August 2022 Summaries

We are pleased to announce that developers can now run SingleStoreDB locally on their Mac ARM — M1, and have simplified the configuration for deploying it. When signing up for a SingleStoreDB account, developers get $600 free credits to use in the cloud managed offering over a 60-day period, as well as the ability to run SingleStoreDB on four units (up to 32 vCPU) on their local machine, including Mac M1 machines, for an extended period. To install SingleStoreDB locally using Docker, developers can follow a set of instructions that include signing up for a free trial account, getting a license key, and copying and pasting specific docker commands in their local CLI. Once installed, developers can run SingleStoreDB from Docker and connect to it with the Studio by accessing `http://localhost:8080/` on their local browser.

SAS Viya with SingleStore is a cloud-enabled, in-memory analytics engine that provides quick, accurate and reliable analytical insights by consolidating multiple data sources into a single, analytically powerful solution. It offers faster processing for huge amounts of data and complex analytics, including machine learning, deep learning, and artificial intelligence, as well as support for cloud, on-site or hybrid environments. The integration enables organizations to integrate transactional and analytical data into a comprehensible analytic ready format for downstream AI use cases, providing benefits such as faster time to decisions, open data framework, simplified data access, increased productivity and cost optimization of analytical systems.

SingleStoreDB is a real-time, distributed SQL database designed for data-intensive applications, offering fast analytics in real time. It powers unified, low-latency access to large datasets and supports both OLAP and OLTP workloads. With its unique features such as Universal Storage, limitless point-in-time recovery, fast ingestion via pipelines, and multi-model capabilities, SingleStoreDB enables businesses to drive real-time actions and stay competitive in a rapidly evolving data landscape. By leveraging connectors like Spark and Kafka, developers can get started with SingleStoreDB and unlock the full potential of their applications, including building new modern applications with operational analytics workloads, augmenting single-node databases, and modernizing legacy datastores.

Gartner predicts that more than half of enterprise IT spend will shift to cloud computing by 2025, as it allows organizations to focus on differentiating their business within their market segment. SingleStoreDB, a multi-model, multi-cloud database, is recognized for its scalability and performance, with hundreds of customers across the globe. The company offers two deployment methods: Singlestore Helios (managed service) and SingleStoreDB Self-Managed (self-managed), which provide flexibility to customers. Many customers are migrating from self-managed to managed services due to the benefits it provides in terms of day-to-day database infrastructure management, allowing them to focus on business differentiation. The migration process involves several steps, including inventory audit, firewall configuration, backup and restore, and validation testing.

It's been an amazing half year for SingleStore, marked by significant collaborations, funding, and expansion of real-time features. The company outshined major database competitors in a GigaOm study, launching a data intensity assessment tool to help customers understand the impact of data on their infrastructure. SingleStore has received industry recognitions from Dresner Advisory Services and TrustRadius for its excellence in product/technology, sales and service, as well as value and confidence. The company's continuous growth and commitment to innovation have earned it consecutive awards from both organizations, with TrustRadius recognizing its value, feature set, and relationship in relational databases. SingleStore is actively seeking new talent and invites customers to try its product at no cost, with open positions available for those interested in joining the team.

We've been using the Git Rebase workflow with GitLab Code Review and CI for over a year now, managing a large repository with 10-40 commits per day across multiple time zones. We use the "Fast-forward" merge method in our Merge Requests configuration to force rebasing of side branches against the main branch before merging. However, this requires us to manually trust that pipelines have succeeded and been recently updated, rather than having a one-click rebase operation. To address this, we're exploring features like Merge Trains, which could allow us to push all code changes into a train instead of the main branch, eliminating pipeline time waste. We've also set up custom roles for our CI pipelines to enable manual deployment and deployment notifications via a Slack bot, making it easier to manage our deployments and track changes.

SingleStoreDB is a distributed SQL database that scales out to power large enterprise data applications, now with its Management API (REST), making it easier to deploy and manage across various parts of the business. The new Management API allows for automation and integration with existing tools, supporting both transactional and analytics workloads, and is available to all customers to enable deployment and management at large enterprise scales. Representational State Transfer (REST) was chosen as the architectural style due to its ease of use and integration with existing enterprise infrastructure, ensuring secure and reliable management operations. The API offers endpoints for creating, retrieving, updating, suspending, resuming, and deleting workspaces, requiring an API key for authentication. Clients can make requests using HTTP verbs such as GET, POST, PATCH, and DELETE, and generate OpenAPI clients to develop client SDKs in various languages.

I will connect to SingleStoreDB from my local notebook, write into a new table and read data from an existing table using Python libraries such as Pymysql, pandas, Plotly, and SQLalchemy. I will also visualize the data using Plotly by selecting specific columns from the database. To start with this tutorial, you should be familiar with working in a local notebook environment, installing necessary libraries, and connecting to SingleStoreDB.

This edition of SingleStore’s engineer showcase puts a spotlight on Vedran Cindric, the founder and CEO of Treblle, an API observability platform. Vedran shares his background in tech, favorite tools, why he chose SingleStore, advice to developers navigating a crowded market, and future plans for his project. He praises SingleStore's performance as its biggest strength, wishing it had foreign key support like MySQL. Vedran is a huge fan of Laravel and Laravel Vapor, which have allowed him to build Treblle. He advises developers to think first, code later, step back when needed, and use tools that make their life easier. PHP is his go-to technology, having been instrumental in launching his startup and completing numerous projects. Vedran's experience with SingleStore has enabled him to add analytical features to Treblle, making it a crucial part of his platform.

SingleStoreDB supports fine-grained access control through its Role Based Access Control (RBAC) mechanism, which ensures that users have only the necessary privileges to access database objects. The system also includes Row-Level Security (RLS), a feature that allows administrators to restrict access to specific rows within tables based on user roles. This enables organizations to balance security and governance while scaling their applications. RLS provides additional benefits such as ease of use, change management, and scalability, making it an attractive solution for multi-tenant applications where data from different users needs to be stored in a common table definition. By using SingleStoreDB's RLS feature, organizations can ensure that sensitive data is only accessible to authorized users.

We're making it easy to load data into SingleStoreDB with a dedicated UI that allows users to select from AWS S3 buckets and load JSON or CSV files with just a few clicks. To use this feature, you need to have a SingleStoreDB workspace up and running and your AWS credentials configured. The Load Data page guides you through the process of defining data structure and operations, allowing for renaming and removal of columns as needed. After loading the data, you can review the SQL queries used to ingest the data and take a sneak peek at the status of your data pipeline, which is a SingleStoreDB feature used to load data from external sources. This UI is currently in preview, with plans to expand its capabilities in the future.

The text discusses improving the speed of queries in MySQL, using a demo database called `airportdb` that provides a structure for the discussion. The author emphasizes the importance of using the `EXPLAIN` command to analyze how the database manager is processing queries and identifies bottlenecks. Optimizations include avoiding FULL SCANS, creating indexes based on primary keys, using indexes with LIKE statements, grouping records with GROUP BY and ORDER BY clauses, and preferring derived tables over subqueries. Additionally, the text introduces SingleStoreDB as a real-time, distributed SQL database that offers improved performance and scalability compared to MySQL.