AllBugsFeaturesChangelogSearchNew post
Sign in with Fluxer
Features

Edit history

Earlier versions of Migrating performance-critical backend components to another language, newest first.

Current version | Edited by Rex
Changes
I propose performance-critical components be moved to a native language such as Rust, which can more efficiently make use of system resources and allow better scaling and reduced system requirements for hosting instances. Rust is a solid choice as it is already used for performance-critical clientside components, however any other native compiled language which offers efficient memory usage and multicore support would also work. Preferably, this is done sooner rather than later, as it's almost inevitable that fluxer will run into performance issues while trying to scale.Removed: ### Notes (optional)Removed: Removed: _No response_Removed: Removed: ### ChecksRemoved: Removed: - ☑ I searched for existing discussions and didn't find a duplicate.Removed:
Show

Migrating performance-critical backend components to another language

Problem

As fluxer becomes more popular and scales, the performance requirements for the backend on both the primary instance and self hosted instances will significantly increase. This is especially a problem with Typescript and Node.js, both of which struggle with high performance especially in terms of memory usage and multicore performance.

Proposed solution

I propose performance-critical components be moved to a native language such as Rust, which can more efficiently make use of system resources and allow better scaling and reduced system requirements for hosting instances. Rust is a solid choice as it is already used for performance-critical clientside components, however any other native compiled language which offers efficient memory usage and multicore support would also work. Preferably, this is done sooner rather than later, as it's almost inevitable that fluxer will run into performance issues while trying to scale.
Original by Rex
Show

Migrating performance-critical backend components to another language

Problem

As fluxer becomes more popular and scales, the performance requirements for the backend on both the primary instance and self hosted instances will significantly increase. This is especially a problem with Typescript and Node.js, both of which struggle with high performance especially in terms of memory usage and multicore performance.

Proposed solution

I propose performance-critical components be moved to a native language such as Rust, which can more efficiently make use of system resources and allow better scaling and reduced system requirements for hosting instances. Rust is a solid choice as it is already used for performance-critical clientside components, however any other native compiled language which offers efficient memory usage and multicore support would also work. Preferably, this is done sooner rather than later, as it's almost inevitable that fluxer will run into performance issues while trying to scale.

Notes (optional)

No response

Checks

  • ☑ I searched for existing discussions and didn't find a duplicate.