mirror of
https://github.com/GreptimeTeam/greptimedb.git
synced 2026-10-14 16:03:27 +00:00
* feat: add per-column COMPRESSION option for SST columns CREATE TABLE accepts COMPRESSION WITH (type = 'zstd', level = N, page_size = '<size>') on a column. The options are stored in the column metadata, carried through gRPC column options, shown by SHOW CREATE TABLE, and applied by the mito SST writer and bulk part encoder as parquet column properties. Unset fields keep the SST defaults. Nested types are rejected. A page size below the dictionary page limit also lowers the column's dictionary page limit: parquet sizes write batches by the dictionary budget and keeps that size after falling back to plain encoding, which would otherwise ignore the smaller data page limit. Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * fix: validate column compression options before persistence Signed-off-by: Dennis Zhuang <killme2008@gmail.com> * fix: reject compression overrides on metric logical tables Signed-off-by: Dennis Zhuang <killme2008@gmail.com> --------- Signed-off-by: Dennis Zhuang <killme2008@gmail.com>
97 lines
3.0 KiB
SQL
97 lines
3.0 KiB
SQL
CREATE TABLE spans (
|
|
ts TIMESTAMP TIME INDEX,
|
|
trace_id STRING SKIPPING INDEX,
|
|
prompt STRING COMPRESSION WITH (type = 'zstd', level = 3, page_size = '8MiB'),
|
|
reply STRING COMPRESSION WITH (level = 9),
|
|
tokens BIGINT COMPRESSION WITH (page_size = '64KiB'),
|
|
) WITH (append_mode = 'true');
|
|
|
|
SHOW CREATE TABLE spans;
|
|
|
|
INSERT INTO spans VALUES
|
|
(1, 't1', 'fix the flaky test', 'running it', 120),
|
|
(2, 't1', 'fix the flaky test\nrunning it', 'patched', 340),
|
|
(3, 't2', 'another prompt', NULL, NULL);
|
|
|
|
ADMIN flush_table('spans');
|
|
|
|
SELECT ts, trace_id, prompt, reply, tokens FROM spans ORDER BY ts;
|
|
|
|
DROP TABLE spans;
|
|
|
|
CREATE TABLE bad (ts TIMESTAMP TIME INDEX, s STRING COMPRESSION WITH (level = 0));
|
|
|
|
CREATE TABLE bad (ts TIMESTAMP TIME INDEX, s STRING COMPRESSION WITH (type = 'lz4'));
|
|
|
|
CREATE TABLE bad (ts TIMESTAMP TIME INDEX, s STRING COMPRESSION WITH (page_size = 'large'));
|
|
|
|
CREATE TABLE bad (ts TIMESTAMP TIME INDEX, s STRING COMPRESSION WITH (codec = 'zstd'));
|
|
|
|
CREATE TABLE bad (ts TIMESTAMP TIME INDEX, j JSON2 COMPRESSION WITH (level = 3));
|
|
|
|
CREATE TABLE bad (ts TIMESTAMP TIME INDEX, s STRING COMPRESSION WITH (level = [3]));
|
|
|
|
CREATE TABLE bad (ts TIMESTAMP TIME INDEX, s STRING COMPRESSION WITH (page_size = NULL));
|
|
|
|
CREATE TABLE bad (
|
|
ts TIMESTAMP TIME INDEX,
|
|
host STRING PRIMARY KEY COMPRESSION WITH (level = 3)
|
|
) WITH ('sst_format' = 'primary_key', 'memtable.type' = 'time_series');
|
|
|
|
CREATE TABLE bad (
|
|
ts TIMESTAMP TIME INDEX,
|
|
host STRING PRIMARY KEY COMPRESSION WITH (level = 3)
|
|
) WITH ('primary_key_encoding' = 'sparse');
|
|
|
|
CREATE TABLE compressed_tags (
|
|
ts TIMESTAMP TIME INDEX,
|
|
host STRING PRIMARY KEY COMPRESSION WITH (level = 3, page_size = '64KiB'),
|
|
val DOUBLE
|
|
) WITH ('sst_format' = 'flat', 'primary_key_encoding' = 'dense');
|
|
|
|
SHOW CREATE TABLE compressed_tags;
|
|
|
|
INSERT INTO compressed_tags VALUES (1, 'host-a', 1), (2, 'host-b', 2);
|
|
|
|
ADMIN flush_table('compressed_tags');
|
|
|
|
SELECT * FROM compressed_tags ORDER BY ts;
|
|
|
|
DROP TABLE compressed_tags;
|
|
|
|
CREATE TABLE compression_phy (
|
|
ts TIMESTAMP TIME INDEX,
|
|
val DOUBLE COMPRESSION WITH (level = 3)
|
|
) ENGINE = metric WITH (physical_metric_table = '');
|
|
|
|
CREATE TABLE compression_conflict (
|
|
ts TIMESTAMP TIME INDEX,
|
|
val DOUBLE COMPRESSION WITH (level = 9)
|
|
) ENGINE = metric WITH (on_physical_table = 'compression_phy');
|
|
|
|
CREATE TABLE compression_same (
|
|
ts TIMESTAMP TIME INDEX,
|
|
val DOUBLE COMPRESSION WITH (level = 3)
|
|
) ENGINE = metric WITH (on_physical_table = 'compression_phy');
|
|
|
|
CREATE TABLE compression_new_tag (
|
|
ts TIMESTAMP TIME INDEX,
|
|
host STRING PRIMARY KEY COMPRESSION WITH (level = 3),
|
|
val DOUBLE
|
|
) ENGINE = metric WITH (on_physical_table = 'compression_phy');
|
|
|
|
CREATE TABLE compression_logical (
|
|
ts TIMESTAMP TIME INDEX,
|
|
val DOUBLE
|
|
) ENGINE = metric WITH (on_physical_table = 'compression_phy');
|
|
|
|
INSERT INTO compression_logical VALUES (1, 42);
|
|
|
|
ADMIN flush_table('compression_phy');
|
|
|
|
SELECT * FROM compression_logical;
|
|
|
|
DROP TABLE compression_logical;
|
|
|
|
DROP TABLE compression_phy;
|