https://deep.ai/publication/machine-learning-containers-are-bloated-and-vulnerable DeepAI AI Chat Log In Sign Up * [?] + Research + Glossary + APIs + Pricing + Sign Up + Log In Machine Learning Containers are Bloated and Vulnerable 12/16/2022 [?] by Huaifeng Zhang, et al. [?] [new_h] [new_r] 0 [?] [new_s] share Today's software is bloated leading to significant resource wastage. This bloat is prevalent across the entire software stack, from the operating system, all the way to software backends, frontends, and web-pages. In this paper, we study how prevalent bloat is in machine learning containers. We develop MMLB, a framework to analyze bloat in machine learning containers, measuring the amount of bloat that exists on the container and package levels. Our tool quantifies the sources of bloat and removes them. We integrate our tool with vulnerability analysis tools to measure how bloat affects container vulnerabilities. We experimentally study 15 machine learning containers from the official Tensorflow, Pytorch, and NVIDIA container registries under different tasks, (i.e., training, tuning, and serving). Our findings show that machine learning containers contain bloat encompassing up to 80% of the container size. We find that debloating machine learning containers speeds provisioning times by up to 3.7x and removes up to 98% of all vulnerabilities detected by vulnerability analysis tools such as Grype. Finally, we relate our results to the larger discussion about technical debt in machine learning systems. READ FULL TEXT * [profile-pl] Huaifeng Zhang 2 publications * [profile-pl] Fahmi Abdulqadir Ahmed 1 publication * [profile-pl] Dyako Fatih 2 publications * [profile-pl] Akayou Kitessa 1 publication * [profile-pl] Mohannad Alhanahnah 4 publications * [profile-pl] Philipp Leitner 11 publications * [profile-pl] Ahmed Ali-Eldin 7 publications [machine-le] page 3 [machine-le] page 11 Related Research research [?] 08/13/2021 VulnEx: Exploring Open-Source Software Vulnerabilities in Large Development Organizations to Understand Risk Exposure The prevalent usage of open-source software (OSS) has led to an increase... [new_h] [new_r] 0 Frederik L. Dennig, et al. [?] [new_s] share research [?] 08/27/2021 A Comparative Study of Vulnerability Reporting by Software Composition Analysis Tools Background: Modern software uses many third-party libraries and framewor... [new_h] [new_r] 0 Nasif Imtiaz, et al. [?] [new_s] share research [?] 03/12/2022 Characterizing and Understanding Software Security Vulnerabilities in Machine Learning Libraries The application of machine learning (ML) libraries has been tremendously... [new_h] [new_r] 0 Nima Shiri Harzevili, et al. [?] [new_s] share research [?] 09/18/2020 On the Threat of npm Vulnerable Dependencies in Node.js Applications Software vulnerabilities have a large negative impact on the software sy... [new_h] [new_r] 0 Mahmoud Alfadel, et al. [?] [new_s] share research [?] 01/11/2021 Understanding the Quality of Container Security Vulnerability Detection Tools Virtualization enables information and communications technology industr... [new_h] [new_r] 0 Omar Javed, et al. [?] [new_s] share research [?] 03/07/2023 Vulnerability Mimicking Mutants With the increasing release of powerful language models trained on large... [new_h] [new_r] 0 Aayush Garg, et al. [?] [new_s] share research [?] 06/28/2018 Machine Learning for Mathematical Software While there has been some discussion on how Symbolic Computation could b... [new_h] [new_r] 0 M. England, et al. [?] [new_s] share Sign up for DeepAI [ ] [ ] [ ] Sign up OR [github-ico]Signup with GitHub [linkedin-i]Signup with LinkedIn [google-ico]Signup with Google more signup options Already have an account? Login here x Login to DeepAI [ ] [ ] Login Forgot password? Click here [github-ico]Login with GitHub [linkedin-i]Login with LinkedIn [google-ico]Login with Google more login options Don't have an account? Signup here x Become a DeepAI PRO [prologo] $4.99/mo [greencheck] 500 generated images per month [greencheck] Private image generation [greencheck] Complete styles library [greencheck] API access [greencheck] Ad-free experience [greencheck] AI Chat access with Genius Mode * This is a recurring payment that will happen monthly * If you exceed more than 500 images, they will be charged at a rate of $5 per 500 images Subscribe Sign up No thanks, I'll do a one time payment Pay as you go $5 per 100 images [greencheck] Private image generation [greencheck] Complete styles library [greencheck] API access [greencheck] Ad-free experience Sign up [$5 USD ] Nevermind, I want to become PRO x SHARE [twitter-ic] [linkedin-i] [facebook-i] x DeepAI contact terms of service [twitter-ic] [linkedin-i]