How does the DRAW algorithm increase the blockchain transaction speed to 100,000+?

July 17, 2018

As we all know, throughput has always been the pain point of the blockchain. It can be said due to low throughput, no killer level applications emerge in the blockchain industry. Similar to the early period of the Internet, the entire underlying system haven’t set up yet, users can only browse some simple web pages. The speed severely restricts the development of decentralized applications and blockchains.


How to solve the problem of low throughput? Currently, there are several options.

Option one, increase the size of the block.

For example, bitcoin, the size of an existing block may only be 1MB of storage space. If you want to trade, you can only add the transaction to this storage space. At the end of last year, SegWit2x enlarged the block size from 1MB to 2MB. But for security, It eventually eliminated the SegWit2x hard fork.

Option 2, over the chain transaction.

By contrasting Lightning Network for Bitcoin with Raiden Network for Ethereum, they have come up a solution. Paying some Ethereum or Bitcoin as a deposit in advance, and then users trade with others under the chain by using other means.

Option 3, the agent consensus agreement solution.

Whether it is 7 agents or 21 agents, or even dozens of agents, everyone will form a small group. Then some consensus agreement or similar function agreement will be run to reach a consensus, generate blocks, and then broadcast the block to the entire network, thus reaching the consensus of the entire network.

The above three solutions are currently good solutions, and the throughput problem is solved from different dimensions and different scenarios. However, the three solutions do not perfectly solve the contradiction between blockchain transaction speed and security.

A new solution – DRAW algorithm and double block refrigerating technology

For the first time, the IFTC team optimized the IFTC (Double Reverse Annulus Work) and IFTC Double Block Refrigerating Technology to optimize the current blockchain algorithm for efficiency and security, increasing blockchain trading speed to 100,000+.

DRAW’s idea: to allow each transaction to participate in the replication and building of the previous two transactions and the next following N transactions, in order to complete the transaction accounting in the reverse transaction sequence. After carry out X account notes, all X transaction records will be confirmed after the completion of block by the last one which has been replicated, and eventually it will form the final transaction group. After the transaction is initiated, this transaction accounting will be completed and the transaction loop will be closed after initiating replication process to previous TX and broadcast it to the entire network directly. After X transactions completed, the block of the whole transactions will be finished by last TX from the tail which has been replicated. After that, all transactions will be confirmed. The bookkeeper for each transaction needs to copy all previous complete blocks before participation in the accounting of each subsequent transaction.

Strategy of Double Block Refrigerating Technology

Design idea of DBRT: In today’s mainstream public blockchain, all public nodes bear the burden of storage transactions, smart contracts and various states. To maintain its normal operation on the blockchain, it may spend large costs to get more storage space. In order to solve this problem, a feasible method of block refrigerating technology has been proposed. The key to this technology is to separate the entire storage so that different nodes can store different parts. Therefore, each node is only responsible for hosting its own block data, rather than storing the whole complete blockchain. The first and most important challenge of refrigerating is to create N-block nodes. There is a need to develop a mechanism to determine which nodes can create indexes in a secure manner, avoiding attacks by people who can control a large number of specific nodes.

The best way to defeat an attacker (in most cases) is to establish randomness. By exploiting randomness, the network can randomly extract nodes to form an index. Such a random sampling approach can prevent overfilling a single index by malicious nodes.

But how can we build randomness? The easiest source of public randomness is the block, for example, the Merkle tree root of the transaction. The randomness provided in the block has been publicly verified and uniform random bits can be extracted by the random extractor.

The IFTC team claims that more technical details will be announced on September 1, 2018. Including smart contracts, digital currency payment transaction settlement system, billing system, Token, decentralized exchange underlying architecture, IFTC has established a basic underlying platform that can be applied in vertical sectors. It’s an actual blockchian technology that can be used by users and realizing the application and construction of the business ecology at the blockchain level. IFTC, reconstructing global business.