etseidl commented on PR #11267:
URL: https://github.com/apache/arrow-rs/pull/11267#issuecomment-5937487731
Cool summary. One thing I noticed
<details>
```
string_dictionary_1pct_cardinality/bloom_filter
1.00 21.3±0.13ms 47.0 MElem/sec
string_dictionary_1pct_cardinality/cdc
1.00 23.9±0.13ms 41.9 MElem/sec
string_dictionary_1pct_cardinality/default
1.00 20.8±0.12ms 48.1 MElem/sec
string_dictionary_1pct_cardinality/number_distinct_values
1.00 39.8±0.22ms 25.1 MElem/sec
string_dictionary_1pct_cardinality/parquet_2
1.00 21.6±0.16ms 46.3 MElem/sec
string_dictionary_1pct_cardinality/zstd
1.00 24.6±0.15ms 40.7 MElem/sec
string_dictionary_1pct_cardinality/zstd_parquet_2
1.00 24.4±0.15ms 41.0 MElem/sec
string_dictionary_low_cardinality_100/bloom_filter 1.02
22.0±0.07ms 181.9 MB/sec 1.00 21.7±0.07ms 46.2 MElem/sec
string_dictionary_low_cardinality_100/cdc 1.00
23.8±0.09ms 168.2 MB/sec 1.01 24.1±0.06ms 41.4 MElem/sec
string_dictionary_low_cardinality_100/default 1.02
21.9±0.07ms 183.3 MB/sec 1.00 21.5±0.08ms 46.4 MElem/sec
string_dictionary_low_cardinality_100/number_distinct_values 1.00
44.5±0.08ms 90.1 MB/sec 1.00 44.5±0.10ms 22.5 MElem/sec
string_dictionary_low_cardinality_100/parquet_2 1.00
21.7±0.08ms 184.7 MB/sec 1.00 21.7±0.05ms 46.0 MElem/sec
string_dictionary_low_cardinality_100/zstd 1.01
22.1±0.06ms 181.5 MB/sec 1.00 22.0±0.07ms 45.5 MElem/sec
string_dictionary_low_cardinality_100/zstd_parquet_2 1.01
22.0±0.06ms 182.1 MB/sec 1.00 21.8±0.05ms 45.8 MElem/sec
string_dictionary_low_cardinality_20/bloom_filter 1.00
21.0±0.05ms 191.1 MB/sec 1.03 21.6±0.09ms 46.4 MElem/sec
string_dictionary_low_cardinality_20/cdc 1.00
23.3±0.08ms 172.1 MB/sec 1.03 23.9±0.06ms 41.9 MElem/sec
string_dictionary_low_cardinality_20/default 1.00
20.9±0.06ms 191.9 MB/sec 1.04 21.7±0.10ms 46.2 MElem/sec
string_dictionary_low_cardinality_20/number_distinct_values 1.00
44.3±0.09ms 90.4 MB/sec 1.00 44.2±0.09ms 22.6 MElem/sec
string_dictionary_low_cardinality_20/parquet_2 1.03
22.0±0.08ms 182.4 MB/sec 1.00 21.3±0.06ms 47.1 MElem/sec
string_dictionary_low_cardinality_20/zstd 1.02
22.1±0.07ms 181.4 MB/sec 1.00 21.6±0.08ms 46.4 MElem/sec
string_dictionary_low_cardinality_20/zstd_parquet_2 1.03
22.1±0.06ms 181.5 MB/sec 1.00 21.4±0.06ms 46.8 MElem/sec
string_dictionary_low_cardinality_400/bloom_filter 1.00
22.3±0.07ms 180.0 MB/sec 1.01 22.4±0.23ms 44.6 MElem/sec
string_dictionary_low_cardinality_400/cdc 1.00
24.2±0.07ms 165.8 MB/sec 1.01 24.4±0.06ms 41.0 MElem/sec
string_dictionary_low_cardinality_400/default 1.00
22.3±0.07ms 180.3 MB/sec 1.02 22.6±0.13ms 44.2 MElem/sec
string_dictionary_low_cardinality_400/number_distinct_values 1.00
45.1±0.14ms 89.0 MB/sec 1.01 45.6±0.18ms 21.9 MElem/sec
string_dictionary_low_cardinality_400/parquet_2 1.00
21.7±0.05ms 185.0 MB/sec 1.04 22.5±0.13ms 44.5 MElem/sec
string_dictionary_low_cardinality_400/zstd 1.00
22.7±0.11ms 176.7 MB/sec 1.00 22.7±0.14ms 44.0 MElem/sec
string_dictionary_low_cardinality_400/zstd_parquet_2 1.00
22.5±0.10ms 178.3 MB/sec 1.01 22.6±0.10ms 44.2 MElem/sec
```
</details>
is the new string_dictionary_1pct markedly different from the other low
cardinality tests? They all seem to have similar execution times.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]