Introduction to gRPC

gRPC:

gRPC is an open source remote procedure call framework that can run any environment. It enables client and server applications to communicate transparently each other with help of HTTP/2. Mostly gRPC supports micro service oriented architecture. 

gRPC supports Java, C++, Dart, Python, C#, Ruby, PHP, Go, NodeJs and JavaScript etc…

for example, you can easily create a gRPC server in c++ with clients in Go, Python, or Ruby.

Features of gRPC:

1.    It supports the HTTP/2.

2.    gRPC messages are serialized using Protobuf by default in a binary format.

3.    gRPC supports client, Server & Bi directional streaming.

4.    Protobuf compiler will generate code automatically.

5.    It supports authentication, load balancing, logging and monitoring, timeouts& cancellation etc…

gRPC Streaming Types:

1.    Unary RPCs: It’s normal function call. the client sends a single request to the server and gets a single response back.

2.    Server streaming RPCs: The client sends a request to the server and gets response stream in the form of sequence of messages. For examples: Suppose if you watch videos in Netflix, YouTube etc. server can send the video content in the form of chunks.

3.    Client streaming RPCs: The client writes a sequence of messages and sends them to the server, for example: If you upload a files to server.

4.    Bidirectional streaming RPCs: Both client & server send a sequence of messages using a read-write stream. The two streams operate independently & they can read & write based on whatever order they like.

The gRPC works with protobufs concept. The Protocol buffers are a flexible, efficient for serializing structured data. The latest version of protobufs is proto3 version. The protobufs file extension is .proto type . It is Contract file which is used to communicate the client & server. 

For example : .Proto file Template 

Here Syntax is version of protobufs. if you don't do this the protocol buffer compiler will assume you are using proto2. 

Package: which helps to prevent naming conflicts between different projects.

RPC: we can define an RPC service interface method. The protocol buffer compiler will generate service interface code.

Message: What type request data you send/receive. (if you compare similar to C# Models class object property names while sending/ receiving data.)

gRPC recommended scenarios:

1.      gRPC is designed for low latency and high throughput communication. So we can use as a Microservices.

2.      IOT devices 

3.      Point to point real time communication because it’s support bidirectional streaming etc…

4.      Cloud services 

Pros:

1.      Inbuilt HTTP/2 protocol support.

2.      It’s support bidirectional streaming with high performance.

3.      The protocol buffer transfers the data in binary format.

4.      It supports multi-platform.

Cons:

1.      It will not support Browser.

2.      It is strictly contract-first.

3.      Debugging of gRPC response payloads is very difficult compared to Restful services.

For more details please visit below link:

For more detailed way to setup gRPC in .net core 3. please visit below link:

How to consume .Net core gRPC service into Python client using gRPC

Note:

I'm sharing this as per my knowledge. Anywhere, I did wrong. please let me know. I want to update myself.