybrid transaction and analytical processing system for real-time internet diagnostics, an AI-based path optimization tool using Wasm, a machine learning-based movie recommender, and a database migration solution to integrate with SingleStoreDB. These projects demonstrated the versatility and capabilities of SingleStoreDB, highlighting its potential for building scalable, high-performance applications in various domains.

The playground of SingleStoreDB allows users to run queries over various datasets, including TPC-H, TPC-DS, and a semi-structured game dataset. The playground provides a SQL editor with pre-loaded databases, enabling users to explore and learn best practices for working with JSON data in SingleStoreDB. Key strategies include using Persisted Computed Columns, leveraging the `::` and `$` symbols for parsing out computed columns, and utilizing the Visual Profiler to compare performance of querying whole JSON columns versus persisted computed columns. Users can create views, join tables, output table rows as JSON, build nested JSON objects from columns on their table, perform analytics queries, and more. By mastering these techniques, developers can power JSON-intensive applications at scale with SingleStoreDB.

In this article, you'll learn about open-source databases, their benefits, and what other options to keep in mind as you scale your business up. Open-source databases are those whose code is available to view, download, modify, and reuse by anyone based on their requirements. They can be categorized into two groups: relational and non-relational. Relational databases store data in key-value pairs and arrange them on a row and column basis, while non-relational (NoSQL) databases are used to store unstructured data like documents and images. Some popular open-source databases include PostgreSQL, MySQL, MongoDB, and MariaDB. These databases have benefits such as speed of implementation, cost-effectiveness, ability to start small, flexibility, and agility. They allow developers to quickly identify if a community version of a database is suitable for their use case, get started, and deliver the use case. Open-source databases are free and more cost-efficient by design, making them an attractive option for businesses. They also provide flexibility and agility, allowing developers to tweak them for free without restrictions. However, they may have limitations when it comes to scaling, which is where a commercial solution might be preferable. SingleStoreDB is a cloud-native, distributed SQL database that's well-suited for real-time applications and analytics, providing benefits such as fast query results, high availability, ability to load data from multiple data sources, and ability to integrate with Tableau for data visualization. It can be connected with MySQL and other databases using the command line or TLS/SSL. The article concludes by highlighting the importance of open-source databases and their benefits, as well as the potential of SingleStoreDB in scaling open-source databases.

SingleStore is excited to participate in AWS re:Invent this year, an event that takes place across six Las Vegas hotels from November 28 to December 2. SingleStore will have multiple activities, including product demos, customer parties, and a magician at their booth, where attendees can see the magic of the database firsthand. The company's SVP of Engineering Shireesh Thota will be interviewed on TheCUBE, discussing real-time databases and modern database capabilities. Additionally, SingleStore will collaborate with Intel to demonstrate how its database performs fast analytics and AI in SQL using Intel processors. Furthermore, SingleStore will host an "Engineering Under the Hood" session, where attendees can hear about high availability and scalability from Shireesh Thota, along with a surprise special guest.

The retail industry is undergoing significant changes, driven by customer expectations for a seamless shopping experience, personalized recommendations, and real-time analytics. To deliver these experiences, retailers need to access data in real-time, but traditional data architectures often struggle to meet this requirement due to slow refresh rates and delayed insights. The webinar "Real-Time Retail With IBM and SingleStoreDB" explores the challenges of building a real-time data infrastructure for interactive applications and analytics, highlighting the limitations of legacy data architectures and the importance of unifying transactions and analytics in a single database. SingleStoreDB is introduced as a fully distributed SQL database that scales with data requirements, powered by IBM Cognos to empower users to make sense of retail data, reducing query latency, hyper-scaling concurrency, and lowering total cost of ownership.

Anuvu, a leading global airline telecommunications provider, struggled with operational complexity, data errors, and inconsistencies due to a patchwork of redundant databases. They initially used MySQL and MariaDB for their unlimited record storage and free community version, but eventually decided to modernize and consolidate their solutions. After testing various options, including Snowflake and Redshift, they chose SingleStoreDB for its MySQL compatibility, rowstore engine, columnstore capabilities, and performance, which outperformed Snowflake and MariaDB by 10-20x and 70-100x respectively. With SingleStoreDB as their primary database, Anuvu is now able to distribute data and insights to all of their clients accurately and in real-time, improving customer experiences and reducing costs.