Google deployed an in-house designed chip to help Youtube videos look better and load faster than ever.
Called Video (Trans) Coding Units, or VCUs, the chips had been in the works for more than half a decade as Google engineers tweaked their optimization algorithms.
The new silicon is said to have helped YouTube cope with a huge spike in usage during the pandemic.
According to Google, the VCU gives users the best possible YouTube video quality on their device without consuming as much bandwidth as before, while helping Google lower infrastructure costs.
In a interview on the YouTube blogGoogle software engineer Jeff Calow, one of the engineers behind the new chip, said Google began work on the chips soon after first noticing increased demand for better quality videos several years ago.
To deliver the videos without penalizing the viewer, Google switched to more data-efficient video codecs such as VP9. However, VP9 required five times more compute resources to encode than the widely used H.264 format, which led the company to design the custom silicon.
Calow pointed out that the VCU is an order of magnitude faster than their previous transcoding solution that ran on traditional processors.
The first version of the VCU, which supports both VP9 and H.264 codecs, has now been deployed in Google data centers around the world, and adds that Google engineers are already working on the next generation. from VCU which supports the new and upcoming AV1 codec.
Going through Knowledge of the data center