AnyLog is a decentralized network that allows members with different roles to join at any time by downloading software and joining the network. The AnyLog network supports 3 types of members – Data Owners, Operators, and Data Consumers. These members are extremely motivated to participate in the network as:
Data Owners are members that generate data. Examples of data owners are: a smart city that has devices that generate data and an owner of a car that generates data. The Data Owners are motivated to push data to the network because for two reasons:
Data Consumers are users and applications that query the data. They are motivated to use the network as they query and view the data and extract value from the data. For the data consumers, the network satisfies their queries using Standard Query Language (SQL) without the need to collect and organize the data.
To maintain the ownership and security of the data, AnyLog is using advanced access control and encryption mechanisms. With this approach, Data Owners maintain full control over their data, even when it is stored and processed by third-party Operators. Data hosted by Operators is encrypted, and queries are performed using encrypted query processing techniques, secure enclaves, or both. Alternatively, if data is hosted by a trusted Operator, the overhead of encrypted query processing can be avoided by providing direct access to the data as a trusted Operator may enforce the client’s security policies directly. Unlike centralized approaches to enforce the Data Owner’s policies, Anylog need not rely on a trusted third party. Authorization may be in the form of a digital signature from the Data Owner authorizing all queries from a particular entity (e.g., an employee of the Data Owner), or as an authorization token obtained by the client (through purchase or other means) that entitles it to a restricted class of queries. Without such authorization, Operators will not process queries, and even if they did, clients would not be able to decrypt the results. Even when data is primarily deployed on trusted operators, Anylog provides a seamless path to increasing both storage and processing capacity beyond the resources of the trusted operators without sacrificing security.
Anylog develops a peer-to-peer network to allow users to query IOT data without reliance on a centralized third party. AnyLog specializes in distributing the data and enabling real-time queries over the distributed data. This approach is provides a backbone of IOT whereas centralized databases are replaced with a decentralized approach. With this approach, the data generated from different devices is organized in log files which are distributed over the network for storage and query.
The vision of the decentralized web began with the emergence of networks like Ethereum which enabled trustless computing and with projects such as Swarm, storj, Sia, and IPFS/Filecoin to enable decentralized storage and content distribution. AnyLog is unique in creating a platform optimized to support Time Series Data. Time Series Data is the type of data that represents data generated from smart meters and devices. By providing an efficient and scalable method to manage Time Series Data, companies can create applications and services leveraging the data, create interactions between devices and processes, support AI and predictive maintenance applications - all of that in a cost-effective way.
Today, to support different applications and services on top of IOT data, systems lack the ability to cope with the increased amounts of the data, the diversity of the data, the real time requirements and the associated costs. The goal of the AnyLog project is to decentralize the management, storage and query of IoT data to allow the following:
In the context of IoT, the computational requirements for CPU and storage are enormous. AnyLog is a platform that offers these resources through a decentralized approach where independent operators are incentivized to join the network and provide the needed resources.
A fully decentralized p2p solution, where nodes contribute their own computation and bandwidth in service of IoT data is more scalable, as there is no practical limit to the amount of CPU and storage that can be provided and the number of connections that can be served.
1. Address the growth of data generated from applications and smart devices.
2. Support real-time interaction with the IoT data.
3. Provide the access control and the security mechanisms that IoT data owners require.
1. AnyLog is a cost-effective way to manage data. The Data Owners alternative is to use databases that service the data (creating schemas, scaling the databases, managing the availability approach etc.). These are long, complex and expensive processes when large data sets are involved. AnyLog removes the need to use databases.
2. AnyLog is a platform that allows Data Owners to monetize their data as data that is placed on the AnyLog Network and can be queried in a unified way. When a user issues a query, a data owner can make his data available to the query and be rewarded for providing the data.