Google Cloud announced that it will launch A3 virtual machine instances next month, equipped with NVIDIA GPUs to meet AI needs

According to news on August 31, Google Cloud announced at the Cloud Next Conference held today that it will launch A3 virtual machine instances next month. The biggest highlight of this new virtual machine instance is that it is equipped with NVIDIA H100 Tensor Core GPU can meet the needs of generative artificial intelligence and large language models.

According to reports, the A3 instance uses the 4th generation Intel Xeon Scalable processor, 2TB DDR5-4800 memory, and 8 NVIDIA H100 "Hopper" GPU. By NVlink 4.0 and NVSwitch, this instance implements 3.6 Bisection bandwidth in TBps. The new A3 supercomputer is designed to train and serve the most demanding artificial intelligence models and power generative artificial intelligence and large-scale language model innovation. It is said that this supercomputer can provide 26 Artificial intelligence performance of exaFlops.

In addition, Google Cloud also introduced the new TPU v5e at the press conference. This cloud TPU is the most cost-effective and accessible yet. These TPUs or custom ASICs are targeted at accelerating artificial intelligence and machine learning workloads. According to SDxCentral, TPU v5e doubles the training performance per dollar and improves the inference performance per dollar by 2.5 times compared to the previous generation

The above is the detailed content of Google Cloud announced that it will launch A3 virtual machine instances next month, equipped with NVIDIA GPUs to meet AI needs. For more information, please follow other related articles on the PHP Chinese website!