【查询时间过长】ngram-bloom索引下模糊查询慢

为了更快的定位您的问题,请提供以下信息,谢谢
【详述】问题详细描述
数据表字段添加了ngram-bloomfilter索引,使用ilike关键字做模糊查询,发现当数据无匹配会比有匹配消耗更长的时间,查看profile发现查询已经全表数据被bloomfilte过滤,正常情况下应该是无匹配更快才对吧?对比同数据规模的ck集群,无匹配数据能够达到毫秒级别返回,但是sr需要几十秒级别
【背景】做过哪些操作?
同上
【业务影响】
查询过慢
【是否存算分离】

【StarRocks版本】
4.0.1
【集群规模】
3fe+3be
【机器信息】CPU虚拟核/内存/网卡,例如:48C/64G/万兆
fe:4c 10g be:12c 80g
【联系方式】为了在解决问题过程中能及时联系到您获取一些日志信息,请补充下您的联系方式,例如:社区群4-小李或者邮箱,谢谢
【附件】

  • Profile信息,如何获取profile
  • 并行度:show variables like ‘%parallel_fragment_exec_instance_num%’;
    show variables like ‘%pipeline_dop%’;
  • pipeline是否开启:show variables like ‘%pipeline%’;
    mysql> show variables like ‘%pipeline%’;
    ±-------------------------------------------±------+
    | Variable_name | Value |
    ±-------------------------------------------±------+
    | enable_pipeline_engine | true |
    | enable_pipeline_event_scheduler | true |
    | enable_pipeline_level_multi_partitioned_rf | false |
    | max_pipeline_dop | 64 |
    | pipeline_dop | 1 |
    | pipeline_profile_level | 1 |
    | pipeline_sink_dop | 0 |
    ±-------------------------------------------±------+
  • 执行计划:explain costs + sql
    profile.txt (39.1 KB)

ngram-bloomfilter索引定义,非全部,只展示部分
INDEX gIndex_userSrcIp (userSrcIp) USING NGRAMBF(“bloom_filter_fpp” = “0.05”, “case_sensitive” = “false”, “gram_num” = “4”) COMMENT ‘’,
INDEX gIndex_pCommand (pCommand) USING NGRAMBF(“bloom_filter_fpp” = “0.05”, “case_sensitive” = “false”, “gram_num” = “4”) COMMENT ‘’,
INDEX gIndex_oldFileName (oldFileName) USING NGRAMBF(“bloom_filter_fpp” = “0.05”, “case_sensitive” = “false”, “gram_num” = “4”) COMMENT ‘’,
INDEX gIndex_dnsAnswers (dnsAnswers) USING NGRAMBF(“bloom_filter_fpp” = “0.05”, “case_sensitive” = “false”, “gram_num” = “4”) COMMENT ‘’,
INDEX gIndex_regeditNewName (regeditNewName) USING NGRAMBF(“bloom_filter_fpp” = “0.05”, “case_sensitive” = “false”, “gram_num” = “4”) COMMENT ‘’,
INDEX gIndex_driverPath (driverPath) USING NGRAMBF(“bloom_filter_fpp” = “0.05”, “case_sensitive” = “false”, “gram_num” = “4”) COMMENT ‘’,
INDEX gIndex_serviceName (serviceName) USING NGRAMBF(“bloom_filter_fpp” = “0.05”, “case_sensitive” = “false”, “gram_num” = “4”) COMMENT ‘’,
INDEX gIndex_pipeName (pipeName) USING NGRAMBF(“bloom_filter_fpp” = “0.05”, “case_sensitive” = “false”, “gram_num” = “4”) COMMENT ‘’,
INDEX gIndex_filterName (filterName) USING NGRAMBF(“bloom_filter_fpp” = “0.05”, “case_sensitive” = “false”, “gram_num” = “4”) COMMENT ‘’,