AWS Graviton Weekly # 156

To support these AI workloads, Pinterest is turning to Amazon custom silicon. The company plans to use AWS Trainium to host and run the large language models and vision-language models that power personalized visual search and AI-assisted discovery. Pinterest is also expanding its use of Graviton—which already powers roughly a third of its compute infrastructure—to run more of the systems that support discovery for hundreds of millions of people every month.

The new architecture delivers meaningfully fewer network devices between any two communicating servers, and fewer devices means fewer potential points of failure. It also means billions of dollars in cost savings and a network that can route around problems more dynamically, using more of its available capacity at any given moment.

In testing across most real-world traffic conditions, the new design moved data roughly a third faster than the hierarchical structures it replaces. The result is a network that is not just more efficient, but more reliable, and more powerful, for the customers who depend on it.

And the efficiency gains extend further. Significantly fewer networking devices means less power consumption, power that can instead be directed toward more compute capacity for customers. AWS expects to reduce electricity consumption for network equipment using the new design by 40% compared to its previous architecture, lowering CO2 emissions across a growing number of grids where it operates.