Google Cloud Seeks to Maintain Fair Pricing Amid Global Chip Shortage

Google Cloud is making efforts to set service prices at the right level amid a global chip shortage and rising costs, according to Country Director Indonesia, Karim Siregar. “If we look at it comprehensively, this impacts all hyperscalers. So, it is certainly not just Google. For our part, we will try our best to ensure the price is at the right level,” he said during a media briefing in Jakarta on Wednesday. He noted that the global computing supply capacity for the next one to two years has already been fully booked, allowing prices to be maintained. However, Karim did not dismiss the possibility of Google Cloud raising its service prices if chip costs continue to soar. Meanwhile, Vice President of Customer Engineering for Google Cloud Asia Pacific, Moe Abdula, stated that the global technology industry is currently facing increased demand for artificial intelligence accelerators such as Graphic Processing Units (GPU) and Tensor Processing Units (TPU). Google Cloud is striving to meet the demand for AI accelerators through partnerships and its own development efforts amid the component supply shortage. “We provide our self-developed TPUs to customers and we also offer them GPUs. We are a strong partner of NVIDIA and we provide those (GPUs). So there is no concern or issue at the moment in working to meet that demand,” Moe said. He explained that the development of TPUs stemmed from Google’s need to provide AI services that require much higher computing capacity than conventional systems. “What I can say is that at Google, we created the TPU a long time ago, when we realised that AI systems needed something different,” he said. Furthermore, Moe noted that GPUs are still required in some regions because not all data centres are designed to support TPUs. Therefore, Google maintains a close partnership with NVIDIA as a GPU supplier.