February 2020 Summaries

This webinar series aims to provide a comprehensive approach to cloud data migration, focusing on strategic decisions, challenges, and best practices. A good cloud data migration involves prioritizing workload types, identifying dependencies, and assessing the shape of data models. The process requires elasticity, automation, and careful planning to minimize risk. Assessment of workload is crucial, considering data shapes, types, and models used. Migration can be done in a partial fashion using a pattern that has been successful with customers, replicating only the analytical workload first. Automation tools play a significant role in the migration process, but manual work may still be required for stored procedures and procedural languages. The webinar also discusses database migration best practices, including managing SingleStore in various environments, and provides insights into customer experiences with migrating from Oracle and other databases to SingleStore.

The webinar discusses the benefits and risks of moving operational databases to the cloud, which is a crucial step in digital transformation. It highlights the motivations for migrating to the cloud, including scalability, cost-effectiveness, and access to new capabilities such as machine learning and IoT integration. However, it also emphasizes the challenges and complexities involved, including data migration best practices, application dependencies, and hardware compatibility. The webinar aims to provide a comprehensive understanding of the process, including assessment, classification, and migration planning, to help organizations make informed decisions about their cloud strategy. It also touches on the differences between SingleStore and Amazon Aurora, highlighting SingleStore's efficiency, scalability, and capabilities in handling a wide range of data types and workloads.

You can run Singlestore Helios with $600 of FREE credits, allowing you to test the database for demos, software testing, developer productivity, and brief proof of concept tests, without worrying about infrastructure maintenance. The free trial includes a 4-node cluster with up to 32GB RAM, and is available in various cloud environments, such as Docker Desktop, Kubernetes, Linux, and Vagrant. Running on Singlestore Helios provides an easy-to-use and managed database experience, outsourcing infrastructure management to the cloud provider, allowing you to focus on developing your application. With a free license that never expires, you can spin up a cluster in minutes, and start being productive with SingleStore Studio, which includes features like SQL Editor, list of databases, and Query Profiler. The trial is ideal for "kick-the-tires" tasks, such as connecting to an analytics dashboard or starting an application, and allows you to scale up to accommodate large data workloads.

The author of the letter is leaving their current vendor, referred to as "Larry", due to various issues such as slow reaction times, inflexibility, and inability to support mixed workloads and data types. The author finds that other vendors, specifically those in the NewSQL crowd, offer better performance, scale, and flexibility at a lower cost. The letter expresses gratitude for Larry's efforts but states that it's time to move on to a more modern and forward-thinking vendor.

Kubernetes is an open-source orchestrator built by Google, now managed by the Cloud Native Computing Foundation (CNCF), designed to manage containers and scale to meet the needs of modern web applications. It is a novel platform that pools machines together into a single cloud OS, allowing for extensibility and no-downtime rolling updates. Kubernetes has made a name for itself as the cloud-native platform for running containers, offering extension points that allow users to replace pieces with community extensions or custom-built components. The orchestrator's architecture consists of the Control Plane (managing machines) and Worker Nodes (running workloads), with various components working together to monitor health, restart services, and ensure desired states are met. With its extensibility and ability to host microservices, Kubernetes is a great platform for hosting web properties and data stores, such as stateful data stores like SingleStore.

The MEAN stack, which includes MongoDB, Express.js, Angular.js, and Node.js, has served as a widely adopted pattern for web development. However, its components have failed to keep up with the times, leading to scalability issues and limitations in handling modern web applications. The JAM stack, which includes JavaScript, APIs, and Markup, is an updated approach that focuses on the thick-client browser environment and marginalizes the backend. In contrast, a modern system prioritizes cloud-native scale, utility billing, infinite scalability, low latency, user-relevance using machine learning, and storing and processing disparate data types and sources. SingleStore, a NewSQL database, has emerged as a solution that meets ACID guarantees, provides durability, supports SQL, and delivers boundless scale in the cloud. The author argues that SingleStore is the modern stack's database of choice for crafting modern app stacks.

Capterra is a leading online resource for business software buyers with over one million verified reviews, growing fast and doubling nearly in mid-2019, showing significant growth in its user base. The site focuses on small and medium businesses (SMB) users, while enterprise users can find rounded comments on G2 Crowd and highlights from Gartner Peer Insights. The most helpful reviews highlight SingleStore's best features, such as speed, scalability, machine learning capabilities, and ease of use, with many users calling it a "great product" and "the go-to option" for real-time analytics and dashboards. Many reviewers praise the software's performance, ease of setup, and support, citing its ability to handle large amounts of data and achieve significant performance wins. Overall, Capterra provides an objective view of SingleStore's suitability for various needs.

To get started with SingleStore on your laptop, you can quickly provision a "cluster-in-a-box" cluster using a Linux virtual machine. This setup allows you to spin up a Small Store cluster, connect it to SingleStore Studio, and start being productive. You'll need at least 8GB RAM and four CPUs to run this minimal version of SingleStore on your laptop. The steps include getting a free SingleStore license, installing VirtualBox, starting a Linux VM, installing SingleStore's apt repository into Linux, installing the latest version of SingleStore Toolbox, SingleStore command-line client, and SingleStore Studio, starting the SingleStore cluster with memsql-deploy, disabling hardware capacity checks, launching SingleStore Studio, and creating a new cluster. With this setup, you can try out SingleStore capabilities, test SingleStore capabilities, prototypes, proof of concept (PoC), production workloads, high availability, and performance testing.

This blog post describes how to run an entire SingleStore cluster on a laptop using Vagrant, a tool for managing virtual machines. The setup involves installing Vagrant and VirtualBox, creating a custom Vagrantfile, provisioning a Linux VM with the necessary dependencies, and starting the SingleStore cluster. With this minimal configuration, developers can quickly test and understand the capabilities of SingleStore without needing expensive hardware. The post highlights the benefits of using a single-machine cluster-in-a-box setup, including ease of use, reduced overhead, and compatibility with various environments.

This is the first part of a two-part blog post that discusses the recent release of SingleStoreDB Self-Managed 7.0, which includes high-throughput, synchronous replication as one of its major features. The database's underlying architecture and how it achieves fast performance are also discussed, including the use of a transaction log to maintain durability and availability. The focus is on transactions, which are used by applications to interact with the database, and how replication is used to ensure that data is always available and can withstand failures. The presentation also touches on the concept of primary-secondary replication and statement replication, and concludes with an overview of the revamped replication system in SingleStoreDB Self-Managed 7.0.

SingleStore's recent release of 7.0 features fast replication as one of its major features, offering high-throughput and synchronous replication that only slows down performance by about 10% compared to asynchronous replication. Rodrigo Gomes, a senior engineer at SingleStore, discusses how the company achieved these results and presents alternatives for replication, including considering primary and secondary nodes, serializing transactions, and using sequence numbers or IDs to order transactions. The presentation also touches on the challenges of implementing fast replication, such as avoiding conflicts between concurrent transactions and ensuring data consistency. SingleStore's approach uses pessimistic concurrency control, reserving disk space for each transaction, and managing metadata separately. The company is working to improve isolation levels in future releases, aiming for higher performance and reliability.