Forward-looking: Hard disk drive technology dates back to the 1950s and still relies on decades-old I/O standards. Seagate plans to usher hard drives into a PCIe-based future a shift driven by the growing demand for faster storage in artificial intelligence training and inferencing, particularly in massive, energy-hungry data centers. Seagate recently introduced a novel approach to HDD development and evolution, bringing magnetic storage and NVMe together in a new storage technology. The manufacturer wants to simplify and improve data storage for AI companies by providing a significantly cheaper solution than solid-state drives.Hard disk drives will never reach the same performance level as SSDs, but Seagate says it can improve them for more efficient use in next-gen data centers. The company is pioneering a "transformational" solution by adding NVMe compatibility to high-capacity hard drives.Seagate explained that by adopting the PCIe-based protocol, NVMe hard drives could eliminate the need for proprietary silicon tied to traditional SAS/SATA interfaces, host bus adapters (HBAs), and controller architectures unsuitable for AI workloads. The new protocol promises high throughput and low latency, though it's hard to imagine a future where hard disk drives move data at gigabytes per second.These NVMe hard drives will have no HBAs or custom controllers, but they will retain the same SAS/SATA connector for improved compatibility. The new tech will simplify the deployment of AI storage solutions, while a single NVMe software driver stack should make HDDs and SSDs work together more efficiently.Seagate is also touting the ability to directly connect NVMe drives to GPUs, bypassing the eventual bottlenecks created by traditional data routes through the CPU. The new architecture is more efficient, so AI models can process massive data sets with reduced I/O delays. Furthermore, a novel NVMe over Fabrics (NVMe-oF) technology should bring HDDs into fully distributed, scalable storage architectures. // Related StoriesSeagate tested the potential impact of NVMe hard drives through a proof-of-concept integration between custom HDDs, NVMe SSDs, Nvidia BlueField data processing units (DPU), and the AIStore software. While the company hasn't provided specific metrics regarding performance, it claims that the new platform provided some key advantages for AI workflows. Direct GPU-to-storage communication via DPUs reduced storage-related latencies while eliminating legacy SAS/SATA overhead in the simpler system architecture.Furthermore, the AIStore software helped optimize data caching for better AI model training performance, while NVMe-oF integration proved its usefulness in multi-rack storage clusters. Seagate said that NVMe HDDs would offer 10 times more efficient embodied carbon per terabyte, four times more efficient power consumption, and "significantly lower" cost per terabyte than SSDs. The company is now working to scale its HAMR-based Mozaic 3+ platform and develop higher-capacity hard disk drives while exploring NVMe storage solutions with its partners.