System recovery experimentation

JKOK005 / nus-monorepo

Repo for NUS courses code commits

1 stars 0 forks source link

System recovery experimentation #14

Open JKOK005 opened 4 years ago

JKOK005 commented 4 years ago

The goal is to understand how throughput is impacted when our system scales down & up

This is to simulate node failure and recovery

One possible suggestion is to display a graph of throughput across time, cut a node off and then restore it

JKOK005 commented 4 years ago

Current system has leader failure recovery in built, such that when a leader dies, the slave will take over. We can use this mechanism to evaluate how long our system takes to recover.