AI 助理
备案控制台
开发者社区 开发与运维 文章 正文

【数据库】Star Schema Benchmark 标准测试集优化（三）

2022-12-05 107
版权
版权声明：
本文内容由阿里云实名注册用户自发贡献，版权归原作者所有，阿里云开发者社区不拥有其著作权，亦不承担相应法律责任。具体规则请查看《 阿里云开发者社区用户服务协议》和 《阿里云开发者社区知识产权保护指引》。如果您发现本社区中有涉嫌抄袭的内容，填写 侵权投诉表单进行举报，一经查实，本社区将立刻删除涉嫌侵权内容。
简介： 【数据库】Star Schema Benchmark 标准测试集优化（三）

正文

这是Star Schema Benchmark 标准测试集优化的第三篇，前一篇我们分析了下表数据，这一篇是最后一篇了。


一、分析算法路径


更新到前一篇的时候，其实专利技术已经集成到数据库中了，这个算法路径，主要是验证一下：在测试环境中的算法路径，是否和开发环境中一致。实际结果如下，13 条SQL语句的算法路径和开发环境中的算法路径，经过验证是完全一致的。

2022-10-20 01:39:53.344 - SQL2: select sum(lo_revenue) as revenue from lineorder,dates where lo_orderdate = d_datekey and d_year = 1993 and lo_discount between 1 and 3 and lo_quantity < 25; ;
  Status: PASS, Elapsed: 3.923, Affected: 1
  Info: Job[67:RESTRICT, tasks:193, time:3,591, size: 3,576,178, [0|193|3,591] ]
Job[69:DIMENSION_JOIN, tasks:193, time:16, size: 3,576,178, [0|193|15] ]
Job[71:OUTPUT_HASH_GROUP_BY, tasks:6, time:166, size: 6, [0|6|167] ]
Job[73:MERGE_HASH_GROUPBY_PARTITION, tasks:1, time:3, size: 1, [0|1|3] ]
Job[75:PROJECT, tasks:1, time:2, size: 1, [0|1|1] ]
total time: 3,788, record count: 1
restrict的时间/整体时间: 3591/3788
result:
REVENUE       |
- - - - - - -
6568512155417 |
2022-10-20 01:39:59.707 - SQL4: select sum(lo_revenue) as revenue from lineorder,dates where lo_orderdate = d_datekey and d_yearmonthnum = 199401 and lo_discount between 4 and 6 and lo_quantity between 26 and 35; ;
  Status: PASS, Elapsed: 1.767, Affected: 1
  Info: Job[99:RESTRICT, tasks:193, time:1,727, size: 126,610, [0|193|1,728] ]
Job[101:DIMENSION_JOIN, tasks:193, time:8, size: 126,610, [0|193|8] ]
Job[103:OUTPUT_HASH_GROUP_BY, tasks:2, time:22, size: 2, [0|2|22] ]
Job[105:MERGE_HASH_GROUPBY_PARTITION, tasks:1, time:1, size: 1, [0|1|1] ]
Job[107:PROJECT, tasks:1, time:1, size: 1, [0|1|1] ]
total time: 1,764, record count: 1
restrict的时间/整体时间: 1727/1764
result:
REVENUE      |
- - - - - - -
550150245374 |
2022-10-20 01:40:04.552 - SQL6: select sum(lo_revenue) as revenue from lineorder,dates where lo_orderdate = d_datekey and d_weeknuminyear = 6 and d_year = 1994 and lo_discount between 5 and 7 and lo_quantity between 26 and 35; ;
  Status: PASS, Elapsed: 2.138, Affected: 1
  Info: Job[131:RESTRICT, tasks:193, time:2,108, size: 28,441, [0|193|2,109] ]
Job[133:DIMENSION_JOIN, tasks:193, time:7, size: 28,441, [0|193|7] ]
Job[135:OUTPUT_HASH_GROUP_BY, tasks:1, time:15, size: 1, [0|1|15] ]
Job[137:MERGE_HASH_GROUPBY_PARTITION, tasks:1, time:1, size: 1, [0|1|1] ]
Job[139:PROJECT, tasks:1, time:1, size: 1, [0|1|1] ]
total time: 2,136, record count: 1
restrict的时间/整体时间: 2108/2136
result:
REVENUE      |
- - - - - - -
122223605792 |
