The fastest growing data processing services today breeds a new form of decentralized cloud that can enable blockchain computing. Plan Flash, a global advanced decentralized blockchain provider, recently launched a next generation infrastructure for decentralizing data processing.
Plan Flash uses a Committee of Computing Facilities (CCF) to select certain computing devices entering into the network. Currently, what Plan Flash chooses to access are mostly small and medium-sized data processing centers, idle server sets of large institutions, and some specific types of servers provided by individuals. Plan Flash chooses a proper distributed-structure for real-world practice to balance efficiency and decentralization.
At the early stage, professionals have to select specialized computing devices to maintain the stability from abruptly withdrawn from the network. Plan Flash can grade the computing performance of devices yet conduct targeted of both software and hardware upgrades for specific data processing tasks to further improve service efficiency.
In the next stage, CCF will make an access rules document and develop a set of test programs to receive devices that meet certain performance requirements to access the network. Any device provider wishing to access Plan Flash will be able to access the network through a standard process and local compliance procedures. After being audited and tested with the device, the provider could access the network and will be supervised by an intelligent contract with a guarantee to provide stable hardware services.
Plan Flash is developing a distributed computing platform with cross-chain API, for DApps or other users to finish data processing and analysis tasks online. When users need to analyze data, they can initiate a data analysis task to Plan Flash network.
After receiving the task, the user’s connecting node will broadcast the task on the network and then Plan Flash processors response to match according to the communication cost (lowest), the size of the data request (affordable) and the required computing resource (available). The connecting node will choose processors through the intelligent scheduling algorithm to track and maintain the status of the computing task. Once the task is finished, the participating processors get corresponding reward.
Plan Flash divides data processors into 8 levels to meet higher computing power such as image recognition, voice print analysis and so on despite low cost. Due to its unique Cluster Fog structure, Plan Flash can receive fairly large-scale data processing tasks as long as larger requirements through simple task sharing, seen to be a bright breakthrough.