Lavinia Spatariu | Jan 7th, 2022
Can you imagine your personal information being stored in a DNA molecule or at the bottom of an abandoned mine? Seems unfathomable but the technologies are already being developed.
The need to preserve data is a fundamental human need. It’s natural to want to leave something behind, to make a difference in the world through innovation, to create something that we will be remembered for. In the same way in which our children are our most valuable legacy for the future, the data we generate will tell stories of events, habits, human and chatbot interactions and probably all the small things we find meaningful on a daily basis. Ancient civilizations used cave carvings and paintings for the same reason. All we need to do is make sure we find the most durable, cost-effective, sustainable and efficient storage solution.
DNA Storage
They say every movie ever made could fit into a space the size of a sugar cube and all the world’s data could fit into a cup of coffee. Storing data in artificial DNA molecules is a very appealing form of storage due to the fact that it is reliable, it can last thousands of years and it takes very little space. A DNA molecule consists of 4 chemical units called adenine, cytosine, guanine and thymine. The technology uses binary code to encode the information and the formatting can vary.
Synthesising and sequencing the DNA blocks takes little time but it is very costly. At present it costs $1 trillion to produce 1 petabyte of DNA data and as a result, this method may not be commercially feasible for some time yet. One other consideration is the required access frequency. This method is best suited for data which needs to be archived for a long time rather than accessed often or rewritten.
Specific maintenance conditions such as low temperature, the fact that DNA storage is currently more error-prone than traditional methods and the fact that once written the data is incredibly difficult to query mean that further research and development are needed before this option will be available at large-scale.
Pros
- Long-lasting;
- Compact;
- Reliable.
Cons
- High cost for read/write;
- Cannot be used for frequently accessed data;
- Low storage temperature required;
- Higher error rate than traditional storage methods;
- Difficult to query.
5D Optical Storage
Who would have thought that terabytes of data can be carved into small glass discs by using lasers? The durability and storage capacity of quartz glass means that data can be stored for billions of years and can survive significantly large-scale natural disasters (very high thermal stability of 1800 degrees F). Data is written and retrieved through a femtosecond laser system and is encoded in multiple layers. Density is also a big advantage as hundreds of terabytes can fit into a single monolithic 12-cm diameter disc. This is possible due to an innovative optical storage technique called birefringence. In simple terms, this refers to the way in which light is reflected through glass by making use of a 5-dimensional model. Another reason why optical storage is a great option is because the glass surface allows stored data to be erased and rewritten. This is a great advantage when thinking about efficient data lifecycle management.
While there are great benefits for using 5D optical storage in the future, there is no certainty that this option will be viable for large-scale commercial use. The technology is currently being developed by the research team at the University of Southampton, UK and it is unlikely that other institutions or organizations will be able to provide the same advanced optical technology thus making scalability challenging. The actual process of writing the data is expensive, takes time and requires a very specific skill set and expertise which is not easy to find or develop.
Pros:
- High durability;
- High density;
- High capacity;
- Erase & rewrite flexibility.
Cons:
- Slow development;
- Not easy to scale;
- Data writing is expensive.
Unique Location: Abandoned Mines
Underground cloud sounds like an oxymoron but it may become reality as abandoned limestone mines actually offer perfect conditions for hosting an efficient and secure data center. This type of environment provides consistent temperatures, humidity and is not affected by extreme weather. As Security is one of the main considerations for a data center, old mines make ideal locations for storing sensitive data used by governments, healthcare or financial organizations as their exact coordinates are kept secret from the public (although this also means that marketing the mine is not easy). Mines also have a single point of entry which makes access more secure.
On the downside, location planning is challenging due to the limited number of suitable mines in existence. For this reason, creating paired regions or availability zones for data redundancy would most likely be impossible. The initial setup is also cumbersome as installing the required infrastructure and networking is both laborious and costly.
Pros:
- Ideal conditions (temperature, humidity);
- Unaffected by extreme weather;
- Security (single entrance, secret location).
Cons:
- Difficulty in installing fibre optic internet connections;
- Secret location makes it difficult to market;
- Location choice and data redundancy options are limited.
Although each storage method has its own advantages and pitfalls, the reality is that storing data securely, reliably and cost-effectively is top-of-mind and a priority for all consumers, organizations and governments and keeping up-to-date with available technologies is critical. IoT, AI and the rise in analytical tools will make data the most valuable asset in the world and storing it efficiently will benefit not just us but generations to come.