Does Aperture add latency on requests?
While Aperture does add some latency, it is a minimal one. The latency of requests to Aperture Cloud is in the order of 10 to 20 ms within the same region as your application. If you are hosting Aperture Agents yourself, the latency is in the order of 1-2ms.
Does Aperture reject requests immediately?
- Rate Limiter always accepts or rejects immediately.
- Scheduler queues the request for a maximum time up to the gRPC timeout of the request to Aperture with a small deadline margin.
If Aperture is rejecting or queuing requests, how will it impact the user experience?
Queuing requests should not affect user experience (apart from increased
latency). When it comes to rejecting requests, clients (whether it is front-end
code or some other service) should be prepared to receive
429 Too Many Requests or
503 Service Unavailable response and react
Remember, that some users receiving 503 means that the service is overloaded and Aperture is protecting your service from going into an unhealthy state.
How can Flow Labels be defined for workload prioritization or rate limiting?
- With SDKs, it's possible to explicitly pass Flow Labels to the Check call.
- In proxy- or web-framework-based Control Point insertion, most request
metadata is already available as Flow Labels, for example
- Already existing baggage is also available as Flow Labels.
- Proxy-based integrations can use a Classifier to define new Flow Labels.
See the Flow Label page for more details.
How does Aperture work with existing auto-scaling?
Rate limiting and caching allows services to stay performant while being cost-effective. Aperture enables developers to bring these capabilities to their service through a single convenient API.
Auto-scaling is used when the service is nearing peak capacity, despite rate limiting and caching. But scaling a service can be slow and expensive. While auto-scaling is happening, Aperture can protect the service from overload by queuing and prioritizing requests while staying within capacity. This also reduces the need to always stay over-provisioned.
Can you host Aperture in your infrastructure?
Yes, Aperture is fully open source and can be hosted on your infrastructure. There are two possible deployment options:
- Install just the Agents and connect the Agents to Aperture Cloud.
- Install the Agents and the Aperture Controller and connect to Aperture Cloud only for sending the telemetry.
Can the Aperture Agent run in a non-containerized environment?
Note: Aperture Cloud provides a hosted Agent for SDK integration, allowing you to use it by API instead of deploying your own Aperture Agents.
What are Aperture Agent's performance numbers?
The Aperture Agent is designed to be lightweight and performant.
With the following setup:
- 1 node Kubernetes cluster
- 1 Aperture Agent installed as a DaemonSet
- 1 policy with a rate limiter, a load scheduler and a flux meter
- 3 services in
demoappnamespace instrumented using Istio Integration
- 5000 RPS at constant arrival rate over 30 minutes
The following results were observed:
|CPU (vCPU core)||Memory (MB)|
|Aperture Agent||0.783 mean, 1.02 max||13.7 mean, 22.0 max|
|Istio Proxy||1.81 mean, 2.11 max||12.5 mean, 20.8 max|