Creating Mass Digital Storage Technology Roadmaps

For over 20 years I have been involved in activities that create industry consortium roadmaps for digital storage technologies. Technology roadmap activities such as these were put on by various groups on various mass storage topics in the 1980’s and 1990’s and into the early 21^st century.

These were pre-competitive meetings that included folks from industry as well as academia who worked on new technologies that were being developed to enable advances in the capabilities of these storage devices. These roadmaps allowed companies and researchers to create an outline of when new technologies were likely to be implemented for these various storage technologies, and thus focusing their efforts.

I was an organizer of an annual industry roadmap meeting organized at Asilomar by the Institute of Information Storage Technology (IIST) out of Santa Clara University for several years that resulted in looking at historical technology developments and projections for increases in areal density (and thus storage capacity) and other important factors in the 1990’s and into the early 2000’s. I also participated in INSIC roadmap meetings and related activities such as standards with the HDD trade group, IDEMA.

In the early 2000’s I joined Roger Hoyt, then working at IBM, as co-chair on a Mass Storage Roadmap Effort with NEMI (now INEMI, International Electronics Manufacturing Initiative), which published mostly semiconductor and packaging roadmaps every couple of years. Roger was chairing this activity since about 1996. The roadmap group included experts on magnetic digital storage (HDDs and magnetic tape) as well as optical storage and solid-state storage.

In 2022, Roger and I moved the former INEMI roadmap to the IEEE Technology Roadmap Initiative. IEEE took over the former ITRS semiconductor roadmaps in 2015-2016 (now called the IEEE IRDS) and has expanded this effort to include a heterogeneous integration roadmap (2.5D and 3D semiconductor device integration), and international network generation roadmap, neurotechnology for brain-machine interfacing, a roadmap on wide bandgap semiconductors and soon a robotics roadmap and a power electronics roadmap.

Roger and I are currently chairing the IEEE Mass Storage Roadmap working group with experts on HDDs, magnetic tape, optical storage and solid-state storage. We also hope to have a section on a DNA storage roadmap for the 2023 IEEE Mass Storage Roadmap white paper. Last week I had a chance to talk with folks from the IRDS roadmaps on non-volatile memories from the Beyond CMOS and More Moore roadmap working groups. We want to make sure to coordinate our roadmap with the work that they have done.

Digital storage (and memory) enables many of the technologies that drive our society and economy. For instance, various AI model training needs lots of memory and storage for the data used in training. After a model has been trained then weightings from the training need to be stored and be available for model inference. Higher resolution video and audio, scientific and engineering data are demanding higher storage capacities as well as performance.

Creating roadmaps for these storage technologies may help in directing work on these technologies as well as informing people developing dependent technologies for when capabilities that they need for their applications will be available and affordable. We plan to release the IEEE Mass Storage Roadmap in the Fall of this year and I will write about those results when it is released.

Technology roadmaps help technologists and users in planning their activities. Digital storage and memory enable many technologies such as AI. The IEEE Mass Storage Roadmap will provide insights on developments in HDD, magnetic tape, SSD and optical storage technologies.