Analysis: AI Data Growth Outpacing Compute, Storage Emerging as New Bottleneck in AI Infrastructure

BlockBeats News, August 15th, a recent analysis by Western Digital pointed out that as the scale of artificial intelligence applications rapidly expands, AI data center construction is shifting from a simple GPU computing power competition to a data storage capacity competition, and storage planning has become a core part of AI infrastructure.

The article cited IDC's forecast, stating that by 2030, the global annual data volume will reach 718ZB. Data generated by AI systems will not disappear when the computing tasks end. Training data, model checkpoints, embedding vectors, inference logs, prompts, output results, and evaluation data will continue to accumulate.

Western Digital stated that many current AI infrastructure plans focus excessively on GPU utilization but overlook data deposition throughout the AI lifecycle. The data generated during training and inference processes will become important assets for model iteration, quality assessment, and compliance audits, and storage costs will directly impact the long-term operational efficiency of AI systems.

As data scales enter the PB or even EB level, a single storage architecture is insufficient to meet the demands. Enterprises need to adopt a tiered storage strategy, using high-performance flash storage for training and real-time inference, and using high-capacity HDDs and object storage for long-term data retention, historical records, and infrequent access scenarios.

The analysis suggests that in the future, a key metric for AI infrastructure competition will not only be the number of GPUs but also the cost of data storage per PB, energy consumption, recovery efficiency, and data lifecycle management capabilities. If enterprises continue to view storage as a post-computing ancillary part, they may face issues such as uncontrolled data costs and decreased model iteration efficiency.