为了更快的定位您的问题,请提供以下信息,谢谢
【详述】单表百万级别,进行group by时会扫描上亿行数据,资源消耗巨大
【背景】重建表,更改表模型,ANALYZE TABLE尝试后扫描行数还是异常
【业务影响】
【是否存算分离】
【StarRocks版本】2.5.12
【集群规模】 3fe(1 follower+2observer)+5be(fe与be混部)
【机器信息】CPU虚拟核/内存/网卡,例如:48C/64G/万兆
【联系方式】m18710923628@163.com
表模型是这个
CREATE TABLE test1
(
t1
date NOT NULL ,
t2
varchar(36) NULL ,
t3
varchar(300) NULL ,
t4
int(11) NULL,
t5
int(11) NULL
) ENGINE=OLAP
UNIQUE KEY(t1
, t2
, t3
, t4
)
PARTITION BY RANGE(t1
)
(PARTITION p20240413 VALUES [(“2024-04-13”), (“2024-04-14”)),
PARTITION p20240414 VALUES [(“2024-04-14”), (“2024-04-15”)),
PARTITION p20240415 VALUES [(“2024-04-15”), (“2024-04-16”)),
PARTITION p20240416 VALUES [(“2024-04-16”), (“2024-04-17”)),
PARTITION p20240417 VALUES [(“2024-04-17”), (“2024-04-18”)),
PARTITION p20240418 VALUES [(“2024-04-18”), (“2024-04-19”)),
PARTITION p20240419 VALUES [(“2024-04-19”), (“2024-04-20”)),
PARTITION p20240420 VALUES [(“2024-04-20”), (“2024-04-21”)),
PARTITION p20240421 VALUES [(“2024-04-21”), (“2024-04-22”)),
PARTITION p20240422 VALUES [(“2024-04-22”), (“2024-04-23”)),
PARTITION p20240423 VALUES [(“2024-04-23”), (“2024-04-24”)),
PARTITION p20240424 VALUES [(“2024-04-24”), (“2024-04-25”)),
PARTITION p20240425 VALUES [(“2024-04-25”), (“2024-04-26”)),
PARTITION p20240426 VALUES [(“2024-04-26”), (“2024-04-27”)),
PARTITION p20240427 VALUES [(“2024-04-27”), (“2024-04-28”)),
PARTITION p20240428 VALUES [(“2024-04-28”), (“2024-04-29”)),
PARTITION p20240429 VALUES [(“2024-04-29”), (“2024-04-30”)),
PARTITION p20240430 VALUES [(“2024-04-30”), (“2024-05-01”)),
PARTITION p20240501 VALUES [(“2024-05-01”), (“2024-05-02”)),
PARTITION p20240502 VALUES [(“2024-05-02”), (“2024-05-03”)),
PARTITION p20240503 VALUES [(“2024-05-03”), (“2024-05-04”)),
PARTITION p20240504 VALUES [(“2024-05-04”), (“2024-05-05”)),
PARTITION p20240505 VALUES [(“2024-05-05”), (“2024-05-06”)),
PARTITION p20240506 VALUES [(“2024-05-06”), (“2024-05-07”)),
PARTITION p20240507 VALUES [(“2024-05-07”), (“2024-05-08”)),
PARTITION p20240508 VALUES [(“2024-05-08”), (“2024-05-09”)),
PARTITION p20240509 VALUES [(“2024-05-09”), (“2024-05-10”)),
PARTITION p20240510 VALUES [(“2024-05-10”), (“2024-05-11”)),
PARTITION p20240511 VALUES [(“2024-05-11”), (“2024-05-12”)),
PARTITION p20240512 VALUES [(“2024-05-12”), (“2024-05-13”)),
PARTITION p20240513 VALUES [(“2024-05-13”), (“2024-05-14”)),
PARTITION p20240514 VALUES [(“2024-05-14”), (“2024-05-15”)),
PARTITION p20240515 VALUES [(“2024-05-15”), (“2024-05-16”)),
PARTITION p20240516 VALUES [(“2024-05-16”), (“2024-05-17”)))
DISTRIBUTED BY HASH(t1
, t2
, t3
, t4
) BUCKETS 4
PROPERTIES (
“replication_num” = “3”,
“dynamic_partition.enable” = “true”,
“dynamic_partition.time_unit” = “DAY”,
“dynamic_partition.time_zone” = “Asia/Shanghai”,
“dynamic_partition.start” = “-30”,
“dynamic_partition.end” = “3”,
“dynamic_partition.prefix” = “p”,
“dynamic_partition.buckets” = “4”,
“dynamic_partition.history_partition_num” = “0”,
“in_memory” = “false”,
“storage_format” = “DEFAULT”,
“enable_persistent_index” = “false”,
“compression” = “LZ4”,
“storage_medium” = “SSD”
);
审计日志结果