For most of us, data storage has only gotten easier. Rather than messing around with computer hard drives with limited storage, floppy disks, and rewritable CDs or DVDs, today we just hit “save” on whatever file we’re working on and let it be whisked off to the cloud. It’s a no muss, no fuss approach to data storage that not only means we don’t run out of space, but that we can also access our files wherever we go. What’s not to love?
Well, it turns out there’s a fair bit. For one thing, we’re running out of storage space. Literally. With the 3.7 billion people who currently use the internet generating around 2.5 quintillion bytes of data each day, more and more data centers are needed to keep up with demand. By 2025, the world is on track to generate 160 zettabytes of data per year. That is more bytes than there are stars in the observable universe. Short of covering every square inch of land with data centers, we’re not going to be able to keep pace with this kind of increase. This means discarding data which could be invaluable — in some cases before we necessarily know whether or not it actually is.
“Simple storage of data isn’t necessarily all we can do with DNA.”
“If you look at large institutions like CERN, which runs the Large Hadron Collider, it generates petabytes of data each second the machine is running,” Nick Gold, VP of marketing at data company Catalog, told Digital Trends. “But there’s no way to store petabytes per second, so they have to throw away more than 90% of the data they generate. They’d love to keep all of that if there was a way to keep it.”
There’s also an environmental factor at work. According to one report, 17% of the total carbon footprint caused by technology is due to data centers. A current single data center can consume more power than a medium-size town. While companies like Apple have taken steps to offset this by embracing more sustainable energy sources, there are still reasons to seek a better alternative.
Fortunately, that’s exactly what some smart researchers around the world are working on. They are busy dreaming up (and, just as importantly, showcasing) some impressive next-gen storage technologies which could solve the world’s data problem in the years to come. And once and for all, too.
Welcome to the world of DNA storage
The idea of storing data in DNA sounds positively futuristic. In some ways, it’s just the opposite. A long time before computers existed (along with the humans needed to invent them), nature had figured out how to store enormous amounts of information in the form of DNA, the building blocks of life as we know it. Now, some investigators are embracing the idea of creating artificial gene sequences that use the four base pairs of DNA — A, C, G, and T — to represent binary bits of information.
Several years ago, researchers at the University of Ljubljana in Slovenia demonstrated that it was possible to encode pieces of computer code into the DNA of tobacco plants. They created a simple computer program and then spliced it into the genetic makeup of a tobacco plant; in essence, cloning it with the computer program still inside. Extracting the plant’s DNA and sequencing it resulted in the message “Hello World” popping up on a computer screen.
Since then, a team at Harvard University used CRISPR gene-editing technology to store a video in the form of bacterial DNA. The video, which more closely resembled a low-resolution GIF than the kind of high-res video most of us are used to watching today, nonetheless represented a significant advance. “We wanted to test whether the CRISPR-Cas system in bacteria could be used to capture complex information with a time component in living bacteria,” Dr. Seth Shipman, a neuroscientist at Harvard who led the experiment, told Digital Trends at the time.
In 2017, Shipman told me that there were no immediate practical applications for the work. “But hopefully on the near horizon,” he added.
“Our goal in the near future is to have a universal computing system …”
That “near horizon” may be now. For the past several years, the pioneering company Catalog has been working to commercialize DNA storage. Their pitch is that it might soon be possible to store the entirety of the world’s data in a space the size of a coat closet. That’s thanks to their approach in which data is coded into a synthetic polymer (rather than something living, such as a plant). This summer, the startup announced that it had managed to compress all 16 GB of English-language Wikipedia into a tiny vial of this material.
That’s not the only thing that makes it exciting, either. “Simple storage of data isn’t necessarily all we can do with DNA,” said CJ Huntzinger, director of communications at Catalog. “It’s not necessarily even the most attractive part of this whole platform. We’re seeing much greater opportunity in computation, and things that can yield even more value for humanity than just being able to store data in a very small volume.”
This includes breakthroughs in the ability to sift through DNA-based data. “When we put information into these DNA molecules, we’ve developed the theoretical framework for how we would manipulate those molecules to carry out basic computation, building up to a complex function,” Huntzinger continued. “Our goal in the near future is to have a universal computing system where we can translate any kind of Boolean logic and function into a set of molecular instructions — so we don’t have to pull information out of the DNA molecules into a digital medium before we compute on them.”
As this technology continues to advance, it offers a whole lot of potential in not just storage, but the ability to carry out things like pattern recognition across petabytes (one thousand million million) or even exabytes (one quintillion bytes) of data.
