An open source HTTP router to increase your network visibility

This article focuses on how Skipper increases network visibility and describes its advantages for both developers of scalable applications and operators of the infrastructure they run on.

Skipper increases network observability for developers and operators alike.

Subscribe now

Get the highlights in your inbox every week.

In my previous article, I introduced Skipper, an open source HTTP router and reverse proxy for service composition. This article focuses on how Skipper increases network visibility and describes its advantages for both developers of scalable applications and operators of the infrastructure they run on.

What is network visibility?

What is visibility, and what do we need to achieve to support it? Charity Majors provides a good answer on Twitter:

As an HTTP routing provider, Skipper wants to provide visibility to application developers. At e-commerce retailer Zalando, our developers in feature teams want to monitor their systems to understand failure rates (%), throughput (request per second, or RPS), and latencies (e.g., p50, p99, p999). Skipper library users who want to instrument their custom proxies can use Skipper's metrics package to instrument their proxies and create custom metrics.

Operators need observability to understand how the system behaves in general and at certain points in time, so they're ready when someone asks, "What happened yesterday at 2:30am?" or "How would this request be processed?" Skipper can help answer these questions by increasing backend applications' visibility.

Monitoring backend applications

For the purposes of this article, I'll define a backend application as an arbitrary HTTP service that is running behind the Skipper HTTP proxy.

Application owners running workloads want to monitor the four golden signals: latency, traffic, errors, and saturation. An HTTP router like Skipper is in a good position in the call path to provide three of the four—latency, traffic, and errors—for backends. But saturation is a complex target in practice; it has to be monitored at the node, control group, or application level, so it is out of scope for this article.

neural networks

Neural networks, as their name implies, are computer algorithms modeled after networks of neurons in the human brain. Learn more about neural networks from Algorithmia.

Recurrent neural networks, also known as RNNs, are a class of neural networks that allow previous outputs to be used as inputs while having hidden states.

The purpose of this project is to build and evaluate Recurrent Neural Networks(RNNs) for sentence-level classification tasks. Let's understand about recurrent neural networks for multilabel text classification tasks.

Artificial neural networks (ANNs), usually simply called neural networks (NNs), are computing systems vaguely inspired by the biological neural networks that constitute animal brains.

Convolutional Neural Network: How is it different from the other networks? What’s so unique about CNNs and what does convolution really do? This is a math-free introduction to the wonders of CNNs.