PoW, Meet DAG: How Conflux Plans to Solve the Scaling Problem
January 10, 2019 by Paul de Havilland
Conflux, set to launch its test phase in February, has proposed a scaling protocol that runs contrary to the growing movement toward Proof of Stake. By interlacing Directed Acyclic Graph (DAG) data with Proof of Work, the startup says it can provide transaction speeds of over 6,500 transactions per second.
Also read: Ahead of Ethereum Constantinople Update, Community Eyes ProgPoW
Subscribe to the Bitsonline YouTube channel for great videos featuring industry insiders & experts
The Lightning Network Faces a Cashed Up Competitor
Lack of pace in transaction throughput has been a constant criticism of bitcoin and similar blockchains. Transaction sloth makes bitcoin unfeasible as a real-world ledger in most transactions. While a number of proposals have been put forth to solve the transaction speed problem, a new protocol has emerged via Conflux.
Conflux is backed by Sequoia China, Metastable, IMO Ventures, FreesFund, Rong 360, Shunwei Capital, F2Pool, and Huobi to the tune of $35 million USD. The technology is a blockchain protocol that is premised on Proof of Work and touts itself as a decentralized, scalable, dApp blockchain solution.
At the root of blockchain’s scalability problem is that blocks are mined and added to the ledger sequentially. If more than one block can be added at a time (which would currently represent a fork via the creation of more than one chain), the ledger would be able to scale and accommodate more transactions at a time.
Conflux Has a Hybrid Solution to Scaling
Conflux’s architecture permits concurrent blocks being worked on and added to the blockchain. Their claim is delivering the speed of centralized blockchains without sacrificing decentralization. It was developed by Andrew Yao, China’s “godfather of computer science,” and Fan Long, a professor at the University of Toronto. Per Long:
“Conflux’s main idea is how to make the whole blockchain scalable. We’ve changed the structure of the blockchain so that it’s no longer a chain in the sense that it records each block based on what its parent block says.”
Best of Both Worlds
A company press release describes this method of achieving speed and decentralization at the same time:
“Contrary to popular belief, true decentralization isn’t sacrificed to increase throughput, highlighting Conflux as the first example that achieves the best of both worlds. By weaving a Directed Acyclic Graph data structure into Conflux’s Proof of Work consensus algorithm, tests on its testnet has achieved a throughput of at least 6,500 Transactions Per Second (TPS), while supporting at least 20,000 nodes.”
The DAG protocol is deployed in IOTA’s tangle architecture. In theory, it is not a blockchain, but a “tangle”, where transactors approve transactions independently of miners. The speed and cost benefits of DAG are significant. (Admittedly, Bitsonline’s Tyson O’Ham has described significant technological problems with IOTA’s tangle.) Nano, similarly fast and transaction-cost free, uses a similar architecture in its block lattice protocol.
By interweaving DAG into a Proof of Work algorithm, Conflux hopes to bring scalability to sluggish blockchains while remaining faithful to the idea of decentralization and immutability.
Have your say. Is the use of DAG crucial to solving the scaling problem?
Images via Pixabay
