GlassFish podcast on Clustering
By alexismp on Nov 06, 2007
In this third episode of the GlassFish Podcast, HA engineering lead Shreedhar Ganapathy discusses project Shoal and how it's applied to GlassFish clustering as a GMS (Group Management System) technology but also how it does in-memory replication to provide High Availability (depending on the profile used).
If you're interested in the max number of nodes in a cluster, buddy replication, recovery failure notification, sticky load-balancing, availability level, or distributed state cache, this interview gets into all of these. It also tries to go from first steps to setup the cluster to tips on running and managing it effectively. Finally it touches on what's next for clustering in GlassFish in the next 18 months.
This is a longer podcast that the earlier two (40 minutes). Feedback is welcome about what you think is the optimal duration (although I suspect this varies depending on one's commute).