February 2023 Summaries

Momento Cache has expanded its support to the AWS ap-south-1 region in Mumbai, enhancing accessibility and performance for customers in India by offering lower latencies and faster response times. This expansion aims to provide users with a swift and efficient experience, complemented by the full range of Momento Cache features, including new collection data types. Indian partners such as AntStack, a serverless-first consulting company, have leveraged Momento Cache to significantly reduce their API latency by up to 40% for applications like HooChat. The move underscores Momento's commitment to broadening its global presence and optimizing cache performance for its users.

In an episode of the Ready, Set, Cloud podcast, host Allen Helton interviews Khawaja Shams, cofounder and CEO of Momento, about the concept of serverless computing and its impact on optimizing user experience. Shams explains that serverless computing eliminates barriers to entry, allowing for easier and cost-effective experimentation, which aligns with the original vision for cloud computing by simplifying configurations and adopting usage-based pricing models. This approach helps reduce costs by scaling down to zero when not in use, thereby enabling faster experimentation without the need for extensive initial investments. The conversation highlights serverless as a transformative evolution in cloud computing, enhancing accessibility and innovation.

The 1.0 release of the Momento Node.js Client introduces a range of features aimed at simplifying cache management and enhancing developer experience. This release emphasizes ease of integration and usage with asynchronous methods, allowing seamless interaction with existing asynchronous or synchronous Node.js applications. It introduces support for new collection data types, enabling the storage of lists, dictionaries, and sets alongside strings and byte arrays. By utilizing TypeScript’s type guards, the client ensures strong type checking and minimizes bugs by providing discoverable response types for various API calls. Additionally, it offers flexible logging through a customizable MomentoLogger interface and pre-built configurations for different environments to streamline setup and tuning. The release invites users to explore the new features via the Momento CLI and Node.js SDK, with extensive documentation and support available to address any user feedback or issues.

Momento, a company focused on providing a serverless caching service, has developed a streamlined key-value service that abstracts complex processes into three simple API calls. This approach, inspired by the operational excellence learned from building DynamoDB, has quickly matured Momento into a key player, earning partnerships with Google Cloud and clients like CBS and Wyze. Over the past six months, the team has prioritized customer feedback, discovering that users prefer an intuitive API with basic capabilities rather than a multitude of features. Responding to this, Momento has begun transforming its service by introducing collection data types (CDTs) such as dictionaries, sets, and lists, which address various workloads and enhance flexibility and scalability. As they continue to iterate and expand their offerings, Momento invites customers to share their needs and feedback on new features, fostering a collaborative development environment.

Momento has introduced three new cacheable collection data types—List, Set, and Dictionary—expanding its capabilities for managing data in applications. Lists maintain the order of inserted elements and allow duplicates, making them ideal for scenarios like chat history in games. Sets ensure unique element storage without regard to order, useful for tracking unique game players. Dictionaries allow selective data retrieval, improving efficiency by avoiding unnecessary data transfer, which is environmentally sustainable and cost-effective. These collection types support features such as auto-truncation, byte support, and individual Time to Live (TTL) settings, enhancing data management flexibility. Additionally, Momento's caching solutions scale seamlessly with serverless applications, accommodating sudden traffic surges without compromising performance. Currently available for Node.js, Python, .NET, and PHP, with plans to extend support to Go, Rust, and Java, Momento's new offerings are designed based on customer feedback and aim to further empower users with innovative features.

Sorted sets have been introduced as a new collection data type (CDT) by Momento, adding to the previously released dictionaries, sets, and lists. This CDT allows for the creation of ordered arrays of distinct elements based on numeric scores, automatically updating as elements change. In the multiplayer game Acorn Hunt, sorted sets can be utilized to manage leaderboards by tracking players' scores and automatically updating rankings as players collect acorns. They can also support rate-limiting features, such as limiting players to three uses of a special ability per game by decrementing scores and removing players from the set when their limit is reached. Additionally, the sorted set API offers versatility for various use cases beyond leaderboards and rate limiting, with commands to manipulate sets and manage time to live (TTL) for cached items. The feature is available in the Go SDK and supports multiple programming languages, enhancing developer experience by responding to user feedback.

This blog post, part of a series on DynamoDB data modeling, delves into optimizing secondary indexes to maximize cost savings and scalability. It emphasizes the importance of carefully selecting which attributes to project into a secondary index, advocating for minimal projection to reduce write unit consumption and storage costs. The concept of "sparse" indexes is highlighted as a means to efficiently filter data, and the pitfalls of "GSI overloading" are discussed, a practice that became obsolete after DynamoDB increased the limit of Global Secondary Indexes (GSIs) per table from 5 to 20. The article criticizes the "single table design" approach when it leads to unnecessary complexity and inefficiency, urging developers to prioritize flexibility and scalability. The author concludes by advising against GSI overloading and encouraging thoughtful index design to optimize costs and performance.

In a Software Daily podcast episode, host Alex DeBrie discusses with Momento Co-Founder and CEO Khawaja Shams the concept of serverless technology and how Momento exemplifies this through its caching solutions. The conversation highlights the shortcomings of traditional caching systems like Redis and Memcached, and how Momento Cache addresses these issues by offering a simplified API-based approach to cache creation and management. The discussion also covers the benefits and challenges of multi-tenant systems, as well as the importance of operational rigor and metrics in maintaining efficient serverless systems. Shams explains that serverless architecture allows users to access services through simple API calls, emphasizing Momento's ease of use where creating a cache involves merely passing a single parameter.