B-tree concurrency control and recovery in page-server database systems

Abstract

We develop new algorithms for the management of transactions in a page-shipping client-server database system in which the physical database is organized as a sparse B-tree index. Our starvation-free fine-grained locking protocol combines adaptive callbacks with key-range locking and guarantees repeatable-read-level isolation (i.e., serializability) for transactions containing any number of record insertions, record deletions, and key-range scans. Partial and total rollbacks of client transactions are performed by the client. Each structure modification such as a page split or merge is defined as an atomic action that affects only two levels of the B-tree and is logged using a single redo-only log record, so that the modification never needs to be undone during transaction rollback or restart recovery. The steal-and-no-force buffering policy is applied by the server when flushing updated pages onto disk and by the clients when shipping updated data pages to the server, while pages involved in a structure modification are forced to the server when the modification is finished. The server performs the restart recovery from client and system failures using an ARIES/CSA-based recovery protocol. Our algorithms avoid accessing stale data but allow a data page to be updated by one client transaction and read by many other client transactions simultaneously, and updates may migrate from a data page to another in structure modifications caused by other transactions while the updating transaction is still active.

Concurrent multidimensional index algorithms in page-server DBMS

“…Therefore, new and advanced R-tree concurrency-control, replica-management, and recovery protocols are needed for page-server systems. The only studies that are similar to ours appeared in [1,9,12,13,15,20]. But, these studies considered B-tree index algorithms in page-server systems.…”

“…Then T performs the structure modifications bottom-up on these pages, where the U locks on these pages are upgraded to X locks. Each tree structure modification is logged using physiological redo-undo log records [10,12,14,15,16]. Hence, in the event of a system failure, all incomplete structure modifications must be undone first so that aborted transactions can be rolled back during the undo phase of the restart recovery.…”

“…As in the first approach, when an updating transaction T reaches a leaf node and finds out that a tree structure modification is needed, then T releases its S lock on that node and retraverses the R-tree from the root using lock-coupling with U locks and executes structure modifications as atomic actions. Each structure modification upgrades the U locks to X-locks and updates three pages at a time on two adjacent levels in case of a page split (and one page on a single level of the R-tree in case of a minimum-bounding-rectangle modification), and is logged using a single physiological redo-only log record [12,14,15,16]. The X locks are released when the execution of an atomic structure modification is completed.…”

Scalable and dynamically balanced shared-everything OLTP with physiological partitioning

“…Partitioning the tree physically with MRBTrees eases the problem by distributing SMOs across subtrees (whose roots are fixed) without having to apply more complicated protocols, as such those described in [22,34]. The benefits of parallel SMOs are apparent in the case of insert-heavy workloads, which we evaluate in Sect.…”

“…Alternatives to traditional B+tree concurrency control are studied to allow multiple SMOs at the same time [22,34]. The MRBTree index structure provides an alternative to such techniques, allowing concurrent SMOs with less code complexity.…”