The Google File System [pdf] (2003)

From good ol' Chat GPT:

The document is a research paper titled "The Google File System" and was published by Google in 2003. It describes the design and implementation of a new file system, called the Google File System (GFS), which was built to handle the challenges of processing large data sets using clusters of commodity hardware. The paper explains the architecture and components of the GFS, including its master and chunkservers, as well as its features such as atomic record append, data replication, and fault tolerance. The paper also discusses the performance and scalability of the GFS, and presents evaluation results of the system's performance. Overall, the paper presents GFS as a highly scalable and fault-tolerant file system that can be used to store and process large data sets.

Google File System (GFS) was a proprietary file system developed by Google and used in their data centers for over a decade before being replaced by Google's new file system, called Colossus. GFS was designed to handle the challenges of processing large data sets using clusters of commodity hardware and was used in many of Google's core services such as Google Search, Google Maps, and YouTube.

Colossus was developed as a successor to GFS, and it was introduced in 2010. It is a distributed file system that is based on the same principles as GFS, but it includes a number of new features and improvements such as increased scalability, improved fault tolerance, and support for larger file sizes. As of my last update GFS is not used by google as its primary file system, but it's important to note that GFS paper was a seminal work in distributed file systems, and it's concepts and ideas have been used in many other distributed file systems that have been developed since.