Shipwright: A Human-in-the-Loop System for Dockerfile Repair

Docker images are like Lego blocks for software developers, providing a standardized way to package applications and their dependencies. However, these images can become bloated with unnecessary files and instructions, leading to slower performance and increased security risks. To address this issue, researchers have developed various techniques for optimizing Docker images, similar to how a carpenter might use different tools to optimize the construction of a building.
In this study, we surveyed 31 Docker image optimization techniques and analyzed their effectiveness. We found that some techniques, such as removing unnecessary files and instructions, can significantly reduce the size of an image without affecting its functionality. Others, like using compression tools or optimizing container runtimes, can also help improve performance.
We also observed that different techniques are better suited for different scenarios, much like how a carpenter might choose different tools depending on the type of building they are constructing. For example, removing unnecessary files is most effective in images with a large number of small files, while optimizing container runtimes is more useful for improving performance in long-running applications.
Overall, our study provides a comprehensive overview of the various techniques available for optimizing Docker images, helping developers make informed decisions about which tools to use when building their applications. By using these techniques, developers can create faster, more secure, and more efficient applications, much like how a skilled carpenter can build a sturdy and comfortable home using the right tools.