What is a Service Mesh

How Does it Compare with Other Service Architectures

Comparing Service Mesh with Other Service Architectures

Service mesh is a response to complex microservice architectures that pose challenges for reliability and efficiency.
A service mesh is a layer of infrastructure to handle service-to-service communication without adding complexity to the applications themselves.
A service mesh creates point-to-point connections between applications to ensure that data and messages are passed reliably.
Dive deeper into what services meshes do and then compare three-tier applications, API gateways, and service mesh to help you understand when you need each.

What Does a Service Mesh Do?

Service-to-service communication for complex software architectures

Digitally transforming organizations have moved from monolithic applications to microservices: loosely coupled, independent, modular units of functionality. A microservices architecture allows for better scaling and more efficient production of features.

When monolithic applications are broken down and packaged into portable containers, the number of connections between applications and services grows exponentially. Monitoring and troubleshooting this complex system for reliability and performance can be a nightmare. This is where a service mesh comes in.

A service mesh is a relatively simple concept: think of it as DNS for microservices.

A service mesh is a layer of infrastructure to handle service-to-service communication without adding complexity to the applications themselves. When an application needs to send data and messages to other applications, the service mesh takes care of it, handling any failures.

A service mesh is usually implemented with proxies that coordinate the traffic between the services, making the communication more efficient and resilient.

A control plane determines where traffic goes and a data plane forwards the requests, retries as needed, and creates logs for debugging.

A service mesh uses dynamic routing rules to balance loads and determine the most reliable and fastest way to serve each request. Existing connections might be used to route traffic more effectively.

A service mesh makes microservices more scalable and dynamic. New services can be discovered and loads can be balanced automatically. The service mesh handles the encryption and authentication between services to prevent unauthorized access.

A service mesh also makes this layer more “visible.” You can configure policies for security. A service mesh captures metrics for each request and response. You can observe service performance, making it easier to find problems and resolve issues that are impacting the business.

Three-Tier Architecture

Monolithic applications are often not flexible enough for today’s business needs

Architectures based on the three-tier model can become inefficient, unreliable, hard to maintain, difficult to adjust for traffic spikes, and a challenge to deploy to different locations.

API Gateway Architecture

API gateways help route traffic “up and down” but not across services

API gateways & application delivery controllers create a central point to route traffic. This helps manage internal-external traffic, but not communication between services.

As the number of containerized applications and microservices grows, better service-to-service communication architectures are needed.

Service Mesh Architecture

Mesh networks have been used in IT for decades — here is another application

In one kind of service mesh, applications have lightweight proxies that manage the communication with other applications.

The platform interlaces communication between services in way that resembles a woven fabric, thus the term “mesh.”

The proxies add extra “hops” but if they are fast and light, they do not significantly affect performance. The added reliability, observability, and scalability are worth it.