In the new age of computers, many things have changed, and one important newly introduced aspect to both our lives and the economy is data collection. Companies like Google, Meta, and Amazon create significant amounts of data. Moreover, the amount of data produced by these world-class companies increases exponentially every year. For example, in 2023, 120 zettabytes of information were created and stored on the internet. To put that into perspective, a zettabyte is equal to a billion times a billion bytes. However, you need space to store the collected data. To power the need for storage, an enormous sector of data storage devices exists; however, the rapid escalation of data collection is getting critically close to the production of storage devices. This calls for us to improvise to continue gathering data at this scale. Currently, a dim yet promising light is being researched: DNA data caching.
Right now, magnetic tape is our best attempt at storing data; our photos, files, and apps are often kept in them. Though it has too many impediments, as mentioned, the data created is slowly surpassing the storage density of magnetic tape. This is partly caused by magnetic tape’s scale-up problems. Furthermore, magnetic tape degrades, and to preserve it, it needs to be in a room with a cool temperature, low humidity, and a stable environment. Even then, the tape used must still be replaced every now and then. Due to these conditions, additional expenses are added on top of the electricity to power the system and sustain it. Sure, magnetic tape is also improving every day, but the lack of sufficient backward compatibility limits it even more.
On the other hand, we have DNA. We have learned how to sequence and synthesize short pieces of single-stranded DNA. In nature, we see them in double-stranded forms, which makes them more resistant than single-stranded DNA. Since our artificial DNA is weaker than the naturally occurring form, reading and writing data may damage it. To counter the fact that reading and writing data are noisy molecular processes, the information being encoded is distributed among many bases, 1 bit per 60 atoms to be exact. In addition, we have far surpassed the first attempt by Microsoft, which wrote “HELLO” in 21 hours. Our ability to perform on DNA and costs relating to it have obviously been improving throughout the decades, keeping up with Moore’s Law. We also already transfer data through DNA, but only in such a way that it makes sense in the context of biology. Though, for actual data storage and transmission, we will need to write tens of thousands of bases worth of sequences that seem to make no sense. Scientists are planning to do this by adapting biological processes and fusing them with semiconductor technology to create high-density devices.
While this technology will be a great aspect of storing data when the principles of DNA are more clearly understood, it looks like we will stick with magnetic tapes until enough research has been done for us to begin scaling up and standardizing its usage.