Clichouse v20.4 performance issue #11482
Description
Describe the situation
I'm seeing slower than expected queries on v20.4.4.18-stable compared to v20.1:20.1.4.14-stable.
My schema:
CREATE TABLE jaeger_index_v2 (
timestamp DateTime CODEC(Delta, ZSTD(1)),
traceID String CODEC(ZSTD(1)),
service LowCardinality(String) CODEC(ZSTD(1)),
operation LowCardinality(String) CODEC(ZSTD(1)),
durationUs UInt64 CODEC(ZSTD(1)),
tags Array(String) CODEC(ZSTD(1)),
INDEX idx_tags tags TYPE bloom_filter(0.01) GRANULARITY 64
) ENGINE MergeTree()
PARTITION BY toDate(timestamp)
ORDER BY (service, timestamp)
SETTINGS index_granularity=1024My query:
SELECT DISTINCT traceID
FROM jaeger_index_v2
PREWHERE service = 'nginx-fl'
AND timestamp >= now() - 3600
AND timestamp <= now()
ORDER BY service DESC,
timestamp DESC
LIMIT 20On v20.1 I see ~0.5s per query
On v20.4 I see a very different picture for 1.5s per query.