This is the scenario you want to achieve: the team working at San Diego and the one at Bangalore need to share the same code base, but they need to run their own servers due to the slow and unreliable network connection. They can afford wasting time when the network goes down, so they both need to run their own servers.

Plastic makes it really simple, the are only 4 steps left!

Assuming the project already exists in San Diego, create a new repository at Bangalore to host it remotely. You can do that from the Plastic GUI or typing the following command if you're a real Plastic hacker:

$ cm mkrep bangalore:8084 replica

Now is time for the real replication: just run a "replicate" command to distribute the original main branch at San Diego to bangalore.

$ cm replicate br:/main@projectRep@sandiego:8084 bangalore:8084

In case your servers don't have direct access, you can create a replication package, deploy it yourself, and import it back at Bangalore.

San Diego:

$ cm replicate br:/main@projectRep@sandiego:8084 --package=replica.pkg

Bangalore:

$ cm replicate bangalore:8084 --import=replica.pkg

Repeat the process for any further branches you need to make available at both sites.

Now the two teams have their working environment correctly set up. Notice you just had to type 2 commands!
So the time to make the job has come... They work like crazy submitting code to Plastic SCM, creating branches, running code reviews, playing around with the 3D version tree...

At any point in time you'll need to reconcile changes back. Easy? Yes, really simple.
You can replicate again from San Diego to Bangalore and viceversa, running the same "replicate commands".

Plastic will detect if changes were made on the same branch and reconcile them back automatically, making use of its merging capabilities if any conflict occurs.

Multi-site development is not an obscure business anymore!