oceanbase

Parses SQL statements using a recursive descent parser that builds an abstract syntax tree (AST), then resolves table references, column names, and function calls against the internal schema system. The resolver validates semantic correctness by cross-referencing the internal table schema (ob_inner_table_schema) and type system before passing to the optimizer. Supports MySQL 5.7+ syntax including window functions, CTEs, and subqueries.

Applies cost-based optimization using cardinality estimation, table statistics, and join order enumeration to generate optimal physical execution plans. The optimizer evaluates multiple join orders (nested loop, hash join, merge join) and access paths (full scan, index scan, partition pruning) using a dynamic programming algorithm. Integrates with the plan cache to avoid re-optimization for identical query patterns.

Coordinates multi-tablet transactions using a two-phase commit (2PC) protocol where the transaction coordinator (typically the leader tablet) collects prepare votes from all participating tablets, then issues a global commit or rollback decision. The protocol uses write-ahead logging to ensure durability of the commit decision, and Paxos replication to ensure the decision survives coordinator failures. Supports both strong consistency (all-or-nothing) and eventual consistency modes for performance tuning.

Analyzes WHERE clause predicates during query optimization to identify which tablet partitions contain matching rows, then prunes partitions that cannot contain results. Pushes filter predicates down to the storage layer so that filtering happens during table scans rather than after rows are retrieved. Supports range pruning (for range-partitioned tables), hash pruning (for hash-partitioned tables), and list pruning (for list-partitioned tables). Integrates with the query optimizer to apply pruning before generating the execution plan.

Collects runtime statistics during query execution (rows processed, actual join cardinalities, predicate selectivity) and uses these statistics to adapt the execution plan mid-query. If actual cardinalities differ significantly from estimates, the executor can switch to a different join algorithm or access path without restarting the query. Statistics are fed back to the plan cache to improve future plan quality. Integrates with the SQL audit system (ob_gv_sql_audit) to track execution metrics.

Isolates multiple tenants within a single OceanBase cluster using logical tenant boundaries, resource quotas (CPU, memory, I/O), and access control lists. Each tenant has its own schema, data, and configuration, but shares underlying hardware resources. The resource manager enforces quotas by throttling queries that exceed allocated resources. Integrates with the session context to track tenant identity and apply tenant-specific configuration.

Executes physical plans across multiple tablet replicas by decomposing queries into remote RPC calls via the RPC communication framework. The executor routes data requests to the correct tablet partition based on the partition key, handles remote execution failures with automatic retry logic, and merges results from multiple tablets. Uses the ObRpcProcessor framework to serialize/deserialize query fragments and coordinate execution across nodes.

Implements multi-version concurrency control (MVCC) using row-level versioning where each row modification creates a new version with a transaction ID (txn_id) and commit timestamp. Readers acquire a consistent snapshot at a specific timestamp and only see versions committed before that timestamp, enabling concurrent reads and writes without blocking. The transaction manager maintains active transaction lists and coordinates version visibility across the cluster using the Paxos consensus protocol.

Replicates tablet data across multiple nodes using the Paxos consensus protocol, ensuring that writes are committed only after a quorum of replicas acknowledge the change. The leader replica coordinates write proposals, followers apply changes in log order, and the protocol handles leader failures by triggering new elections. Integrates with the tablet management system to track replica locations and membership changes.

Builds inverted indexes on text columns that map terms to row IDs, supporting phrase queries, boolean operators, and relevance ranking. The indexing system tokenizes text during INSERT/UPDATE operations and stores term frequencies for BM25 ranking. Query execution uses the inverted index to quickly locate matching rows, then applies ranking functions to order results by relevance. Integrates with the DDL system to support CREATE FULLTEXT INDEX statements.

Stores dense vector embeddings and supports approximate nearest neighbor (ANN) search using hierarchical navigable small-world (HNSW) or product quantization indexes. The vector engine computes similarity metrics (L2, cosine, inner product) and returns the K nearest neighbors ranked by distance. Integrates with the storage engine to support vector columns and with the query optimizer to push vector distance calculations into the execution plan.

Compiles PL/SQL procedures, functions, and packages into bytecode using the PL/SQL compiler (ob_pl_compile), then executes the bytecode in a virtual machine. The compiler performs syntax checking, type resolution, and code generation, storing compiled code in the package manager. Execution supports control flow (loops, conditionals), cursor operations, and calls to SQL statements and built-in packages (DBMS_SQL, DBMS_OUTPUT). Integrates with the session context to maintain procedure state and variable bindings.

Executes DDL operations (CREATE TABLE, ALTER TABLE, DROP TABLE) without blocking concurrent reads and writes using the DDL task scheduler. For column additions, uses a zero-copy approach where new columns are added to the schema metadata without rewriting existing rows; old rows lazily populate default values on read. The DDL service coordinates schema changes across all replicas using Paxos consensus, ensuring consistency. Supports online index creation and constraint addition without table locks.

Supports both row-oriented storage (for OLTP) and column-oriented storage (for OLAP) within the same database, allowing users to choose the optimal format per table. Row storage uses B+ trees for fast single-row lookups and updates; column storage uses compressed columnar format for fast analytical scans. The query optimizer automatically selects the appropriate storage format based on the query pattern (single-row vs. full-scan). Integrates with the tablet management system to store row and column data in separate tablet replicas.