Showing posts from November, 2017

Lyft’s New Application Service Mesh

How best should an organization transition its monolith architecture into a set of microservices?  Apparently, it might not have to. Matt Klein, principal software engineer at car-sharing service Lyft, told The New Stack at PagerDuty Summit 2017 that a startup can develop its own monolith more easily than it can develop complex microservices. But with an underlying service mesh architecture, such as Lyft’s Envoy, that monolith can still be providing service-oriented functions to customers in the same way, and probably without service degradation.

High Performance, Small Footprint Edge and Service Proxy

Envoy Proxy is a high performance, small footprint edge and service proxy, designed for modern cloud-native architectures. Built by the engineering team at Lyft.

Connect, Manage, and Secure Microservices

Google, IBM, and Lyft released Istio an open source project that provides a uniform way to connect, secure, manage and monitor microservices.

The current release is targeted at the Kubernetes environment but support for other environments including virtual machines and cloud foundry is promised in the coming months.

Chaos Engineering Brings Stability to Your Distributed Systems

Jennifer Riggins from TheNewStack talks through Chaos Engineering and how it can be used to bring stability to large distributed container style systems.

I really like this articular in particle the Chaos Monkey toolkit they wrote for Docker called Pumba

Pumba is a chaos testing and network emulation tool for Docker.

with Pumba you can

Stop running Docker containers.Kill the send termination signal. Remove containers.Stop a random container once every ten minutes.Kill a MySQL container every 15 minutes.Kill random containers every 5 minutes.Pause the queue for 15 seconds every 3 minutes.
In chaos engineering, as you try to achieve stability at scale, you experiment following these four steps:

Define that ideal state of the system’s normal behavior.
Create a control group and an experimental group.
Introduce real-world wrenches, like changing servers.
Try to find the difference or weakness between the control and what is crashing.

The Future of Microservices Monitoring & Instrumentation

The future of microservices monitoring depends on what kind of solutions become standardized in the industry and what new features will we see in the future that will make your applications much better.

In this article Peter Marton takes a look at trends for 2018 and some interesting aspects of the article include

Vendor Neutral Agents
Distributed Tracing
Extracting metrics from OpenTracing API

Ultralight Edge Microservices Framework

Project Flogo™ lets developers build applications that run on edge devices and integrate them with IoT gateways. With the Project Flogo framework, you can extend the reach of core applications and infrastructure to edge devices to interconnect everything anywhere.