After Hard Drives—What Comes Next?

Abstract

There are numerous emerging nonvolatile memory technologies, which have been proposed as being capable of replacing hard disk drives (HDDs). In this paper, the prospects for these alternative technologies to displace HDDs in 2020 are analyzed. In order to compare technologies, projections were made of storage density and performance in year 2020 for both hard disks and the alternative technologies, assuming the alternative technologies could solve their remaining problems and assuming that hard drives would continue to advance areal density at a pace of about 40% per year, which would result in a two-disk 2.5-in disk drive that stores approximately 40 Terabytes and costs about $40. A major conclusion of the study is that to compete with hard drives on a cost per terabyte basis will be challenging for any solid state technology, because the ITRS lithography roadmap limits the density that most alternative technologies can achieve. Those technologies with the best opportunity have a small cell size and the capability of storing multiple bits per cell. Phase change random access memory (PCRAM) and spin transfer torque random access memory (STTRAM) appear to meet these criteria. PCRAMs are being marketed by at least one supplier and therefore appear to be closer to practical realization. On the other hand, STTRAMs would appear to have a performance edge assuming they, too, can be brought to market with multiple bits per cell. Although there are technologies that are not limited by the lithography roadmap and thus have greater areal density potential, they tend to be further from practical realization.