I have a few questions about how BriskSimpleSnitch works:
1) Say I have a Cassandra cluster across regions with the Cassandra vanilla cluster in Amazon's Virginia Datacenter and a Brisk cluster in California. When configuring the seed nodes for Cassandra & Brisk, do I have to use Amazon Elastic IPs?
2) Is BriskSimpleSnitch the only cross-region snitch that works right now? Can you link me to any documentation about the brisksnitch? (the user-mailing list for Cassandra shows 0 results for "BriskSimpleSnitch") BriskSimpleSnitch treats regions as data centers and availability zones as racks, right?
3) The DataStax Brisk AMI seems to launch all nodes (Cassandra vanilla & brisk) into the same region and availability zone. Any plans on allowing an option to launch Cassandra vanilla in one region (with half nodes in one Availability zone and other half in another Availability zone) and the Brisk nodes in a different region? This would be the best way to provide HA+DR when using Cassandra.
Currently, we just pass the parameters into the text user data part of the AWS Launch GUI. Either we'd need to add the option to pass more parameters (regions/availability zones) or need a different "master script" that launches sub-scripts for clusters in different regions. Thoughts?
If no plans, then we'll look into modifying the AMI and share it with you guys.