Even not working optimally?

Nobody has claimed this yet.

The `even` mode in `balance` is the following:
```
even - This is a specialized type for clusters which have topics that have a number of partitions that are a multiple of the number of brokers. The module assures that every broker has exactly the same number of partitions for each topic.
```

Yet the code tries to SSH to my brokers?
```
[ec2-user@ip-172-31-12-216 ~]$ kafka-assigner -z $ZK_HOST -e balance --types even
[INFO] Connecting to zookeeper 172.31.9.21:2181
[INFO] Getting partition list from Zookeeper
[INFO] Closing connection to zookeeper
[INFO] Getting partition sizes via SSH for 172.31.32.31
[INFO] Getting partition sizes via SSH for 172.31.9.32
[INFO] Getting partition sizes via SSH for 172.31.16.33
[INFO] Getting partition sizes via SSH for 172.31.16.34
[INFO] Starting even partition balance
```

I would have guessed the `even` module could just connect to Zookeeper to distribute partitions evenly?

Start at the `even` mode in `balance` and reproduce the documented command with the shown SSH logs. Compare the mode's documented partition-distribution behavior with its SSH-based partition-size collection; done means the implementation and documentation agree about whether SSH is required.

Written by the indexing model from the issue text.