Metrics to track average transaction latency

[cmwaters]

Feature Request

Summary

Based on recent discussions, what might perhaps be the most important metric we want to optimise for outside of throughput is average transaction latency. This is the average time from when a transaction is received until when it is committed.

We should track this. This probably needs to be done by tracking total latency and dividing by the amount of transactions submitted. This will provide a rough answer but it may also be helpful to know other attributes like the standard deviation. One last important consideration is that we may want to make a distinction between when a transaction comes from a peer and when it comes from the original sender

---

For Admin Use

• Not duplicate issue
• Appropriate labels applied
• Appropriate contributors tagged
• Contributor assigned/self-assigned