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 want a test case that will evaluate congestion in a multi-shard environment.
The main difference from the local congestion control is that the load that results in the congestion will be coming from other shards to the hot shard as incoming receipts.
The test should check that after a bursty load that is over the capacity of all shards, the transactions are properly queued up and eventually executed.
A good starting point will be reusing ComputeSum transaction, but issuing the requests to it from accounts on different shards.
The text was updated successfully, but these errors were encountered:
This is a continuation of work in #8920
We want a test case that will evaluate congestion in a multi-shard environment.
The main difference from the local congestion control is that the load that results in the congestion will be coming from other shards to the hot shard as incoming receipts.
We will reuse the same Locust setup, extending the workload defined in https://github.com/near/nearcore/blob/a5ede1d1cd54906ac91cbb614028b5323ddee0c0/pytest/tests/loadtest/locust/common/congestion.py.
The test should check that after a bursty load that is over the capacity of all shards, the transactions are properly queued up and eventually executed.
A good starting point will be reusing
ComputeSum
transaction, but issuing the requests to it from accounts on different shards.The text was updated successfully, but these errors were encountered: