Once Upon a Time…

For digital animation and VFX studios, I/O performance can directly impact the speed of production by dictating both server farm and creative team efficiency. In addition, modern rendering workloads need to embrace AI, machine learning and deep learning frameworks in order to automate larger portions of the creative process. To train these new models, studios are finding the need to provide high-speed access to content archives for high-throughput computation, which is a challenge for an industry where the largest data sets are typically kept on the slowest storage.

With VAST Data, animation studios and VFX shops can finally afford flash for all their working and render data. With affordable flash feeding the pipeline, each digital artist and all the AI-based applications can access the entirety of a studio’s stored data at speeds up to 10 times faster than HDD-based scale-out storage.

VAST Data breaks the decades-old tradeoff between storage performance and capacity with a new storage architecture that enables unlimited processing on affordable flash, at up to exabyte scale. VAST’s DASE architecture allows studios to scale performance and capacity independently. By disaggregating the capacity provided by QLC flash from the performance provided by the stateless VAST Servers, digital artists have greater access to assets, rendering pipelines run faster, AI learns more effectively, and the entire studio works more efficiently.