Here’s Google Cloud’s latest offering – coming next month in beta – Cloud Filestore. This new Google Cloud product is a managed file storage system for applications that need a file system interface & a shared file system. This means that companies can now easily run applications that need a traditional file system interface on the Google Cloud Platform.
According to Google, this is for folks out there who need “to create, read & write large files with low latency.” Which includes the community of film makers who are looking looking to render movies & create CGI images faster & more efficiently.
“So alongside our LA region launch, we’re pleased to enable these creative projects by bringing file storage capabilities to GCP for the first time with Cloud Filestore,” said Google.
Cloud Filestorebeta gives users a simple, integrated, native experience for standing up fully managed network-attached storage (NAS) with their Google Compute Engine & Kubernetes Engine instances.
Filestore promises a high throughput, low latency & high IOPS. The service will come in two tiers: premium & standard. The 1st will cost $0.30 per GB per month at speeds of 700 MB/s & 30,000 IOPS, no matter the storage capacity. Standard-tier Filestore storage will cost US $0.20 per GB per month, but performance scales with capacity & doesn’t hit peak performance until you store more than 10TB of data in Filestore.
Cloud Filestore also provides the reliability & consistency that latency-sensitive workloads need. One example is fuzzing, the process of running millions of permutations to identify security vulnerabilities in code. At Google, ClusterFuzz is the distributed fuzzing infrastructure behind Chrome & OSS-Fuzz that’s built for fuzzing at scale. The ClusterFuzz team needed a shared storage platform to store the millions of files that are used as input for fuzzing mutations.
To get started, create a Cloud Filestore instance, & seed it with the 3D models & raw footage for the render. Set up your Compute Engine instance templates to mount the Cloud Filestore instance. Once that’s set, spin up your render farm with however many nodes you need, & kick off your rendering job. The render nodes all concurrently read the same source data set from the Network File System (NFS) share, perform the rendering computations & write the output artifacts back to the share. Finally, your reassembly process reads the artifacts from Cloud Filestore & assembles it and writes into the final form.
Image Credit: Google
