MEMO — Decentralized Amazon Cloud Storage
The story of modern cloud storage must begin with Amazon Cloud Storage. As the originator of cloud storage, Amazon Cloud Storage has been transformative to this day, allowing data storage to move away from the long construction cycles and large capital investments required as making bread from wheat, and opening up the era of “out-of-the-box” cloud storage with no fixed investment.
Looking back at recent history, the creation of a centralized infrastructure is one of the key signs of modern civilization. The construction of infrastructure for water and electricity resources has made a huge difference to people’s lives. The convenience of ready access to these resources has eliminated the need for families and individuals to spend a lot of time and effort on accessing them, freeing up hands and freeing people to spend more time on more meaningful work.
The construction of a data storage infrastructure heralds the arrival of a higher order digital civilization. Amazon offers a cloud storage service that allows users to access their data from anywhere in the world using the internet without the need to configure any hardware or to consider any operation and maintenance expansion issues. Behind the convenience of the service is Amazon Cloud Storage’s technical drilling and centralized maintenance of storage facilities. Amazon Cloud Storage has always been at the forefront of the industry when it comes to technology. It has developed different storage solutions for products with different lifecycles and different usage requirements, and has set up multiple data centres in different local partitions to improve disaster tolerance. It has also conducted careful design and in-depth research on access control, redundancy methods, integrity checks, and repair mechanisms. The disaster tolerance technology it uses has made storage persistent and available at 99.999999999% and the technical specifications it has developed have almost become the industry standard in cloud storage.
Data is a special resource with strong user attributes and requires a special infrastructure
Amazon Cloud Storage, which is constantly innovating and meeting the individual needs of its users, seems to be able to always meet the needs of different users in different scenarios. However, human society has always evolved under the impact of change and new needs, and the specificity of data has led to new demands on data storage infrastructures.
This is because, in contrast to homogenous resources such as water and electricity, data is special in that it has strong user attributes. There is no difference between the water in the same reservoir, nor is there any difference in the water used by each household. However, data has a non-homogeneous attribute of user identification. Each data centre stores a variety of data, which belongs to different users and presents a one-to-one correspondence with them. Data is the aggregation and output of user knowledge and wisdom, and therefore has strong privacy and high value characteristics, and it is this characteristic that makes Amazon cloud storage subject to a twofold challenge.
1. Privacy challenges
The privacy challenge is that if data is stored on the Amazon cloud storage network, the user does not actually have real privacy. Because the storage devices and service facilities are owned by Amazon, there is still a risk of a breach despite the user encrypting the data. Although there have been no official active leaks in the Amazon cloud storage data breaches that have occurred so far, the absence does not mean that they will never happen in the future, as this risk exists in any centralized system, and Amazon needs to consume its own resources to resist this moral hazard. Users rely entirely on their trust in the platform, and behind that trust is credit, which has been passed down through human history for thousands of years, but trust has not proved to be as secure as it should be. Internet giant Facebook was strongly condemned for leaking user data, and the reason for this was the low cost of trust created by centralization and monopoly. So as soon as data is stored in any centralized system, privacy ceases to exist and the user has no real ownership of the data.
As a key resource in the digital age, data plays a pivotal role in the functioning of the digital economy. The flow of data is the flow of value, and any insecure factors can hinder this flow of value, which in turn can disrupt and affect the economy and the normal course of life. However, centralized systems are exposed to natural disasters (e.g. fire), hacking and downtime. Although many cloud storage providers have adopted the decentralized theory, they are still not completely free from this inherent risk. 2017 saw the “ultra-high error rate” downtime of Amazon AWS, where a large number of servers were mistakenly deleted, bringing down almost half of the global internet. In 2021, the French cloud services giant 0VH suffered a fire that brought down millions of websites. Data platforms of all different categories are also experiencing various hacking incidents of all kinds. The source of these risks is centralization and the inherent risk that exists in centralized systems, which can only be mitigated by human intervention and cannot be completely removed.
Why is MEMO the decentralized Amazon cloud storage?
Spanning to the early days of Web 3.0 today, the world is increasingly calling for a value storage network with strong privacy attributes and greater security. It must have the advanced storage technology of Amazon’s cloud storage and also must also be decentralized and give users ownership of their data.
