In September, Zilliqa , a project to develop next-generation, high-throughput distributed ledger technology (DLT), announced that a testnet of 2,400 nodes had achieved 1,389 transactions per second, a preliminary result that the project believes it can improve upon.
In fact, a following post reported a peak throughput of 2,488 transactions per second in a test network of 3,600 nodes. The Zilliqa team plans to release the source code of the new DLT and a public testnet in December.
Other notable features of Zilliqa's DLT are a relatively low energy footprint and a light, non-Turing-complete smart contracts system. But Zilliqa's truly killer feature is the promise of achieving a throughput much higher than other open DLTs like Bitcoin or Ethereum. It seems likely, indeed, that Zilliqa's high throughput will make it a strong candidate for implementation in legacy enterprises focused on financial markets, global trade and other sectors that require high-performance DLT.
Changing Enterprise Through Sharding
According to Zilliqa, throughput scalability is a major challenge for today's DLT, which could severely limit adoption.
"Existing blockchains are not able to scale for the next generation of internet-style applications," notes Zilliqa's first announcement post. "An often cited example is the 7-10 transactions/second (TX/s) available in Bitcoin and Ethereum today, and the demands of payment processing in centralized operators (e.g., VISA, MasterCard) for supporting thousands of TX/s."
Zilliqa aims to achieve scalable throughput, which rivals traditional centralized payment methods in an open, permissionless distributed network, without compromising resilience and security.
Zilliqa's DLT is based on the concept of "sharding" - dividing the network into several smaller component networks (called shards) capable of processing transactions in parallel. Sharding, which is seen by many DLT developers as a promising way to achieve high throughput, is also part of Ethereum's roadmap for the future.
In 2015, University of Singapore computer scientist Prateek Saxena, now chief scientific advisor at Zilliqa, co-authored a paper titled " A Secure Sharding Protocol For Open Blockchain s," which was the first proposal for sharding in the context of open DLTs.
The paper concluded that the protocol "scales up the agreement throughput near linearly with the computational power of the network and tolerates byzantine adversaries … It offers promising scalability in experiments and suggest[s] strong usability in next-generation cryptocurrencies."
Since then, Zilliqa's sharding protocol has been under research, refinement and active development.
In a short video published by CNBC , Zilliqa's CEO Xinshu Dong discussed the project's plan to achieve 10,000 transactions per second, a throughput comparable to the major credit card networks. Similar or higher performances have been achieved , but not in an open, permissionless blockchain.
If a network of 1,000 nodes is divided into 10 shards, each with 100 nodes, and each shard is capable of processing 10 transactions per second, then all shards together can process 100 transactions per second. Thus, the sharded network achieves a throughput that scales linearly with the size of the network.
Of course, this is easier said than done and there are technical challenges to overcome. Subsidiary sharding techniques, indicated as "transaction sharding" and "computational sharding," are built on the network sharding layer.
The " Practical Byzantine Fault Tolerance " (PBFT) consensus protocol, which has been chosen for the Zilliqa blockchain mainly because its efficiency depends on the size of the network, can exploit the small size of the shards in the Zilliqa network. A related feature is that, while miners use proof-of-work (PoW) to establish their identities on the Zilliqa blockchain, the PBFT consensus protocol limits the amount of computing that must be done to establish PoW. Therefore, mining in the Zilliqa blockchain will be cheaper than in PoW blockchains such as Bitcoin and Ethereum, and offer a smaller energy footprint and less impact on the environment as well.
The Ziliqa blockchain will support smart contracts with a dataflow programming system that, while not Turing-complete like Ethereum, offers better scalability and can be more easily parallelized for a sharded blockchain. Dataflow programming is visual and used in demanding commercial projects, including game development, which ensures the availability of talent.
Recently, Zilliqa added high-profile fintech and banking experts to its advisory board and secured a strategic partnership with blockchain digital asset management firm FBG Capital. A draft position paper describes the current Zilliqa roadmap, with the plan to launch public test and main networks in the second quarter of 2018 and then further increase the throughput of the blockchain and the decentralized applications running on it.