SLSM: An Efficient Strategy for Lazy Schema Migration on Shared-Nothing Databases

By introducing intermediate states for metadata changes, shared-nothing databases are capable of making online, asynchronous schema changes. However, this method leads to delays in the deployment of new schema since it requires waiting for massive data backfill. To shorten the service vacuum period before the new schema is available, this paper proposes a strategy named SLSM for zero-downtime schema migration on shared-nothing databases. SLSM migrates data from the old schema on-demand before processing incoming transactions on the new schema and addresses the performance issue of this lazy migration. First, it keeps the old and new schema with the same data distribution, reducing the node communication overhead of data migration. Further, SLSM combines data migration with transaction processing by extending the execution plan to allow the data involved in the migration to directly serve the transaction, greatly reducing the waiting time of transaction execution. Experiments demonstrate that our strategy can greatly reduce the latency of transactions and improve the efficiency of data migration compared to existing schemes.