By Irene Choi
Photo by Sangharsh Lohakare on Unsplash
From writing on papers to saving on computers, our way of storing data massively changed over the past. Hardwares, floppy drives, and SSDs do have benefits compared to data storage on paper; for example, they are more durable of outside changes in condition and they take up less space and time compared to writing one by one. However, would it be our best method of data storage in the long term? What might be other alternatives?
How does the current system work?
First of all, let’s look at the problems of the current system. We typically use data storage centers to store any crucial data that needs to be protected for a long period of time. Data storage centers serve three main purposes: compute, storage and networking. Essentially, they process the data to run the application, physically store the data with backups, and connect the data with different centers when necessary. Storing the data in the long term is especially important for corporations or organizations which requires frequent recall of information stored. The equipment in large scale data storage centers allow organizations to continue storing information. (Networkworld, 2020)
What are the downsides of the current system?
Not Robust
Storage centers not only require technical equipment to protect the information, but also require extensive facilities to keep the hardware and software running: including power subsystems, ventilation and cooling systems, backup generators, uninterruptible power supplies (UPS), and cables to connect external network operations. These facilities are open to the impact of external changes in condition, and can be consumed - meaning they have to be changed or cleaned constantly while using. This makes it less cost-effective to use in the long term.
2. Low information density
Storage centers also require a large amount of space to store the information. Compared to the space they take up, however, the information they can hold isn’t much. This makes the use of space ineffective and unnecessary; this also contributes to a higher cost.
3. Each media requires a special reader
Just like how a key and a lock works in daily lives, each media composed of data requires a specific reader to analyze or recall the data. The space and cost to buy or build every reader can be negligible if the media is small enough; however, because we need a great amount of information to be stored, the money and space we need can’t simply be ignored. This contributes to a greater inefficiency of using money, especially by profit seeking corporations.
Data storage center in Oregon
As shown in the picture, about 5700m^2 big data storage centers can contain 1 exabyte of information (100,000,000 gigabyte). Considering that a middle size soccer field can be 5000m^2, the data storage center can be interpreted as taking up a great amount of space. (Matange, 2021)
DNA Data Storage?
Understanding the limitations of current data storage techniques, why is DNA data storage an effective alternative? First of all, data storage in DNA can store up to 1018 bytes per mm^3, which is about 6 times denser than the densest storage method available today. This increase in density allows the preservation of data for a long time at low energy costs. DNA storage method also makes it easier to replicate the data using the PCR technique. The cost and time, which is the problem currently faced, is expected to improve in the future.
There are three main types of DNA data protection: archival storage method, working storage method, and short-term dynamic storage. Archival storage method, one of the long-term storage methods - encapsulates the DNA to protect the data that is not often accessed. The working storage method is a short term storage method which freezes or dries the data, and the short-term dynamic storage method, used for the most frequently accessed data, secures the data in a liquid form. Essentially, all three data storage methods secures the data by transforming the original binary data to a DNA code (G/C = 0, A/T = 1). This not only increases the efficiency of data protection, but also allows better data replication as DNA has a characteristic of self-replication. (Ceze, 2019)
DNA storage method diagram
Limitations?
Although DNA storage methods are expected to develop, there are limitations to current DNA storage methods that limits our ability to fully utilize the system. For example, the methods are not fully stable, as the DNA is still exposed to external conditions and can be damaged by small condition changes in humidity, temperature, and many other factors. Moreover, DNAs cannot fully secure the data if the data exceeds the 10*10 binary form. Thus, it is yet unable to contain and replicate the data we need. These limitations are to be addressed, and with future research, DNA data storage methods can be a better means for data storage.
References
Networkworld. What are data centers? How they work and how they are changing in size and scope. 2020. [online] https://www.networkworld.com/article/3599213/what-are-data-centers-how-they-work-and-how-they-are-changing-in-size-and-scope.html [Latest accessed: 15th May]
Matange. DNA stability: a central design consideration for DNA data storage systems. 2021. [online]
https://www.nature.com/articles/s41467-021-21587-5 [Latest accessed: 15th May]
Ceze. Molecular digital data storage using DNA. 2019. [online]
https://www.nature.com/articles/s41576-019-0125-3 [Latest accessed: 15th May]
Comments