A smarter way of benchmarking blockchains and smart contract systems
Posted 21 Dec 2020Since the early modern period, people have been obsessing over how to make all human inventions perfect. Human innovation has been the thing to drive the creation of technological machines. As soon as we have created something new, we want to improve it by making it more powerful and faster. This way of thinking is something humans have implemented into the blockchain software industry by always trying to improve the technology by making the speed faster.
In the IT world, hardware and software engineers use benchmarking methodologies that tend to cover a variety of things such as CPU (central processing units), GPU (graphic processing units), DBMS (database management system) and I/O (input and output). The most relevant one in today’s enterprise blockchain landscape is CPU as it is still the most important resource for dev-ops.
As an example, there is a software called Ethereum which was originally advertised as The World Computer. On this, smart contracts can be created and deploy virtually at any time and any place. The desired code will be executed by thousands of computers, simultaneously all around the world. In order to make the CPU footprint of the code as small as possible, continuously benchmarking and optimising the blockchain software must take place. Unfortunately this method seems to be very costly and contains elements throughout the process that can cause a large amount of development days. The element responsible for this particular CPU footprint is the virtual machine (VM) in which the smart contract is executed. In order to find the best approach to optimise the CPU usage of all smart contracts, it needs to virtualise them and run them in a VM system like WebAssembly as it is safe and native-speed fast. It then needs to look at the best and most efficient option, before being ready to go.
To understand in more depth how to compare and benchmark blockchains and smart contract systems, read Digital Catapult’s recent Medium article: “A Smarter Way of Benchmarking Blockchains: From Overall Measurements of Chains to Smart Contract Timings” written by Andy Baloiu.