Blockchain Scalability: The Issues, and Proposed Solutions
As everyone knows, cryptocurrencies had a fantastic year in 2017.
But now, with more tokens, users, investors, exchanges and startups involved than ever, scalability is emerging as a serious issue. With transaction data piling up, the current system is beginning to strain under its own weight.
With every purchase, the blockchain adds one more block to its ladder of transactions, and every block increases with data as it carries the history of the blocks before it. As more users join the networks and the transaction histories of individual coins grow, the current system is in danger of buckling.
Bitcoin blocks were originally hard-capped at 1MB, or around 2,020 transactions, but there is the potential for an infinite number of transactions to be made on each coin and therefore to be recorded on each block.
As transactions continue and records grow, block sizes increase, eventually exceeding any limits set. Although Bitcoin has increased its cap per-block, and Ethereum has no hard cap and therefore can adjust to scale much better, this still presents problems: blocks will continue to grow with use, and each transaction will need more time to be processed.
Every recordable transaction requires peer-to-peer verification, which can become time-consuming with the number of blocks involved. Bitcoin is currently verifying, or creating, one block every ten minutes.
As more users incur more transactions, it takes longer for verification, with waiting times increasing sharply at peak times.
Bitcoin can handle approximately 60 transactions per second, which pales in comparison to Visa’s peak rate of 47,000 per second. In order for it to expand into the same ubiquitous role as fiat currency, cryptocurrency must be able to process much higher numbers of transactions. For Bitcoin to reach Visa’s numbers, it would be equivalent to trading four terabytes of data per year. Ether, however, despite its lack of block limit, takes approximately 14 seconds to generate a block.
To handle the increased traffic caused by more users and more transactions, more nodes are needed to process them, and the running costs aren’t cheap. Miners also shows preference for transactions with higher fees, meaning that to have a transaction verified in a timely fashion during peak times can escalate from the fraction of a cent to a number of dollars, and there’s no telling how high it might climb when scaled. Ether is considerably cheaper than its primary rival, with transaction fees less than 2% of Bitcoin’s.
When a platform drastically branches away from its initial ideals, this is known as a hard fork. Following the hacking of the DAO (decentralized autonomous organization) on the Ethereum network, where $53 million of crowdfunded cryptocurrency was “stolen”, Ethereum took a hard fork in order to reclaim the money and continued as Ethereum, while the existing course maintained the original blockchain as Ethereum
Bitcoin recently adopted a hard fork in its capped block size, which means that old and new software are incompatible with each other and renders the old outputs invalid. Bitcoin has already forked previously, such as with Bitcoin Cash, and there are more planned for 2018.
The proposed Bitcoin hard forks will all incorporate SegWit (the Segregated Witness soft-fork), which is software designed to primarily to solve transaction malleability but also improves the capped block size issue.
It increases the block size limit to 4Mb, meaning a single block can hold the records of over 8,000 transactions. However, although the block increase provides short-term respite in scalability issues, it will still eventually present the same restrictions once transactions have exceeded the limit.
The Lightning Network is a payment protocol that relies on the SegWit platform and which introduces instantaneous transactions.
Rather than verifying transactions through the cryptocurrency network miners, which is particularly time-consuming for Bitcoin, transactions occur on a separate channel, and the blocks are updated after. The Lightning Network is being trialled as of last month, with TorGuard being the first VPN provider to utilise it.
Distributed Ledger Technology
DLT, developed by Polish fintech company Billon that deals with digitised versions of currency rather than cryptocurrency, offers fast transactions at a speed of 130 million per day, a number comparable to traditional banking giants like Visa. Billon has said that its platform is “faster, highly scalable and more cost-effective [….] than any alternative approach.”
Zilliqa is an alternative blockchain platform that utilises sharding (splitting the block verification process and running parallel subcommittees to collate the completed data) to handle 2,400 transactions per second and is aiming to ultimately match Visa’s average of 8,000 transactions per second.
Perhaps most importantly, Zilliqa reacts appropriately to scaling as its throughput increases with its network size, as opposed to Bitcoin becoming clogged with transactions. With a node size equivalent to Ethereum, Zilliqa predicts it could handle twice the transactions of Visa per second.
Ethereum’s Plasma will trim unnecessary data from the root chain, instead of working from the blockheader hashes to verify the transaction, which will minimise the time and power needed to process transactions. This will present only the completed transactions to the public Ethereum blockchain for a more streamlined verification process. Since Plasma focuses on the verification of the blockchain, a two-fold process could be employed whereby the Lightning Network processes the transaction and Plasma updates the blockchain.
The Future of Scalability
Since cryptocurrencies are decentralised, a myriad of possibilities may be presented due to the platforms being, so user developed. Political principles influence how much a service may stray from its original ideals, but if cryptocurrency is to fully integrate and perhaps exceed fiat currency, it must be prepared for the billions of transactions it must support and log.