Hu Yoshida

Stars in the Universe and Hippocampus Storage

Blog Post created by Hu Yoshida Employee on Apr 11, 2016


Roberto Basilio, our VP of next Gen Products shows the following chart to illustrate the fact that we are rapidly losing the ability to store data, and the challenge will be to determine what data we store versus what we throw away. The rate at which we could possibly build storage devices will never be fast enough to catch up to the rate at which we generate data. Just the click of an iPhone can create MBs of digital data in a thousandth of a second. Today we can only store about a third of the digital data that is created. In four short years, by 2020, we will only be able to store 15 % of this data, and the rest falls off the edge.


The other day I was discussing this with Suresh Nair, our country manager from Indonesia and we were speculating about what might be possible if we had a “hippocampus memory system”. Hippocampus is the area of our brain that stores memory. (There are two hippocampi, one on each side of our brain, and damage to the hippocampi causes loss of memory).


Our minds seem to be able to store an unlimited amount of information. Our brain gets input from our five senses and processes the input to create information, which is selectively stored in short or long term memory, depending on factors that are consciously or unconsciously important to the individual. The five senses also act as meta data for retrieval. A certain smell or sound can trigger the retrieval of a memory. Memories can also trigger autonomic reactions like batting a ball or riding a bicycle. We can also communicate and build on the memories that we share with others.


The datacenter of the future could become a self-analyzing brain, built on an open infrastructure platform that delivers the services that can grow and scale to optimize business and services with greater efficiency and lower costs. There will be analytic engines that sit on the edge and glean information from data before it is lost like the tips of our fingers tell us what we felt. Or maybe it will be able to recreate the data, based upon shards of information. The data center will be a distributed cloud, increasingly software defined, open source, ubiquitous, borderless, scalable, and self healing. It will use telemetry to sense information and telematics to control remote devices. Technologies like block chain, distributed ledgers, might be used for global consistency and transparency.


In the future the focus will be on storing information and not data, not as individuals but as a collective of individuals.  Instead of data centers we will have an autonomous information cloud.


Please share your views on the future of data centers, storage systems, and information.