The First Year of Distributed Storage: the Data Revolution of EpiK Knowledge Graph
A cover article in The Economist pointed out that data has replaced oil as the most valuable resource in the world today. In the current competition of Internet giants, traffic and data have become the core of competition, and data has become the “oil” of the digital era. But data is only valuable when it is stored, and as a new force in future storage, distributed storage is highly expected.
The launch of Filecoin, a popular distributed storage project, has become a milestone event, and for this reason, 2020 is also called the first year of distributed storage.
In addition to Filecoin, distributed storage projects are generally favored by the industry, including EpiK, which is considered by many users to be the “dark horse” of distributed storage projects and the “pioneer ox of the smart era”. Why are distributed storage projects favored, and what does EpiK bring to distributed storage?
Why Distributed Storage Projects Are Favored
As the saying goes, at every point in history, something new is born to drive the wheel of history forward, and similarly, the emergence of distributed storage has its own special background.
In the Internet era, there is data being generated quietly almost every second. Looking at the history of Internet development, we will find that: from LAN to Internet, information dissemination has gone through many stages from code to text, to pictures, to video, to live streaming, etc. In the future, VR and holographic projection technology will be popular for information dissemination, and with the arrival of 5G and IoT, big data will grow explosively.
In the case of the geometric explosion of big data, the retention of these data has become an important issue, and storage has become the primary problem to solve.
About storage, for everyone, the simplest example is the U disk, hard disk, computer, etc. Of course, this kind of storage can only meet personal storage, and can’t meet enterprise storage or industry storage. But storage services are also constantly innovating, developing into online servers that store data centrally through network transmission, that is, Tencent cloud, Ali cloud, and other cloud servers. And their revenue has also exploded.
According to IDC, global data generation reached 1.8 ZB (or 1.8 trillion gigabytes) in 2011 alone, and the total amount of data worldwide will continue to rise. According to Western Digital, there will be 32 ZB of data generated and 5 ZB of data stored in 2018, and 103 ZB of data generated, and 12 ZB of data stored in 2023. As data generation accelerates and storage is “squeezed into a dead end”, the emergence of distributed storage becomes an important turning point.
After a series of changes, Filecoin finally launched the mainnet this year, which also gave a boost to the industry that was looking to distributed storage. For this reason, 2020 has become the “Year of Distributed Storage”.
With that said, what exactly does distributed storage solve?
Take enterprise storage, for example, the storage services provided by enterprise-based servers obviously require more hardware investment, operation, and maintenance, while the excessive scale also faces excessive liability, data security and confidentiality is particularly important.
IPFS distributed storage, can mobilize idle storage space in the network to help those in need of data storage, through the incentive layer to obtain the corresponding profits. Its fragmented distributed storage and self-healing features also make data more secure and confidential.
For this reason, a vast distributed storage wind is blowing slowly, and more distributed projects have been in the tide.
EpiK Looks at the Value of Data Storage
The current boom of distributed storage projects makes the whole track look more and more prosperous. But behind the boom, there is not a single well-known storage project in the market that is not dealing with the useless data, for the sole purpose of getting rewards.
How significant is data storage for the purpose of mining? Has it run counter to the original idea of distributed storage?
On this issue, EpiK has been designed from the beginning with the goal of building a “knowledge graph” and the role of field experts. This is different from almost all storage projects in the market.
The major improvement of EpiK lies in capturing the value of data. The goal of EpiK is to guide all community participants to co-construct and share knowledge graph data in various fields, to open up AI cognition.
Knowledge graph data is the best way for robots to understand human knowledge. The more knowledge graph data there is, the more knowledge robots master, and their intelligence rises exponentially. EpiK adopts decentralized storage technology to capture the exponentially growing data value with linearly increasing hardware cost.
For data storage, EpiK’s underlying distributed storage ensures data authentication and non-tampering. In terms of knowledge graph construction, EpiK has built three major platforms, namely knowledge graph construction tool platform, decentralized collaboration platform, and knowledge vault application platform, of which the tool platform contains four modules: knowledge acquisition, knowledge error correction, knowledge fusion, and knowledge storage. The purpose is to process these unstructured data into a triadic network, thus reducing the conversion cost of knowledge.
Since the organizing of knowledge graph is a labor-intensive job, the process of super large-scale knowledge graph construction includes the collaboration of multiple parties, including field experts who organize the knowledge skeleton of each field, bounty hunters who help field experts process and detect data, miners who provide storage space, users who use the knowledge graph, etc. The decentralized collaboration platform organizes the roles of all parties well and motivates each participant to continuously enrich the human knowledge vault for their own benefit through an incentive-compatible token economy model.
To sum up, EpiK is not just a distributed storage project. It is leading the knowledge graph data revolution and is striving to open up AI cognition, which is an epic sermon of at least 50 years of evolution from carbon-based life to silicon-based life.