2 min

Tags in this article


Google announces its new new “Argos” brand of video transcoding silicon.

YouTube requires such a huge video transcoding workload that Google has decided it needs to build its own server chips. We are all familiar with the specialised GPUs for graphics workloads. Many of us are also familiar with Google’s own TPUs (tensor processing units) for AI workloads. Well now the infrastructure team at YouTube claims to have created the first “VCU” or “Video (trans)Coding Unit”. These chips help YouTube transcode a single video into over a dozen versions, so that the site that remains bandwidth-efficient and hence profitable.

Meeting the video demand boom from the pandemic

Google detailed the new chip deployment in a blog post this week. The post was a sort of interview with Google’s Jeff Calow, the lead developer on the Argos project. “We’ve seen up to 20-33x improvements in compute efficiency compared to our previous optimized system, which was running software on traditional servers,” he said.

“”During the Covid-19 pandemic we saw surges in video consumption as people sheltered at home. In the first quarter of last year, we saw a 25 percent increase in watchtime around the world. And for the first half of last year, total daily livestreams grew by 45 percent,” he explained.

“Because we had this system in place, we were able to rapidly scale up to meet this surge. Practically, this meant that videos were available for viewers promptly after the creator uploaded them.”

The start of a new development journey

The VCU package is a full-length PCI-E card and looks a lot like a graphics card. A board has two Argos ASIC chips buried under a gigantic, passively cooled aluminum heat sink. There’s even what looks like an 8-pin power connector on the end because PCI-E does not meet the power demands of the VCU platform.

Carlow continues: “one of the key things that we’re doing in the next-generation chip is adding in AV1, a new advanced coding standard that compresses more efficiently than VP9, and has an even higher computation load to encode.” 

“As for me, I’ll be continuing my work on this project, developing future generations, which will keep me busy for a while.”