Team Details

The core of QRYIX system is the link library separation mechanism and functionsub-chain design. Decentralized applications can store data in on-chain and databasesystems respectively according to different trust and public verification levels of data.

QRYIX system provides different types and levels of data collaborative management.Moreover, the multi-database cluster system is a Permissionless environment. QRYIXsystem also completes access control mechanism based on multi-authority attributebased encryption, as well as complete proof of possession of stored data.

The main reason for the design of link database separation is to consider the futureupgrade and update of the system, because the update of the blockchain system willlead to the fork of the system, which will cause irreversible impact on the entireeconomic system, so we put the main data processing capacity on the database system,and the access control system of the database system through the function of thesub-chain to complete. Functional sub-chain design one is for the future expansion,more is to complete the two core functions of the decentralized storage system: privacyprotection and data holding proof. We use an efficient multi-authority attribute-basedencryption scheme to achieve the cloud storage data access control function andencryption function.

Accounts

QRYIX uses the concept of state to store a series of accounts, each with its own identityand unique data. In some cases, the accepting account has a code that needs to beexecuted, and the transaction triggers the execution of that code, then the internalmemory of the account may change, and additional information may even be createdto be sent to other accounts, resulting in a new transaction.

Merkle tree

Bitcoin does this through something called a Merkle tree, and IPFS also stores datathrough a directed acyclic graph data structure of the Merkle tree. The Merkle tree issimply that when the file is large, the IPFS system will split the file when the useruploads the file to the IPFS node, and then each file will use a hash value as its file name.And then these files are saved by a number, in terms of the total, for example, thisnumber is like there are many leaves, and then the branch with two leaves is actually ahash of the two leaves, so from the leaf to the branch, and then from the branch to theroot. And in this way, you can make sure that when the data on a particular leaf changes,the hash value of the root changes as well. This method is actually the same as the datastorage method of Bitcoin. Its purpose is to allow the entire network to verify theintegrity of a piece of data as quickly as possible. Because we don't need to comparethe whole file, we just need to see if the root value is still the same. If yes, the differentnodes can prove that the data has not been tampered with.

The Merkle Patricia tree/trie, conceived by Alan Reiner and implemented in the Ripple QRYIX White paper protocol, is the main data structure of the QRYIX system and is used to store all accountstates, as well as transaction and receipt data in each block. The Merkle Patricia tree/trieis a combined abbreviation of the Merkel tree and the Patricia tree, and the structurecreated by combining the two trees has the following properties:

• A hash value that maps uniquely to the root for each unique key-value pair; In Merkle Patriciatree/trie, it is not possible to fool members with only one key-value pair (unless the attacker has~2A128 computing power);

• The time complexity of adding, deleting or changing a key-value pair is logarithmic.

Merkle Patricia tree/trie provides QRYIX with an efficient, easily updated fingerprint thatrepresents the entire state tree.

RLP encoding

RLP is intended to be a highly simplified serialization format whose sole purpose is tostore nested arrays of bytes. Unlike existing solutions such as protobuf BSON, RLP doesnot define any specified data type - such as Boolean, 20floa, double, or integer. It simplystores the structure as a nested array and leaves it up to the protocol to determine whatthe array means. RLP also does not explicitly support map sets, semi-officialrecommendations are to adopt [[kl, vl], [k2, v2],...] Is a nested array of [kl,k2...] Sort bythe standard string.

A scheme that has the same functionality as RLP is protobuf or BSON, which are thealgorithms that are used all the time. However, we prefer to use RLP because:

• It is easy to implement;

• Byte consistency is absolutely guaranteed.

Node architecture

QRYIX system itself is an isomorphic multi-chain design, including verification node(super node) and storage node (voting node). Super node is the "block producer",referring to those nodes that collect, package and verify transaction information to theblock. It is the basis of stable operation of QRYIX system network, which is based onPOS consensus mechanism. The working principle of POS is as follows: similar toproperty stored in a bank, this model will allocate interest to the user according to theamount and time you hold digital currency.

In simple terms, it is a system that pays interest to users according to the amount andtime of currency held by users. Under the proof of equity POS mode, there is a nouncalled coin age, and each coin generates 1 coin age every day. For example, users hold100 coins for a total of 30 days, then the coin age is 3000. If the user finds a POS block,the coin age will be emptied to 0. For every 365 coin age emptied, the user will receive0.05 coin interest from the block (assuming that the interest can be understood as 5%per annum), then in this case, the interest = 3000 * 5% / 365 = 0.41 coins, which is theinterest on the coins. In addition, POS-based pledge lending will also have an efficient,safe and stable system performance.