Feature Experimentation edge workers and agents

Edge workers are small programs or functions that execute at the edge of a network, closer to the user's location rather than in a centralized data center. This approach reduces latency and improves application performance by processing data near the source or end user.

Benefits of edge workers

• Reduced latency – Process data closer to you to minimize travel time, resulting in faster response times.
• Improved performance – Handle tasks like caching, data processing, and request handling at the edge for more efficient content delivery and to manage higher loads.
• Enhanced reliability – Distribute tasks across multiple edge locations to increase redundancy and reduce the risk of a single point of failure.
• Scalability – Deploy across various edge locations to accommodate growing user demands without overloading a central server.

Starter kit edge workers

Starter kit edge workers utilize a modified version of the Feature Experimentation JavaScript SDK. These starter kits are designed to be lightweight and compatible with various CDN providers. The primary goal of these edge workers is to intercept requests at the CDN level, make decisions on content variations, and cache the content for subsequent requests.

Key features

• Lightweight JavaScript SDKs – Optimized for performance by reducing the number of network requests and making necessary modifications for compatibility with different CDN providers.
• Custom code requirements – You need to write custom code to handle request interception, decision-making, and content caching.

Target audience

Edge workers are beneficial for product and engineering teams with the technical skills to write custom code and handle the complexities of CDN interactions. These edge works are especially advantageous for organizations with specific CDN needs. The following starter kit edge workers are available:

• Akamai EdgeWorkers
• AWS Lambda@Edge
• Cloudflare Workers
• Fastly Compute@Edge
• Vercel Functions

Optimizely Agent and Edge Agent

Optimizely also offers the following tools for Feature Experimentation, Optimizely Agent and the Optimizely Edge Agent. Each tool offers unique capabilities and advantages, making them suitable for different use cases.

👍

Beta

Optimizely Edge Agent is in beta. Apply on the Optimizely beta signup page or contact your Customer Success Manager.

Optimizely Edge Agent

Optimizely Edge Agent is a microservice that combines the capabilities of Optimizely's Feature Experimentation edge starter kit SDKs with the versatility and scalability of Optimizely Agent. It enables developers to conduct experiments entirely on the edge, leveraging the speed of CDN edge workers and the scalability of serverless architecture.

Use cases

• Edge computing capabilities – Lets you conduct experiments entirely on the edge, combining the speed of CDN edge workers with the scalability of serverless architecture.
• Optimizely Edge Agent has the following two modes of operation:
  • Edge mode (GET requests) – Operates as an edge SDK, making decisions at the edge before forwarding requests to the origin or retrieving content from the origin.
  • Agent mode (POST requests) – Handles feature flag evaluations and experiments using a scalable, serverless infrastructure.

Target audience

Optimizely Edge Agent is ideal for the following users:

• Developers looking to leverage edge computing capabilities to conduct experiments with minimal latency. Especially for those who want to combine the speed of CDN edge workers with the scalability of serverless architecture.
• Experimentation teams that require instant decision-making and persistent bucketing for users. The Edge Agent's ability to make decisions at the edge and maintain user bucketing information in cookies ensures a seamless and consistent user experience.
• Marketing teams that need to deliver optimized digital experiences quickly and efficiently. The Edge Agent's capabilities to run flicker-free experiments and serve cached content based on previous decisions help reduce latency and improve user experience.
• Organizations that require real-time optimization of user experiences at the edge. The Edge Agent's ability to conduct experiments entirely on the edge makes it suitable for businesses that need to optimize user experiences in real-time.

Optimizely Agent

Optimizely Agent is a standalone, open-source microservice designed to provide significant benefits over using Optimizely Feature Experimentation SDKs in certain use cases. It provides REST APIs that enable running feature flag rules, such as A/B tests and targeted flag deliveries.

The Optimizely Agent does not run on the edge. It is suitable for traditional server environments, service-oriented architectures, and scenarios where network centralization and standardized access across teams are priorities

Use cases

• Simplified integration – Optimizely Agent simplifies the integration process by providing a REST HTTP server. Engineers can implement the logic to make HTTP requests to the agent for flag evaluation on every client.
• High availability and scalability – As a microservice, Optimizely Agent is designed to be highly available and scalable, ensuring reliable feature flag evaluations and experiments even under high traffic conditions.
• Network Centralization – Centralizes the network connection to avoid multiple SDK instances connecting to Optimizely's cloud service. Only one cluster of agent instances connects to Optimizely for tasks like updating datafiles and dispatching events.

Target audience

Optimizely Agent is ideal for the following users:

• Developers who prefer using REST APIs for feature flag evaluations and experiments, avoiding the complexities of integrating SDKs directly into applications.
• Product managers who need a reliable and scalable solution for managing feature flags and experiments. Optimizely Agent offers high availability and flexibility, ideal for teams requiring consistent performance under high traffic.
• Companies with high traffic websites seeking a scalable and reliable solution for feature flag evaluations and experiments. Optimizely Agent's high availability ensures reliable experiments even during peak traffic.

Conclusion

Optimizely leverages agents and edge workers to enhance its Feature Experimentation platform. By running experiments at the network edge, Optimizely ensures a swift user experience. This capability lets you conduct experiments without sacrificing performance, maintaining speed even with complex testing scenarios.

Whether starter kit edge workers, Optimizely Agent, or Optimizely Edge Agents, edge workers are essential for users and organizations looking to boost application performance, reduce latency, and improve user experience across distributed networks. Whether you are a developer, a business with a global presence, or need to manage high-volume transactions, edge workers and agents provide the tools needed to optimize your digital experiences.