Treffen der SIG JBoss
Thema: Memory is the new disk, disk is the new tape
ReplCache is a tool to store data in a cluster. It is essentially a hashmap spanning multiple hosts in a cluster.
There are put(), get() and remove() methods which distribute data over the cluster.
The unique feature of ReplCache is that it allows to define a replication count per element, e.g. how many times should an element be stored (on different nodes) for redundancy. When the cluster changes, e.g. due to new nodes joining or nodes crashing, elements are automatically rebalanced to satisfy replication counts.
Given enough nodes and the assumption that not all nodes ever crash at the exact same time, we can get rid of our database, and use the aggregated memory as our large virtual memory-based database!
(For the paranoid folks, we can still stream the data away to disk/DB).
We will also discuss the architectural approach for applications leveraging ReplCache and which for which kind of application this architecture will be suitable.
About Bela Ban
Bela Ban completed his PhD at the University of Zurich, Switzerland. After some time at IBM Research, he did a post-doc at Cornell. Then he worked on NMS/EMS for Fujitsu Network Communications in San Jose, California. In 2003, he joined JBoss to work full-time on open source. Bela manages the Clustering Team at JBoss and created and leads the JGroups project.
Bela"s interests include network protocols, performance, group communication, trail running, biking and beerathlon. If not hacking code he spends time with his family.