2022-10-20 01:40:31.923 - SQL8: select sum(lo_revenue) as lo_revenue, d_year, p_brand from lineorder ,dates,part,supplier where lo_orderdate = d_datekey and lo_partkey = p_partkey and lo_suppkey = s_suppkey and p_category = 'MFGR#12' and s_region = 'AMERICA' group by d_year, p_brand order by d_year, p_brand; ;
  Status: PASS, Elapsed: 4.714, Affected: 280
  Info: Job[193:RESTRICT, tasks:193, time:4,342, size: 1,442,564, [0|193|4,342] ]
Job[195:DIMENSION_JOIN, tasks:193, time:9, size: 1,442,564, [0|193|10] ]
Job[197:DIMENSION_JOIN, tasks:193, time:5, size: 1,442,564, [0|193|5] ]
Job[199:DIMENSION_JOIN, tasks:193, time:5, size: 1,442,564, [0|193|5] ]
Job[202:OUTPUT_HASH_GROUP_BY, tasks:6, time:332, size: 1,680, [0|6|332] ]
Job[204:MERGE_HASH_GROUPBY_PARTITION, tasks:7, time:4, size: 280, [0|7|4] ]
Job[206:PROJECT, tasks:7, time:3, size: 280, [0|7|3] ]
Job[208:RANGE_SORT, tasks:7, time:2, size: 280, [0|7|2] ]
Job[212:PARTITION_TABLET, tasks:7, time:0, size: 280, [0|7|0] ]
Job[214:MERGE_ORDERBY_RANGES, tasks:1, time:1, size: 70, [0|1|1] ]
total time: 4,711, record count: 280
restrict的时间/整体时间: 4342/4711
result:
LO_REVENUE  |D_YEAR     |P_BRAND    |
- - - - - - - - - - - - - - - - - -
18712903257 |1992       |MFGR#121   |
20576919851 |1992       |MFGR#1210  |
20452654696 |1992       |MFGR#1211  |
2022-10-20 01:40:37.664 - SQL10: select sum(lo_revenue) as lo_revenue, d_year, p_brand from lineorder,dates,part,supplier where lo_orderdate = d_datekey and lo_partkey = p_partkey and lo_suppkey = s_suppkey and p_brand between 'MFGR#2221' and 'MFGR#2228' and s_region = 'ASIA' group by d_year, p_brand order by d_year, p_brand; ;
  Status: PASS, Elapsed: 1.24, Affected: 56
  Info: Job[292:RESTRICT, tasks:193, time:1,017, size: 287,885, [0|193|1,017] ]
Job[294:DIMENSION_JOIN, tasks:193, time:5, size: 287,885, [0|193|5] ]
Job[296:DIMENSION_JOIN, tasks:193, time:5, size: 287,885, [0|193|5] ]
Job[298:DIMENSION_JOIN, tasks:193, time:4, size: 287,885, [0|193|4] ]
Job[301:OUTPUT_HASH_GROUP_BY, tasks:3, time:193, size: 168, [0|3|193] ]
Job[303:MERGE_HASH_GROUPBY_PARTITION, tasks:7, time:2, size: 56, [0|7|2] ]
Job[305:PROJECT, tasks:7, time:1, size: 56, [0|7|1] ]
Job[307:RANGE_SORT, tasks:7, time:1, size: 56, [0|7|1] ]
Job[311:PARTITION_TABLET, tasks:7, time:0, size: 56, [0|7|0] ]
Job[313:MERGE_ORDERBY_RANGES, tasks:1, time:0, size: 12, [0|1|0] ]
total time: 1,235, record count: 56
restrict的时间/整体时间: 1017/1235
result:
LO_REVENUE  |D_YEAR     |P_BRAND    |
- - - - - - - - - - - - - - - - - -
19803695538 |1992       |MFGR#2221  |
19639734537 |1992       |MFGR#2222  |
19945070508 |1992       |MFGR#2223  |
