Performance Benchmark

What overhead does being on the service mesh incur?

Benchmark the performance of your application across different service meshes and compare their overhead.

performance tests run!


Which service mesh should I use and how do I get started?

Learn about the functionality of different service meshes and visually manipulate mesh configuration.

Benchmark, Configure, Adopt

  • - Compare apples-to-apples performance across service meshes.
  • - Track your service mesh performance from release to release.
  • - Understand behavioral differences between service meshes.
  • - Track your application performance from version to version.

Meshery provides

  • the user interface to authenticate users.
  • accepts and maintains the Kubernetes cluster config and context.
  • enables users to generate load against their applications.
  • collects the results of the performance tests from its load generator and stores in the cloud for historical viewing.
  • interfaces with the service mesh adapters dynamically and enables users to play with service mesh.

Service Mesh Adapters

Service mesh adapters provision, configure, and manage their respective service meshes.

  1. Meshery adapter for Istio (stable)
  2. Meshery adapter for Linkerd (stable)
  3. Meshery adapter for Consul (stable)
  4. Meshery adapter for Octarine (beta)
  5. Meshery adapter for Network Service Mesh (beta)
  6. Meshery adapter for Maesh (alpha)
  7. Meshery adapter for App Mesh (alpha)
  8. Meshery adapter for Kuma (alpha)

Service Mesh Interface (SMI) is a specification for service meshes that run on Kubernetes. SMI defines a common standard that can be implemented by a variety of service mesh projects and vendors. SMI’s aim for consistent APIs facilitates Meshery’s same goals, allowing for users and tools to flourish. As SMI was unveiled at KubeCon + CloudNativeCon Europe 2019, so did Meshery’s compatibility.

Layer5 is pleased to partner with Microsoft in support of SMI!

See Meshery at

Run Meshery

Step 1: Install and start

Using Docker, install Meshery on your local machine by running the following:

 sudo curl -L -o /usr/local/bin/meshery
 sudo chmod a+x /usr/local/bin/meshery
 meshery start        

Using Kubernetes, install Meshery on your cluster by cloning the Meshery repo:

 $ git clone; cd meshery        

Install Meshery on your cluster by running the following:

  $ kubectl create ns meshery
  $ kubectl -n meshery apply -f deployment_yamls/k8s

If you want to use a different namespace, please change the name of the namespace in the ClusterRoleBinding section appropriately.

Step 2: Login

Access Meshery in your browser at http://localhost:9081.


Fork me on GitHub This project is community-built and welcomes collaboration!

Benchmark Specification

Fork me on GitHub The Benchmark Specification is a common format for describing and capturing performance benchmark tests and results. This specification is currently being advanced through discussion in the Meshery project.


  • Does Meshery manage service meshes?
    • Yes, Meshery acts as a management plane to provision and manage the lifecycle of service meshes.

  • Why do I need to authenticate my social account to use Meshery?
    • Create your account - avoid having to reconfigure Meshery each time you deploy it. When you use a social account to authenticate to Meshery, your environment configuration persists from session to session.
      • Meshery sources from Mixer Prometheus adapter and uses Prometheus node-exporter.

    • Rank your deployment - results of performance tests are anonymously collected using the Service Mesh Performance Benchmark, so that Meshery may provide the public service of ranking the speed of your deployment against that of the rest of the world's users.
  • Why use Meshery?
    • Meshery is an open source, vendor neutral projects that facilitates testing across meshes.
    • Other performance benchmark tools are not packaged into a mesh testing utility, butare only a load-generators unto their own.

Mailing List

Stay up-to-date on service mesh content...