I am working with Django and want to maximize the ability to handle web requests per second that interact with a db

Gordon Long 20

I am working with Django and want to maximize the ability to handle web requests per second that interact with a db. How would you recommend setting this up? I am using a postgresql db and can try to spin up more CPUs to have more worker threads, but so far struggling to get the application to work asynchronously (likely because of db requests). I have thus far max'd out at 30 requests/second but hoping to get much higher.

Laxman Reddy Revuri 800 Reputation points Microsoft Vendor

2024-11-07T17:48:44.33+00:00
Hi @Gordon Long
Thanks for the question and using MS Q&A platform.

Optimizing a Django application to handle a higher number of web requests per second, especially when interacting with a PostgreSQL database, involves several strategies.

Database Optimization

Indexing: Ensure that your database tables are properly indexed. This can significantly speed up query performance.

Query Optimization: Use Django's select_related and prefetch_related to reduce the number of queries. Avoid N+1 query problems by fetching related objects in a single query.

Statement Timeout: Set a statement timeout to prevent long-running queries from consuming resources.

Asynchronous Processing

Async Views: Use Django's asynchronous views to handle non-blocking I/O operations. This can help manage more requests concurrently.

Celery: Offload long-running tasks to background workers using Celery. This can help keep your web requests fast and responsive.

Caching

Database Caching: Use Django’s caching framework to cache query results. This can reduce the load on your database.

View Caching: Cache entire views or parts of views to serve repeated requests quickly.

Load Balancing and Scaling

Horizontal Scaling: Deploy multiple instances of your Django application behind a load balancer. This can distribute the load and improve your application's ability to handle more requests.

Vertical Scaling: Increase the resources (CPU, RAM) of your existing servers. However, this has limits and should be combined with horizontal scaling.

Connection Pooling

Database Connection Pooling: Use a connection pooler like pgbouncer to manage database connections efficiently. This can reduce the overhead of establishing new connections.

Profiling and Monitoring

Profiling: Use tools like Django Debug Toolbar and New Relic to profile your application and identify bottlenecks.

Monitoring: Continuously monitor your application's performance to catch and address issues promptly.

I hope this information is helpful.
Gordon Long 20 Reputation points

2024-11-08T13:05:39.1133333+00:00

Thanks for the response Laxman, I have been trying to get async views to work but apparently django wants database queries to be sychronous due to its ORM. I have tried using sync_to_async but keep running into issues with it. As for load balancing (especially Horizontal Scaling), are there easy resources for how to do this on Azure? Thanks for all of the great tips! Would you also recommend have a standalone database server which the separate django servers reach out to for db operations?
Laxman Reddy Revuri 800 Reputation points Microsoft Vendor

2024-11-12T16:50:08.33+00:00

Hi @Gordon Long
We haven’t heard from you on the last response and was just checking back to see if you have a resolution yet. In case if you have any resolution please do share that same with the community as it can be helpful to others. Otherwise, will respond with more details and we will try to help.
Gordon Long 20 Reputation points

2024-11-12T22:48:49.3466667+00:00

Hi Laxman, I haven't had time to try to implement these things yet, but I think this answers my questions for now. Thanks!

Accepted answer

Laxman Reddy Revuri 800 Reputation points Microsoft Vendor

2024-11-11T15:41:08.4166667+00:00
Hi @Gordon Long
Load Balancing (Horizontal Scaling) on Azure

Azure App Service:

Scale horizontally by adding more app instances.

Azure automatically handles load balancing between instances. You can configure auto-scaling based on metrics like CPU usage.
For more information about horizontal scaling:
Autoscaling guidance - Best practices for cloud applications | Microsoft Learn

Azure Front Door / Application Gateway: Azure Front Door: Global load balancer with SSL termination, URL-based routing, and WAF.
https://learn.microsoft.com/en-us/azure/frontdoor/front-door-overview
Azure Application Gateway: Layer 7 load balancing, SSL offloading, and traffic routing for web apps. What is Azure Application Gateway | Microsoft Learn

Standalone Database Server for Multiple Django Servers

Centralized Database:

Use a separate, standalone database server accessible by all app instances.

Ensure the database can handle connections from multiple servers (use connection pooling). Set up replication (read replicas) to distribute read queries and reduce load on the primary database. Implement caching (e.g., Redis) to offload frequent queries and reduce database load. Azure Database for PostgreSQL/MySQL: Provides automatic backups, scaling, and high availability. Azure Redis Cache: Integrate Redis for caching queries and storing sessions.
If the answer is helpful, please click "Accept Answer" and kindly upvote it.
Please sign in to rate this answer.

1 person found this answer helpful.
Laxman Reddy Revuri 800 Reputation points Microsoft Vendor

2024-11-14T16:18:04.2533333+00:00

Hi @Gordon Long
If this answers your query, do click Accept Answer and Yes for was this answer helpful. And, if you have any further query do let us know.

Laxman Reddy Revuri 800 Reputation points Microsoft Vendor

2024-11-15T08:25:12.7466667+00:00

Hi @Gordon Long
If this answers your query, do click Accept Answer and Yes for was this answer helpful. And, if you have any further query do let us know.

Report a concern
Sign in to comment

Use comments to ask for clarification, additional information, or improvements to the question.

Share via

I am working with Django and want to maximize the ability to handle web requests per second that interact with a db

0 additional answers

Your answer