The Road To Adoption (Part 1): The Scalability Trilemma
For blockchain to continue on its path to enterprise and mainstream adoption, it first has significant obstacles to overcome. One of the most pressing issues is scalability. In this three-part series, we will examine the factors impeding the scaling of blockchain networks and the best technological solutions posited to dealing with them.
The Current State of Major Networks
Bitcoin has a maximum transaction-per-second rate is the equivalent of seven simple transactions or three complex ones. With 86400 seconds in a day and Bitcoin averaging approximately 300,000 transactions per day, the network is operating at near capacity.
Ethereum can handle an improved 15 transactions per second. While the hypothetical capacity for a single Ethereum node is over 1000 transactions, in practice, existing “gas limits” prevent the network from functioning at close to its optimal rate.
EOS, the network focused on speed and scalability by using a Designated Proof-of-Stake consensus mechanism, is experiencing “blockchain bloat”. After less than a year of existence, it’s grown to 4 terabytes — dwarfing the older Ethereum network which weighs in at a total of only 200 gigabytes.
With these limitations in both inherent processing speed and a size restrictions serve as major obstacles to adoption. With major, traditional payment networks like Visa processing ~24,000 transactions per second, there needs to be a much higher TPS achieved by networks to be able to compete with established payment providers.
However, it is not only the finance and cryptocurrency based projects that need high TPS. For applications such as gaming and advertising, which generally require huge amounts of throughput, there have only been recent advances that are allowing traditional industry players to even consider adding one of the more established blockchains to their tech stack.
In late 2017, Ethereum transaction times skyrocketed with the release of the CryptoKitties dapp. With multiple networks vying to be the leader in the decentralized application space, creating an environment that allows for huge amounts of transactions-per-second (TPS) is necessary for the success of dapps when they find their way to the mainstream.
High speeds are achievable on private blockchains as there are far fewer validators needed for each transaction. However, this lower decentralization requires more trust for the network to operate. This concept will be covered more in future blogs.
The Scalability Trilemma
The fundamental challenge to scaling up blockchain projects can be best illustrated through ‘The Scalability Trilemma’, a term coined by Ethereum’s Vitalik Buterin.
The trilemma is a manifestation of a project management triangle — a tool used to demonstrate how products and services can only achieve two of the following three characteristics:
— Good
— Cheap
— Fast
For example, you can implement a cheap, fast project but you will lack quality, or you can implement a great project at low cost but it will take time.
Using the same logic, the Scalability Trilemma states that a blockchain can only be two of the following:
- Decentralized
Decentralization means that the technology is not housed on a single server but instead distributed across multiple nodes (and thereby participants) in the network. Crucial to the blockchain philosophy, decentralization removes the need for a single point of control — a central authority to validate transactions. Instead, the consensus protocol should immutably, incorruptibly validate data; creating a trustless network.
- Security
Security does not pertain to bugs or error-codes but to defense against bad actors. A network must be safe from hacks and exploitation, it should be able to withstand “51% attacks” and “Sybil attacks”.
- Scalability
Scalability and speed in a blockchain is measured by how many transactions a network can process regardless of the volume of network participants. Independence between speed and participant quantity is essential to scalability.
Public blockchain solutions will always struggle to be faster than traditional databases running on a single (or distributed) server, meaning that the benefits of decentralization and security must outweigh speed limitations.
The trilemma can be illustrated with the example of Bitcoin and its Proof-of-Work (PoW) consensus protocol. To add a block to the chain, PoW requires significant computing power and therefore high amounts of energy consumption. By maintaining these requirements, Bitcoin’s network maintains a high level of security and decentralization, but by necessity, sacrifices speed.
Inversely, EOS and Ripple are secure and fast but fall behind on decentralization. Buterin’s own Ethereum network, like Bitcoin, is secure and fully decentralized but lacks speed.
The development of viable solutions to ease the strain of different sides of the triangle is ongoing. As the enterprise sector continues to investigate blockchain solutions, the ability for them to mitigate the different aspects of the triangle to fit their needs remains one of the primary focal points for integration.
In part two of this mini-series, we’ll be examining Bitcoin Cash’s expanding of block sizes and Ethereum’s upcoming plan to switch to a Proof-of-Stake (PoS) consensus model codenamed “Casper”. In part three, we’ll investigate other layered solutions such as sidechains that allow smaller faster networks to connect to the larger, more secure networks, and dedicated scaling solutions.
Feel free to share this story on your social networks to keep you and your friends in the loop!
To make sure you don’t miss future articles, follow and favourite us on LinkedIn, Twitter, and Medium.