Storage gets colder
Researchers at the University of Manchester in the U.K. (the same university behind all-round wonder material graphene) have also developed some impressive next-gen storage technology. They’ve created molecules that could one day store hundreds of times more data than current hard drives in a significantly smaller form factor. The catch: it needs to be kept incredibly cold in order to function. However, while data centers would require supercooling technology to use it, they would also be able to significantly reduce their footprint; becoming cheaper to run, more energy-efficient, and less damaging to the environment.
“The advantage of our technology is extreme durability as we use quartz glass as a storage medium, which can survive disasters like fires or solar flares.”
“We are interested in making molecules that can store magnetic information,” said Dr. Nicholas Chilton, a senior lecturer and Royal Society University Research Fellow in the Department of Chemistry at the University of Manchester, told Digital Trends. “This could lead to a very useful technology if it works because molecules are very, very small; far smaller than existing magnetic materials that are used to store information. Using single-molecule magnets, we could potentially make data storage media that is 100 times more dense than current technologies such as HDDs and SSDs, which are facing their own limitations for data density.”
Single-molecule magnets can be “written” due to their ability to remember the direction of an applied magnetic field over relatively long periods of time after the magnetic field has been switched off. In 2017, Chilton and Manchester colleague Dr. David Mills prepared and studied the first “dysprosocenium” molecule: a dysprosium ion sandwiched between two five-membered carbon rings. Since then, two other groups at the university have built on this work by preparing additional dysprosocenium molecules, following the designs laid out by Chilton and Mills. This month, they published a new paper describing the work.
“The recent result published in Science shows magnetic memory up to 80 Kelvin, which is a significant milestone, as it is above the temperature of liquid nitrogen — which is a cheap and plentiful resource, unlike liquid helium,” Chilton explained. “However, this does not yet mean that data storage at liquid nitrogen temperatures is practical in molecules. The length of time that data can be stored at 80 K is on the order of seconds, and we need this to be on the scale of years for practical applications.”
At the cutting edge of this research, Chilton and Mills are experimenting with replacing carbon in the rings for the heavier element phosphorus. While their first result is not an improvement over all-carbon rings, the pair hope this research will provide insights into how to develop superior molecular magnets.
There are still bottlenecks to be solved, such as how to place these magnetic molecules on surfaces without affecting their performance, and how to protect the memory in each molecule from interacting with its neighbors. Nonetheless, while Chilton acknowledged that there remains “a long way to go” before this is commercially viable, these are exciting developments readers should keep their eyes on.
Get ready for 5D optical storage
Of course, if super cold storage isn’t exciting enough for you, how about the possibility of revolutionizing data storage by using lasers to carve terabytes of data into tiny glass discs? That’s the mission statement of researchers at the U.K.’s University of Southampton. In a quest to develop digital data storage that can potentially survive for billions of years, they have created a recording and retrieval process that relies on femtosecond laser writing.
“We are developing data storage technology primary for archiving and cold storage of large amounts of data, [such as] for data centers and cloud,” Dr. Peter Kazansky, a Professor in Southampton’s Optoelectronics Research Center, told Digital Trends. “One of our goals is to replace magnetic tape, which currently is used for such applications. The advantage of our technology is extreme durability as we use quartz glass as a storage medium, which can survive disasters like fires or solar flares, potentially harmful for data centers. Another advantage is that we use additional degrees of freedom for the data storage, which help to increase capacity.”
The storage solution is described as being five-dimensional. Information is encoded in multiple layers, including the usual three dimensions. However, it is also encoded in orientation and size of imprinted structures — thereby giving it five degrees of freedom for data storage. The storage allows for hundreds of terabytes per disc in data capacity. It’s also got thermal stability up to 1,800 degrees Fahrenheit. Compared to the vulnerability of magnetic tape, which lasts only for around one decade, this approach seems nigh-indestructible by comparison.
The University of Southampton’s work has garnered the interest of Microsoft. The initiative, called Project Silica, seeks to exploit 5D optical storage in glass for the first-ever storage technology designed and built for the cloud from the media up. “The main current bottleneck is the increase of writing speed,” Kazansky acknowledged.
Which will come out on top?
Right now, all three of these approaches are at various stages of future-ness. Each has its own unique timelines, challenges, and potential benefits and reasons.
Which will we ultimately see shape the future of data storage as we know it? It’s hard to know for certain. There’s still plenty more work to be done before today’s storage methods go the way of the 3 ½ floppy disk; replaced by something infinitely more science fiction in concept. But things are moving quickly.
Frankly, whichever of these fascinating approaches comes out on top, the real winners will be those of us who get to use the technology. And never have to delete anything for good ever again as a result.