Standardizing Service Mesh Value Measurement

SMP In-Action

Service Mesh Performance

Service Mesh Performance is a standard for capturing and characterizing the details of infrastructure capacity, service mesh configuration, and workload metadata.

Service Mesh Performance facilitates:

  • the ability to reason over the efficiency by which cloud native infrastructure is run, specifically in context of a service mesh and its network functions, include custom filters and/or protocol translators using WebAssembly or other.
  • benchmarking of service mesh performance
  • common vernacular and measurement for exchange of performance information from system-to-system and mesh-to-mesh
  • apples-to-apples performance comparisons of service mesh deployments and tooling to trend workload performance.
  • a universal performance index to gauge a service mesh’s efficiency against deployments in other organizations’ environments.

Using SMP, MeshMark provides a universal performance index to gauge your mesh’s efficiency against deployments in other organizations’ environments.

Service Mesh Performance (SMP) is a vendor-neutral specification for capturing details of:

  • Environment and infrastructure details
  • Number and size of nodes, orchestrator
  • Service mesh and its configuration
  • workload / application details
  • Statistical analysis to characterize performance

The canonical implementation of this specification is Meshery. The following snippet provides a further insight into the fact that the specification defines a common collection of statistical analysis to be calculated for every performance test:

message PerformanceTestResult {
              message Latency {
                double min = 1;
                double average = 2;
                double p50 = 3;
                double p90 = 4;
                double p99 = 5;
                double max = 6;
              } ...

Snippet of the Service Mesh Performance describing how to capture statistical analysis.





Project Overview





Presentation: Service Meshes: At What Cost?


KubeCon EU 2019 - Lee Calcote & Girish Ranganathan




Presentation: Discreetly Studying the Effects of Individual Traffic Control Functions


KubeCon EU 2020 - Lee Calcote & Prateek Sahu




Presentation: ServiceMeshCon-EU Conference


ServiceMeshCon-EU Conference - Otto van der Schaaf, Sunku Ranganath & Mrittika Ganguli

Contributors

Layer5
UEM
HashiCorp
CNCF
Red Hat
Fiserv
Intel
solarwinds
University of Texas
Layer5
UEM
HashiCorp
CNCF
Red Hat
Fiserv
Intel
solarwinds
University of Texas

Resources

SMP Community Meetings biweekly on Thursdays at 9 - 10 AM Central.

Example implementation

Meshery is the canonical implementation of SMP

Exchange performance tests and results in a standard format.

performance tests run!

Blog

Learn more about Service Mesh Performance

Service Mesh Offers Promising Solution for Cloud Native Networking

Service Mesh can be thought of as a next generation of Software Defined Networking (SDN) for the cloud. Here are the main approaches.

by The Newsroom

KubeCon+CloudNativeCon

Service Mesh Battle Stories and Fixes

by The Newsroom

Analyzing with SMP

Standardizing mesh analysis practices

by Lee Calcote

MeshMark explained

Through the eyes of a consumer

by Lee Calcote

Newsletter

Stay abreast of the progress of the Service Mesh Performance initiative

Mailing Lists

Engage in the Service Mesh Performance project. Join any of the mailing lists.




Service Mesh Performance is a Cloud Native Computing Foundation project
CNCF Logo

Created by Layer5 Logo and its open source community.