Solo.io, a cloud native software company, launched the first industry service mesh hub. The hub provides resources to help users adopt service mesh technology in hybrid and multi-cloud environments and features service mesh tools such as Istio, Linkerd, Envoy, AWS App Mesh, and HashiCorp Consul. Gloo Shot, Solo.io's chaos engineering tool for service mesh applications, has also been released and made available via the hub.
Service mesh technology assists with managing microservices by providing centralized tooling for common features shared across microservices, such as routing, load balancing, and logging and monitoring. A service mesh is generally implemented as a service proxy, called a sidecar, that handles all interservice communications.
Inspiration for the service mesh hub stemmed from the release of the open source project SuperGloo, which unifies service mesh lifecycle management. The hub builds off of SuperGloo to provide a centralized place for collaboration and management of service mesh technology. According to Solo.io founder and CEO Idit Levine:
The vision for Service Mesh Hub is more than technology, it is to support the end user through their entire adoption lifecycle from exploration, proof of concept, integration and through to production for their service mesh environment.
A key component of the service mesh hub is the dashboard, which provides a UI for users to search and install a range of available service mesh technology. The dashboard can be used to manage and configure meshes across cloud service providers and to monitor the health of all active meshes. The service mesh hub can be installed on Kubernetes with kubectl commands. The hub also provides an extensions catalog that enables third party vendors and open source providers to promote service mesh tools and make them available for installation via the hub.
Alongside the launch of the service mesh hub, Solo.io released Gloo Shot, a tool for chaos engineering. Gloo Shot leverages service mesh technology to conduct chaos experiments and currently supports service failure and latency based tests. Chaos experiments are designed with the Gloo Shot API and can be implemented against all upstream requests or a specified percentage. Experiments terminate based on a user defined failure condition, such as a Prometheus metric, or a time limit.
Gloo Shot is language agnostic and does not require library imports or code changes because it uses the service mesh interface to implement experiments. Gloo Shot leverages SuperGloo, a service mesh orchestration platform, in order to use the mesh without directly interfacing with it. Experiments can be run with Kubernetes kubectl commands and are executed in Kubernetes custom resources. When an experiment is run, Prometheus is installed and used for metrics collection during the test and can be used to review metrics when an experiment completes.
The service mesh hub documentation provides further information on getting start and contributing to the project. To learn more about Gloo Shot, visit the project documentation. An open source version of Gloo Shot is available at the service mesh hub and an enterprise version will be available in the near future.