DZone
Thanks for visiting DZone today,
Edit Profile
  • Manage Email Subscriptions
  • How to Post to DZone
  • Article Submission Guidelines
Sign Out View Profile
  • Post an Article
  • Manage My Drafts
Over 2 million developers have joined DZone.
Log In / Join
Refcards Trend Reports
Events Video Library
Refcards
Trend Reports

Events

View Events Video Library

Related

  • When One Giant Payload Must Serve Many Small Consumers: Designing a Scalable Fanout Service
  • How to Enhance the Performance of .NET Core Applications for Large Responses
  • Automatic 1111: Custom Sketch-to-Image API
  • On-Demand-Schedulers With MuleSoft CloudHub APIs

Trending

  • Building a RAG-Powered Bug Triage Agent With AWS Bedrock and OpenSearch k-NN
  • Amazon OpenSearch Vector Search Explained for RAG Systems
  • From 24 Hours to 2 Hours: How We Fixed a Broken BI System With Apache Airflow
  • Reproducible Development Environments, One Command Away: Introducing CodingBooth

Determine Payload Size Using Nginx

We wanted to evaluate which protocol is suitable for our micro-service architecture as our payload size was getting increased quite considerably.

By 
Milind Deobhankar user avatar
Milind Deobhankar
·
Jul. 20, 20 · Analysis
Likes (2)
Comment
Save
Tweet
Share
6.8K Views

Join the DZone community and get the full member experience.

Join For Free

Nowadays any application is API based or distributed where one request is not just served by one application but a series of applications either parallel or serial. Application talks to each other using various protocols like REST, RPC, WebSocket, and the payload formats varies from JSON / XML / Binary / propriety. 

Recently we wanted to evaluate which protocol is suitable for our micro-service architecture as our payload size was getting increased quite considerably. To determine that we want to know maximum payload size and minimum, as our payload size is directly proportional to items selected. 

Our tech stack is open source and our deployment is containerized polyglot micro-service architecture using docker swarm. Some of the services are using REST, and some are using gRPC (protobuff) and the payload is JSON, JSON+gzip, and binary. So to determine the payload size, we need a common proxy that will able to serve HTTP request and gRPC request. After looking at all tools, we decided to front our swarm service using Nginx so that we can capture payload size and uniquely logs each service request. 

overlay network

Let’s start how to configure Nginx for the same. The first challenge is to serve all types of the payload so we are going to use stream rather than HTTP. For each service, we are going to dedicate the incoming ports and upstream ports. There will be separate log files for each service as Nginx can only log the upstream IP address. If we combine the logs of all service, it will be difficult to identify based on IP as these will be container IP which can change if we restart or scale.

Nginx configuration:

Plain Text
 




xxxxxxxxxx
1
25


 
1
worker_processes  1;
2
events {
3
    worker_connections  1024;
4
}
5

          
6
stream {
7
    upstream stream_backend_service1 {
8
        server swarm_service1:8080;
9
    }
10
    upstream stream_backend_service2 {
11
        server swarm_service2:9090;
12
    }   
13
    log_format combined '$upstream_addr = Request Payload Size - $upstream_bytes_sent                           Response Payload Size - $upstream_bytes_received';
14
    
15
    server {
16
        listen            127.0.0.1:8090;
17
        proxy_pass        stream_backend_service1;
18
        access_log  logs/payloadsize.log  combined;
19
    }
20
    server {
21
        listen            127.0.0.1:8091;
22
        proxy_pass        stream_backend_service2;
23
    access_log  logs/payloadsize_sever2.log  combined;
24
    }
25
}



If you go to the respective logs following will get printed.

Plain Text
xxxxxxxxxx
1
 
1
payloadsize.log
2
10.0.1.2:8080 = Request Payload Size – 377 Response Payload Size – 180
3
payloadsize_server2.log
4
10.0.1.3:9090 = Request Payload Size – 837 Response Payload Size – 282


Our swarm_service1 is using REST over HTTP and swarm_service2 is using gRPC.
Hope this will help others who want to discover Nginx benefits.

Happy coding.

Payload (computing)

Published at DZone with permission of Milind Deobhankar. See the original article here.

Opinions expressed by DZone contributors are their own.

Related

  • When One Giant Payload Must Serve Many Small Consumers: Designing a Scalable Fanout Service
  • How to Enhance the Performance of .NET Core Applications for Large Responses
  • Automatic 1111: Custom Sketch-to-Image API
  • On-Demand-Schedulers With MuleSoft CloudHub APIs

Partner Resources

×

Comments

The likes didn't load as expected. Please refresh the page and try again.

  • RSS
  • X
  • Facebook

ABOUT US

  • About DZone
  • Support and feedback
  • Community research

ADVERTISE

  • Advertise with DZone

CONTRIBUTE ON DZONE

  • Article Submission Guidelines
  • Become a Contributor
  • Core Program
  • Visit the Writers' Zone

LEGAL

  • Terms of Service
  • Privacy Policy

CONTACT US

  • 3343 Perimeter Hill Drive
  • Suite 215
  • Nashville, TN 37211
  • [email protected]

Let's be friends:

  • RSS
  • X
  • Facebook