Skip to content

A set of utilities for building, deploying and operating a high performance NFS Cache in Google Cloud. This is designed to be used for certain HPC and burst compute use-cases where there is a requirement for a high performance NFS cache between a NFS Server and its downstream NFS Clients.

License

Notifications You must be signed in to change notification settings

GoogleCloudPlatform/knfsd-cache-utils

Repository files navigation

knfsd-cache-utils

Note: This is not an officially supported Google product. NFS re-exporting is only suited for very limited use cases. You should be extremely familiar with NFS, FS-Cache and the other components used in this repository before deploying and/or using these utilities.

For questions & comments, reach out to knfsd-cache-utils@google.com.

Overview

This repository contains a set of utilities for building, deploying and operating a high performance NFS cache in Google Cloud. It is designed to be used for certain HPC and burst compute use-cases where there is a requirement for a high performance NFS cache between a NFS server and its downstream NFS clients.

This solution is based on existing Linux kernel modules, including nfs-kernel-server which supports NFS re-exporting and cachefilesd (FS-Cache) which provides a persistent cache of network filesystems on disk.

The solution works by mounting NFS exports from a source NFS filer (typically located on-premise) and re-exporting the mount points to downstream NFS clients (typically in Google Cloud).

Performing this re-export provides two layers of caching:

  • Level 1: The standard block cache of the operating system, residing in RAM.
  • Level 2: FS-Cache. A Linux kernel module which caches data from network filesystems locally on disk. When the volume of data exceeds available RAM (L1), the data is cached on the disk by FS-Cache. In this deployment, we use Local SSD's for the L2 cache, although this could be configured in a number of ways.

Using the deployment scripts in this repository, we further extend this architecture by creating multiple NFS proxies in a Managed Instance Group and utilising a Internal TCP Load Balancer to manage connections between the NFS clients and NFS cache.

The NFS caching solution is collectively referred to as Knfsd in this repository.

Building and Deploying

This repository is broken down into two key sections:

  1. Build of Knfsd Image
  2. Deployment and Operation of Knfsd Cluster on GCP

You should start with the Build of Knfsd Image. Once built, you can use this image and the deployment scripts to deploy and operate a Knfsd Cluster on GCP.

Testing the Image

There is a basic suite of smoke tests that can be run after building a new image. These tests check for common configuration issues such as the correct kernel version, cachefilesd is enabled and active, etc.

About

A set of utilities for building, deploying and operating a high performance NFS Cache in Google Cloud. This is designed to be used for certain HPC and burst compute use-cases where there is a requirement for a high performance NFS cache between a NFS Server and its downstream NFS Clients.

Resources

License

Stars

Watchers

Forks

Packages

No packages published