You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Develop a parsing tool for #4260, capturing all data in SQL format and allowing visualizations & queries for better identifying performance bottlenecks.
What problem would be solved by this feature?
This enables faster & more accurate debugging, allowing developers to find and improve bottlenecks, deadlocks, & timeouts more effectively. One example from some of the work done so far is the identification of potential correlations between process-confirmed and confirm_ack_received messages. So in theory, better prioritizing & processing incoming confirm_acks over new elections could improve stability and/or throughput under load.
This contributes towards the ultimate goal of achieving reliably steady TPS (even under saturation).
Summary
Develop a parsing tool for #4260, capturing all data in SQL format and allowing visualizations & queries for better identifying performance bottlenecks.
What problem would be solved by this feature?
This enables faster & more accurate debugging, allowing developers to find and improve bottlenecks, deadlocks, & timeouts more effectively. One example from some of the work done so far is the identification of potential correlations between process-confirmed and confirm_ack_received messages. So in theory, better prioritizing & processing incoming confirm_acks over new elections could improve stability and/or throughput under load.
This contributes towards the ultimate goal of achieving reliably steady TPS (even under saturation).
Initial Implementation
https://github.com/gr0vity-dev/nanolog-parser
The text was updated successfully, but these errors were encountered: