October 2017 Summaries

The explosive growth in data has led companies to adopt common formats such as XML and JSON to manage this data. Despite its popularity, JSON presents a challenge when ingested into relational databases due to its unstructured nature. This is where tools like jq come in handy, which can transform, extract, filter, and change the structure of JSON data completely. SingleStore has integrated jq as a native JSON transformer, providing users with a performant and easy-to-use solution to handle one of the most common data formats today. By leveraging features in jq, companies can ingest and process large amounts of unstructured JSON data from platforms like CloudTrail without needing extra tools, allowing them to reap the benefits of both structured and unstructured data paradigms.

The latest version of SingleStoreDB Self-Managed 6 has been released, offering breakthrough query performance, enhanced online operations, and extensibility features. The new Extensibility feature was built based on market demand and enables customers to move database workloads that require stored procedure functionality into SingleStore. With Extensibility, customers can write custom logic in the database for data manipulation and processing, allowing them to take advantage of what the database already does well, such as transactions and role-based security. MPSQL, a custom language introduced with this release, provides benefits like compilation down to machine code, distribution, and Turing completeness, making it suitable for large-scale applications like machine learning, which can utilize these new features to process data at high performance speeds.

SingleStore powers a number of IoT applications to help manage operating costs while improving customer experiences by providing core database capabilities including streaming data ingest and store, fast query response, proven compatibility, scalability and availability. The database must collect and store multiple streams of data into relational formats to support real-time and historical analysis, perform instant queries across millions of events or devices, leverage the familiarity of ANSI SQL with full data persistence, and utilize a modern shared nothing architecture to scale out with industry-standard hardware. SingleStore has helped industry-leading organizations such as Uber, Verizon, Comcast, and Cisco deliver IoT applications powered by analytics at scale, including real-time monitoring and detection, predictive maintenance applications, and fleet optimization.

At SingleStore, product design is a holistic process that involves gathering requests, ideating, and delivering pixel-perfect products. The company uses a full-stack approach where each designer takes ownership of a product or feature and engages with other team members to ensure constant feedback and iteration. This includes research, user flow diagrams, wireframes, high-fidelity mockups, prototypes, writing, and implementation. The design process is iterative, requiring patience and commitment, and involves validation through data analysis, task analysis, and customer feedback. Designing as a team and encouraging contributions from multiple stakeholders across departments are crucial to producing effective design solutions that meet customers' real needs.