PostgreSQL
Versions
| Run as | Image | Server | DuckDB | Digest |
|---|---|---|---|---|
| PostgreSQL | postgres:18 | PostgreSQL 18 | n/a | sha256:edb0c08dc33befd0262ab3b5cf1c4f3f8cfd8b09e88c05b45adee29d6761ddb6 |
Observed in this run
Generated the same way for every system from its result files: record counts per measure and status, and the reasons given for anything that is not ok.
PostgreSQL
- M1: ok 20, timeout 2
- M2: n/a 1, ok 1
- M3: ok 2
- M5: n/a 120
- M1 q18 (timeout): QueryCanceled: canceling statement due to statement timeout
- M1 q20 (timeout): QueryCanceled: canceling statement due to statement timeout
- M2 load_convert (n/a): row-store baseline: no DuckDB copy
- M5 cte_multiple (n/a): row-store baseline; reference for the probe suite
- M5 cte_plain (n/a): row-store baseline; reference for the probe suite
- M5 cte_recursive_counter (n/a): row-store baseline; reference for the probe suite
- M5 cte_recursive_hierarchy (n/a): row-store baseline; reference for the probe suite
- M5 cte_recursive_path (n/a): row-store baseline; reference for the probe suite
- M5 cte_referenced_twice (n/a): row-store baseline; reference for the probe suite
- M5 cte_tpch_aggregate (n/a): row-store baseline; reference for the probe suite
- M5 ddl_add_column (n/a): row-store baseline; reference for the probe suite
- M5 ddl_add_column_default (n/a): row-store baseline; reference for the probe suite
- and 111 more
Settings that differ from the defaults
| Setting | Value | Reason | File |
|---|---|---|---|
listen_addresses | '*' | The runner container connects over the docker bridge. | systems/postgres/postgresql.conf |
shared_buffers | 16GB | 25% of container memory, the usual starting point; the OS page cache holds the rest. | systems/postgres/postgresql.conf |
effective_cache_size | 48GB | Planner's estimate of buffer plus page cache: 75% of container memory. | systems/postgres/postgresql.conf |
max_parallel_workers_per_gather | 16 | Allow one query to use every core given to the database. | systems/postgres/postgresql.conf |
max_parallel_workers | 16 | Cap on parallel workers across all sessions: one per core. | systems/postgres/postgresql.conf |
max_worker_processes | 16 | Must be at least max_parallel_workers for those workers to start. | systems/postgres/postgresql.conf |
work_mem | 256MB | Analytical sorts and hashes stay in memory instead of spilling at the 4MB default. | systems/postgres/postgresql.conf |
max_connections | 200 | Headroom for the 32-stream throughput run plus probes and cancel connections. | systems/postgres/postgresql.conf |
max_wal_size | 16GB | WAL of mem_gb/4 keeps checkpoints from throttling bulk load at the 1GB default. | systems/postgres/postgresql.conf |
checkpoint_timeout | 30min | Timed checkpoints every 30 minutes, so load and query runs are not interrupted by them. | systems/postgres/postgresql.conf |
README
PostgreSQL 18 baseline
Image postgres:18 (digest recorded in every result as versions.image_digest), started with -c config_file=/etc/postgresql/postgresql.conf, the file rendered from postgresql.conf and mounted read-only. The dataset is mounted read-only at /data, shm_size is 8 GB, and authentication is trust on the private docker network.
Non-default settings
Rendered from the container limits (dbench/systems/baseline_conf.py); at 64 GB and 16 cores:
| Setting | Value | Reason |
|---|---|---|
listen_addresses | * | The runner container connects over the docker bridge (the custom config file replaces the default one). |
shared_buffers | 0.25 x container GB = 16GB | The usual starting point; the OS page cache holds the rest. |
effective_cache_size | 0.75 x container GB = 48GB | Planner's estimate of buffer plus page cache. |
max_parallel_workers_per_gather | cores in the DB cpuset = 16 | One query may use every core given to the database. |
max_parallel_workers | 16 | Cap across sessions: one worker per core. |
max_worker_processes | 16 | Must be at least max_parallel_workers. |
work_mem | 256MB | Analytical sorts and hashes stay in memory instead of spilling at the 4MB default. |
max_wal_size | mem_gb/4 = 16GB | Keeps checkpoints from throttling bulk load at the 1GB default. |
checkpoint_timeout | 30min | Timed checkpoints do not interrupt load and query runs. |
max_connections | 200 | Headroom over the 32-stream throughput run plus probe and cancel connections. |
Everything else stays at the PostgreSQL 18 default (including JIT).
Loading and measurement notes
COPY t FROM '/data/...' WITH (FORMAT csv, DELIMITER '|', QUOTE '"')per CSV part.- Row counts are checked against the manifest, then
VACUUM (FREEZE, ANALYZE)runs per table. Load time covers the copies and the vacuums, not the count check. The vacuum sets hint bits and freezes tuples so cold runs do not pay for it. - Tables are loaded with the primary key and foreign-key indexes already in place.
- Queries run after
SET statement_timeout;cancel()ispg_cancel_backendfrom a second connection. storage_bytes()isdu -sb $PGDATAafter aCHECKPOINT. Storage rule for the Postgres family: WAL is counted (replication logs are excluded in the MySQL family, redo/WAL are counted).- Queries also have a client-side deadline at timeout + 30 s:
pg_cancel_backend, thenpg_terminate_backendafter 10 s, recorded astimeout. This protects engines that ignorestatement_timeout.
Schema
TPC-H tables with primary keys, plus a plain index on every foreign-key column (row-store baseline rule); no other indexes and no declared foreign-key constraints. Strings are VARCHAR(n) with the spec lengths. The default collation of the image is kept. There are no schema deviations.
Dialect
The 22 DuckDB query texts run unchanged.
M11 and M12 are n/a: there is no DuckDB copy to convert or push down to.