Google Ironwood is the company’s new chip for artificial intelligence operations

Google unveiled a new AI acceleration chip called Ironwood at its Cloud Next conference. It’s the company’s seventh-generation Tensor Processing Unit (TPU) and the first optimized for AI inferencing. The chip will be available to Google Cloud customers this year in two configurations: a 256-chip cluster and a 9216-chip cluster.

Ironwood delivers up to 4,614 TFLOPs of performance, has 192GB of dedicated RAM, and has a bandwidth of up to 7.4Tbps. It also includes an advanced SparseCore core designed to handle specific data. The chip’s architecture is designed to minimize data movement and latency on the chip, which helps reduce power consumption.

Ironwood will be a key part of Google’s strategy to integrate AI into its services. The chip will be integrated with the AI ​​Hypercomputer, a modular cluster of computing resources in Google Cloud that will allow for even more efficient processing of large amounts of data for AI.

Google announced Ironwood against the backdrop of increasing competition in the field of AI accelerators. While NVIDIA leads the market, other companies such as Amazon and Microsoft are also actively developing their own solutions. Amazon offers Trainium, Inferentia and Graviton processors through its AWS platform, and Microsoft has its Cobalt 100 chip for AI, available through Azure.

Google emphasizes that Ironwood is the most powerful and energy-efficient TPU to date, and its architecture allows you to efficiently run inferential AI models.