Looking at the current crypto field, most centralized storage projects are labeled as decentralization, however, their storage technology is generally still in the early stages or only uses the original multi-copy redundancy method, or lacks the necessary recovery mechanism. Because decentralization is only the foundation that meets the storage needs of Web 3.0 and the future digital age, it must also have advanced storage technology and scalability. In this respect, MEMO decentralized storage is the first to go beyond in terms of concept and technical architecture.
1、A decentralized economic system empowered by blockchain
Web 3.0 is based on blockchain decentralization. MEMO provides storage services by organizing a massive number of edge storage devices and enables transactions and decentralized autonomy by deploying smart contracts on the chain. The protocol is designed with three roles: User is the storage demander, Provider is the edge storage device provider, and Keeper is the intermediate manager for information matching and verification, and the three roles interact automatically through smart contracts. The decentralized nature of the massive edge storage devices distributed around the world has an inherent advantage in combating natural disasters or accidents, as the failure of a few nodes does not affect the overall access and download of data. And once the smart contracts deployed on the chain are operational, no organization or individual has the right to change them, including the MEMO itself, making the data truly owned by the users themselves. The agreement suggests that Providers and Keepers with longer online time, quicker response and higher integrity will be rewarded more, while those with low integrity will be punished, thus promoting benign ecological development. Blockchain and its underlying technology enable the decentralization of governance and minimization of trust. The decentralization of geography and economic governance becomes the first barrier to isolating fires, hacking and downtime incidents.
2. Diversified disaster tolerance mechanisms bring high reliability
Decentralized storage is essentially decentralization + storage, with decentralization being the prefix and storage being the core. Therefore only with advanced storage technology can it be called advanced decentralized storage. The technical keyword of traditional centralized storage is the disaster tolerance mechanism. Over the years, disaster tolerance mechanisms such as redundancy mechanisms and data recovery have been developed to cope with natural disasters or accidents. The MEMO protocol has integrated these advanced technologies and upgraded them, giving users more choices. In terms of redundancy mechanism, MEMO protocol adopts erasure code mode by default. Users can set the number of erasure codes by themselves. For example, it can be set to “50+50” mode, which can tolerate data failure of up to 50 nodes. Although the probability of a 50% node failure is almost zero in a decentralized environment, it gives a qualitative improvement to the reliability of the storage. In addition, Users can also choose a multi-copy model, but regardless of the redundancy approach, RAFI technology acts as a data recovery guarantee. The diverse and selectable disaster tolerance mechanisms bring high reliability to the MEMO storage protocol.
3. Scalability created by layering
To improve availability and scalability, MEMO storage uses a strategy of layered scaling, i.e. storing smart contracts and transaction information on the blockchain and storing the actual user data on the edge nodes. Also to ensure data integrity, Memolabs has developed a public verification mechanism led by the Keeper role. This strategy of decoupling storage from the blockchain is primarily driven by the need to meet future needs for large-scale storage and increased availability, as current mainstream on-chain storage not only has the weaknesses of high redundancy and high latency, but is also expensive, and the layering approach not only reduces redundancy and latency, but can also significantly reduce storage costs. Throughout the blockchain field, the layering strategy is a major strategy to improve scalability, for example, the Bitcoin Lightning Network and Ethereum’s Layer2 adopt a layering mindset, which inherits the security of L1 while providing unique efficiency and availability. The scalability created by layering will drive the entire blockchain field closer to the mainstream.
4. Encryption from the source and editable privacy to meet the needs of different scenarios of use
With blockchain-enabled decentralized economic governance guarantees, upgraded storage technology for traditional cloud storage, and a unique strategy designed for privacy protection, the innovative combination makes the MEMO cloud storage protocol unique in both the traditional cloud storage field and the emerging decentralized storage field. Compared to traditional Amazon cloud storage, the distinguishing features of MEMO are decentralized governance and the empowerment of self-sovereignty over user data. Compared to other projects in the encryption field, MEMO is distinguished by its diverse disaster tolerance mechanisms + scalability created by tiering + privacy encrypted from the source and editable privacy strategy. MEMO is therefore a fusion of Amazon cloud storage and emerging blockchain decentralization technology, retaining the advanced storage technology of traditional cloud storage while giving privacy features and data ownership that are missing from traditional cloud storage. If one were to define the MEMO cloud storage protocol, it would best be defined as “decentralized Amazon cloud storage” — a future storage protocol that continues the best of traditional Amazon cloud storage while incorporating blockchain decentralized governance technology.