aperture

README

🥷 FluxNinja Aperture

Aperture is an intelligent load management platform for modern cloud applications.

⚙️ Features

• Adaptive service protection: Guard cloud applications against sudden overloads and cascading failures by scheduling incoming requests based on live service health metrics such as latency or error-rates. Without such a critical capability, it's impossible to ensure uptime and reliable operations of services and API endpoints.
• Workload prioritization: Optimize user experience and resource utilization by prioritizing workloads based on business value and urgency. Workload requests (for example, API calls) are labeled and prioritized using declarative policies and scheduled using weighted fair queuing. This ensures graceful degradation by preserving key user experience pathways even in the face of application failures.
• Distributed rate-limiting: Protect downstream services from overload by enforcing rate limits at the edge. Aperture's distributed rate limiter enforces limits based on fine-grained labels and is higher performance than alternatives based on centralized Redis deployments.
• Load-based auto-scaling: Aperture's policies are expressed as circuit graphs that continuously track deviations from service-level objectives and calculate recovery or escalation actions. Escalations such as auto-scaling can be easily configured to trigger based on load throttling, eliminating the need for costly over-provisioning.
• Automated load ramping: Safely and gradually ramp up load to a new feature or an API endpoint while monitoring for performance issues. Automatically ramp down the load when errors or latency spikes are detected.

🏗️ Architecture

• 🚦 Aperture Agents are typically deployed next to the services and provide high-performance flow-control capabilities. They are responsible for scheduling and rate-limiting incoming requests based on the policies defined in the Aperture Controller. In addition, they also collect service health and flow metrics.

• 🤖 Aperture Controller executes observability-driven control policies and configures Aperture Agents. The control policies are expressed as circuit graphs consisting of interconnected signal processing blocks. The Aperture policies provide a programmable way to tailor sophisticated control policies to match any application's needs.

🏁 Getting Started

🎮 Playground

To try Aperture in a local Kubernetes environment, refer to Playground docs.

🏎️ Installation

To install the Aperture system, please follow the Installation guide.

📖 Tutorials

To learn how to write Aperture policies, please read the Tutorials.

🎥 Videos

• Chaos Carnival 2023-Graceful Degradation:Keeping The Lights On When Everything Goes Wrong
• Graceful Degradation: When All Goes Wrong | Tanveer Gill | Conf42 Chaos Engineering 2023
• How Concurrency Limits Help Protect Against Cascading Failures
• Build Indestructible Applications with Aperture Flow Control

👷 Contributing

Reporting bugs helps us improve Aperture to be more reliable and user-friendly. Please make sure to include all the required information to reproduce and understand the bug you are reporting. Follow helper questions in the bug report template to make it easier. If you see a way to improve Aperture, use the feature request template to create an issue.

To contribute code, please read the Contribution guide.