Define etcd registry methods for scale subresource in Deployments and
ReplicaSets. Register them with the API server.

kube-up for GCE chooses master size based on number of nodes

Fix a breakage of kubelet in Trusty

Add the GCE cluster name to instance metadata

Pass through EXTRA_DOCKER_OPTS.

Update heapster to v0.20.0.alpha11

Revert "Move discovery client to its own package"

Auto commit by PR queue bot

Move discovery client to its own package

Update the logging images to parse sub-second docker timestamps.

Auto commit by PR queue bot

Auto commit by PR queue bot

Auto commit by PR queue bot

Auto commit by PR queue bot

Auto commit by PR queue bot

Auto commit by PR queue bot

Auto commit by PR queue bot

Auto commit by PR queue bot

Auto commit by PR queue bot

Auto commit by PR queue bot

Move scalability suite to dedicated project

Auto commit by PR queue bot

Auto commit by PR queue bot

Auto commit by PR queue bot

Auto commit by PR queue bot

Auto commit by PR queue bot

Auto commit by PR queue bot

Auto commit by PR queue bot

Auto commit by PR queue bot

Auto commit by PR queue bot

Auto commit by PR queue bot

Add some logging around deletion timestamps for RC scaling.