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
We are looking into synthetic noise generators (issue #12), which would help test collectors (e.g., issue #11). In addition, we can show how larger collections of services behave when faced with memory noisy neighbor, by measuring application layer latency with/without synthetic generators.
Some options:
Many observability deployments use GCP's microservices-demo as a demonstration workload, and I think that would be a good start.
We are looking into synthetic noise generators (issue #12), which would help test collectors (e.g., issue #11). In addition, we can show how larger collections of services behave when faced with memory noisy neighbor, by measuring application layer latency with/without synthetic generators.
Some options:
cc @cdelimitrou as author of DeathStarBench, if you have an opinion of workload to show memory bandwidth and cache noisy neighbor.
The text was updated successfully, but these errors were encountered: