InfoQ Homepage Reliability Content on InfoQ
-
MongoDB 1.8 Improves Reliability with Journaling
MongoDB's new journaling feature improves reliability with write-ahead redo logs. Log entries are written before permanent storage is updated. When a server restarts after a crash outstanding journal files will be replayed before the server goes online. Other changes include sharding performance boosts, shell tab completion, and the addition of covering and sparse indexes.
-
Google Releases the High Replication Datastore for App Engine
Google offers now two options for storage on its App Engine, the Master/Slave Datastore and the new High Replication Datastore, which remains available during downtime and offers a higher degree of resiliency to catastrophic failures.