From: thepipeline_xyz
Successfully managing high transaction volumes, especially during events like NFT mints, is a critical challenge for blockchain networks [00:00:04].
Challenges with High Transaction Volumes
When many users concurrently attempt to engage in high-stress activities, such as an NFT mint, it can lead to system failures [00:00:04]. Common issues observed on social media during such events include:
- RPC (Remote Procedure Call) failures [00:00:23]
- Transactions getting reverted [00:00:25]
The user experience on Solana in the fall of 2021, for instance, was characterized by widespread issues where people globally were “spam clicking mint buttons,” leading to system performance problems [00:00:37]. These kinds of high-stress scenarios are difficult to anticipate in advance [00:00:43].
Desired Outcomes for Robust Systems
The goal for performant blockchain infrastructure like Monad is to gracefully handle situations with 100,000 concurrent users sending 100 transactions per second [00:00:00]. The aim is to achieve a state where, even under super high-stress conditions with everyone “slamming the system,” there is no need to make excuses for why the system failed [00:00:16]. Continuous monitoring and data analysis are essential to further optimize performance [00:00:11].
Future Stress Testing with AI Agents
There is interest in using AI agents on the testnet to combine AI capabilities with high throughput systems, potentially yielding interesting results [00:00:50]. Incentivizing AI to intentionally try and crash the testnet could provide valuable insights into system weaknesses, differentiating from human clicking patterns by employing automated “robot clicking” [00:00:58].