#14 ASSEMBLE Technology 2
Procedures for processing, tracking and storing data explained
ASSEMBLE protocol is a blockchain-based global point integration platform that exploits ASM utility tokens, whilst establishing a business ecosystem that can integrate, utilize existing points and miles with point providers, consumers and retailers.
In the previous post, we looked at essential tips that taught us the best way to spend the reward points we’ve collected through loyalty programs. Today, we are back to the discussion about the technological aspects of the ASSEMBLE Protocol. In this article, we will explain the procedures, such as tracking, storing, and processing data, that are carried out using the IPFS (a distributed storage system), DIP and Spark technologies.
Data Storage
Accumulated data in the ASSEMBLE Protocol can be stored and shared in a distributed storage system called the IPFS. The abbreviation stands for InterPlanetary File System, which is a distributed blockchain file system, in other words an encrypted-blockchain-based database. Here, data can be stored with a certain encryption and shared with selected users. Users can encrypt data using their own pair of keys (asymmetric encryption) and further save the encrypted data in the IPFS. With asymmetric encryption, ASSEMBLE users can encrypt data using the public key of other users with whom they want to share their data with. Then, selected users can access this data using their individual keys. Users who do not have access cannot decode this data, which guarantees the protection of personal information.
Image 13 above illustrates how the user’s consent is obtained and how the data is shared in a selective manner. For instance, User A is willing to selectively share his data, and only allows User B to access it. On ASSEMBLE, User A encrypts data using User B’s public key, uploads an encrypted data file to the IPFS, and receives a hash value of the file. Then, User B can find and also access this file since he has an individual key of the public key which was used to encrypt the file.
The IPFS can be regarded as a storage protocol similar to BitTorrent. It performs multiple tasks through hash reference in order to achieve deeper program interaction using a completely distributed interaction. Blockchains usually have a dedicated BPM module that can log simple text records very effectively, which makes it suitable for digital assets to be executed on blockchains. In a digital asset application scenario, the BPM module can be executed in a very effective way because it just needs to record information of the sender, receiver, and digital asset. However, if you need to save a large volume of other types of data like text data or personal information, you should calculate and check all of the hashes every time you create a block, which significantly reduces storage efficiency. Preserving chain integrity results in very inefficient block creation.
To address this issue, people are coming up with new strategies that utilize the combination of IPFS and blockchains. ASSEMBLE saves a hash value of the IPFS creation storage file, which is the user’s data, on the Ethereum blockchain instead of BPM. This guarantees the simplicity of data which is necessary for blockchains. In the meantime, this also provides the benefit of the complete decentralization of IPFS.
Data Tracking
It is possible to track all the records by saving a hash of encrypted data on the blockchain. For example, it can be used to track users’ personal data. Project configuration files (DIP) are provided to give a unique number to ASSEMBLE’s user information, and save it in an encrypted-blockchain-based database like the IPFS mentioned above. This is then mapped onto the Ethereum network and blockchain (side chain) through DIP hash value.
DIP consists of digital containers that store reference information on records and digital documents. All of this data can be collected and traced. Records in DIP are saved in chronological order and create small-sized blockchains that form transactions in DIP including timestamps and the hash value of former records. Logging records requires the signature of one’s individual key, which enhances the ability to verify each record.
There are three characteristics in ASSEMBLE’s data tracking. They are explained below.
User Identification
Each user should have a unique identity (ID), so the key concept is the DIP of ASSEMBLE users. Records stored in ASSEMBLE are based on a unique user DIP.
Data Encryption and Integrity
ASSEMBLE stores data on the blockchain. Anyone who has authorization can trace this data, but it is impossible to have access to it without authorization. Also, no one can modify this data.
Tracking Tool
For quick and automatic tracking of certain data, users should use a convenient system. ASSEMBLE builds a safe tracking system, shown in the image above, using blockchain technology.
Data Processing
ASSEMBLE builds a big data processing platform using Spark. Spark is ideal for building large-scale, low-latency big data analysis and machine learning applications. Big data users can use this platform for data mining and machine learning, which are related to other events and customer behaviors. For example, you can study the behavioral patterns of consumers who make a certain purchase and use that data for more targeted marketing through ASSEMBLE’s big data processing platform.
To be more specific, this is like an open-source cluster computing environment which is similar to Spark and Hadoop, but there is a difference between these two. Due to this difference, Spark performs better under certain workloads. Especially, Spark can activate a data set that is distributed to memory, providing interactive inquires and optimizing repetitive workload. Below is the structure of Spark on the ASSEMBLE Platform.
