Application contact email(s)
ahmed.aley@gmail.com, huaifengzhang@outlook.com
Trademark and accounts
Contributing or sponsoring entity contact email(s)
ahmed.aley@gmail.com, huaifengzhang@outlook.com
Project summary
A container debloating tool that works with complex workloads
Project description
Most containers are bloated with unused files that inflate image sizes, increase provisioning times, and waste resources. Existing debloating tools are limited in scope and effectiveness, failing to address the widespread issue of container bloat at scale. BLAFS addresses bloat at the filesystem level by introducing a flexible debloating layer that preserves the layered structure of container filesystems. The debloating layer can be organized in different ways to meet diverse requirements including security hardening, enabled layered-sharing, and partial layer sharing.
Org repo URL (provide if all repos under the org are in scope of the application)
N/A
Project repo URL in scope of application
https://github.com/negativa-ai/BLAFS
Additional repos in scope of the application
No response
Website URL
https://github.com/negativa-ai/BLAFS
Roadmap
https://github.com/negativa-ai/BLAFS/issues?q=is%3Aissue%20state%3Aopen%20type%3AFeature
Roadmap context
We will be integrating BLAFS with multiple CI/CD platforms, supporting platforms such as Podman, and improving the deployability of the tool
Contributing guide
https://github.com/negativa-ai/BLAFS/blob/main/CONTRIBUTING.md
Code of Conduct (CoC)
https://github.com/negativa-ai/BLAFS/blob/main/CONDUCT.md
Adopters
No response
Maintainers file
https://github.com/negativa-ai/BLAFS/blob/main/MAINTAINERS.md
Security policy file
https://github.com/negativa-ai/BLAFS/blob/main/SECURITY.md
IP policy
Will the project require a license exception?
We are currently using MIT license, which we believe is in the allowlist. We are happy to switch to Apache 2.0, if necessary.
Standard or specification?
N/A
Why CNCF?
Our project was started as an academic research project. However, from talking to many people in different industries, we were told that this will be useful for their workloads, but they would like to see this as part of CNCF to be able to use it as it will mean there will hopefully be more maintainers than two academics. We would like to be able to create a community around the project, integrate it with as many CI/CD systems as we can, and make sure that the project remains open-sourced rather than, e.g., keeping it closed at a company.
Benefit to the landscape
We have done a very thorough evaluation on exisiting deblaoting tools here: https://arxiv.org/abs/2305.04641
Basically, our tool is the only tool we know of capable of supporting very complex and large containers such as Machine Learning containers and able to support layer sharing. Our tool is also much more reliable than all other exisiting tools we know of.
Cloud native 'fit'
BLAFS fits in the "Container Registry", "Automation & Configuration", and "Security & Compliance" landscape categories.
Cloud native 'integration'
N/A
Cloud native overlap
SlimToolKit: https://github.com/slimtoolkit/slim
Similar projects
Cimplifier which is an academic project recently open-sourced also by us: https://github.com/negativa-ai/MMLB/tree/main/external/cimplifier
Landscape
No
Business Product or Service to Project separation
At the moment, this project is unrelated to any product or service.
Project "Domain Technical Review"
No response
CNCF contacts
No response
Additional information
This project started as part of Huaifeng's PhD and has a full academic report with detailed evaluations: https://arxiv.org/abs/2305.04641
Application contact email(s)
ahmed.aley@gmail.com, huaifengzhang@outlook.com
Trademark and accounts
Contributing or sponsoring entity contact email(s)
ahmed.aley@gmail.com, huaifengzhang@outlook.com
Project summary
A container debloating tool that works with complex workloads
Project description
Most containers are bloated with unused files that inflate image sizes, increase provisioning times, and waste resources. Existing debloating tools are limited in scope and effectiveness, failing to address the widespread issue of container bloat at scale. BLAFS addresses bloat at the filesystem level by introducing a flexible debloating layer that preserves the layered structure of container filesystems. The debloating layer can be organized in different ways to meet diverse requirements including security hardening, enabled layered-sharing, and partial layer sharing.
Org repo URL (provide if all repos under the org are in scope of the application)
N/A
Project repo URL in scope of application
https://github.com/negativa-ai/BLAFS
Additional repos in scope of the application
No response
Website URL
https://github.com/negativa-ai/BLAFS
Roadmap
https://github.com/negativa-ai/BLAFS/issues?q=is%3Aissue%20state%3Aopen%20type%3AFeature
Roadmap context
We will be integrating BLAFS with multiple CI/CD platforms, supporting platforms such as Podman, and improving the deployability of the tool
Contributing guide
https://github.com/negativa-ai/BLAFS/blob/main/CONTRIBUTING.md
Code of Conduct (CoC)
https://github.com/negativa-ai/BLAFS/blob/main/CONDUCT.md
Adopters
No response
Maintainers file
https://github.com/negativa-ai/BLAFS/blob/main/MAINTAINERS.md
Security policy file
https://github.com/negativa-ai/BLAFS/blob/main/SECURITY.md
IP policy
Will the project require a license exception?
We are currently using MIT license, which we believe is in the allowlist. We are happy to switch to Apache 2.0, if necessary.
Standard or specification?
N/A
Why CNCF?
Our project was started as an academic research project. However, from talking to many people in different industries, we were told that this will be useful for their workloads, but they would like to see this as part of CNCF to be able to use it as it will mean there will hopefully be more maintainers than two academics. We would like to be able to create a community around the project, integrate it with as many CI/CD systems as we can, and make sure that the project remains open-sourced rather than, e.g., keeping it closed at a company.
Benefit to the landscape
We have done a very thorough evaluation on exisiting deblaoting tools here: https://arxiv.org/abs/2305.04641
Basically, our tool is the only tool we know of capable of supporting very complex and large containers such as Machine Learning containers and able to support layer sharing. Our tool is also much more reliable than all other exisiting tools we know of.
Cloud native 'fit'
BLAFS fits in the "Container Registry", "Automation & Configuration", and "Security & Compliance" landscape categories.
Cloud native 'integration'
N/A
Cloud native overlap
SlimToolKit: https://github.com/slimtoolkit/slim
Similar projects
Cimplifier which is an academic project recently open-sourced also by us: https://github.com/negativa-ai/MMLB/tree/main/external/cimplifier
Landscape
No
Business Product or Service to Project separation
At the moment, this project is unrelated to any product or service.
Project "Domain Technical Review"
No response
CNCF contacts
No response
Additional information
This project started as part of Huaifeng's PhD and has a full academic report with detailed evaluations: https://arxiv.org/abs/2305.04641