2022-10-20 01:40:39.888 - SQL12: select sum(lo_revenue) as lo_revenue, d_year, p_brand from lineorder,dates,part,supplier where lo_orderdate = d_datekey and lo_partkey = p_partkey and lo_suppkey = s_suppkey and p_brand = 'MFGR#2239' and s_region = 'EUROPE' group by d_year, p_brand order by d_year, p_brand; ;
  Status: PASS, Elapsed: 0.862, Affected: 7
  Info: Job[397:RESTRICT, tasks:193, time:785, size: 35,599, [0|193|785] ]
Job[399:DIMENSION_JOIN, tasks:193, time:4, size: 35,599, [0|193|4] ]
Job[401:DIMENSION_JOIN, tasks:193, time:3, size: 35,599, [0|193|3] ]
Job[403:DIMENSION_JOIN, tasks:193, time:3, size: 35,599, [0|193|3] ]
Job[406:OUTPUT_HASH_GROUP_BY, tasks:1, time:49, size: 7, [0|1|49] ]
Job[408:MERGE_HASH_GROUPBY_PARTITION, tasks:7, time:8, size: 7, [0|7|8] ]
Job[410:PROJECT, tasks:7, time:0, size: 7, [0|7|0] ]
Job[412:RANGE_SORT, tasks:5, time:1, size: 7, [0|5|1] ]
Job[416:PARTITION_TABLET, tasks:5, time:0, size: 7, [0|5|0] ]
Job[418:MERGE_ORDERBY_RANGES, tasks:1, time:0, size: 2, [0|1|0] ]
total time: 859, record count: 7
restrict的时间/整体时间: 785/859
result:
LO_REVENUE  |D_YEAR     |P_BRAND    |
- - - - - - - - - - - - - - - - - -
19700225276 |1992       |MFGR#2239  |
19306484466 |1993       |MFGR#2239  |
19398411013 |1994       |MFGR#2239  |
2022-10-20 01:41:02.935 - SQL14: select c_nation, s_nation, d_year, sum(lo_revenue) as lo_revenue from lineorder,dates,customer,supplier where lo_orderdate = d_datekey and lo_custkey = c_custkey and lo_suppkey = s_suppkey and c_region = 'ASIA' and s_region = 'ASIA'and d_year >= 1992 and d_year <= 1997 group by c_nation, s_nation, d_year order by d_year asc, lo_revenue desc; ;
  Status: PASS, Elapsed: 5.015, Affected: 150
  Info: Job[502:RESTRICT, tasks:193, time:3,856, size: 6,570,093, [0|193|3,857] ]
Job[504:DIMENSION_JOIN, tasks:193, time:5, size: 6,570,093, [0|193|4] ]
Job[506:DIMENSION_JOIN, tasks:193, time:3, size: 6,570,093, [0|193|3] ]
Job[508:DIMENSION_JOIN, tasks:193, time:3, size: 6,570,093, [0|193|3] ]
Job[511:OUTPUT_HASH_GROUP_BY, tasks:6, time:1,132, size: 900, [0|6|1,132] ]
Job[513:MERGE_HASH_GROUPBY_PARTITION, tasks:7, time:3, size: 150, [0|7|3] ]
Job[515:PROJECT, tasks:7, time:1, size: 150, [0|7|1] ]
Job[517:RANGE_SORT, tasks:7, time:1, size: 150, [0|7|1] ]
Job[521:PARTITION_TABLET, tasks:7, time:0, size: 150, [0|7|0] ]
Job[523:MERGE_ORDERBY_RANGES, tasks:1, time:1, size: 22, [0|1|1] ]
total time: 5,012, record count: 150
restrict的时间/整体时间: 3856/5012
result:
C_NATION   |S_NATION   |D_YEAR     |LO_REVENUE   |
- - - - - - - - - - - - - - - - - - - - - - - - -
JAPAN      |INDIA      |1992       |163691240866 |
CHINA      |INDONESIA  |1992       |163434081261 |
CHINA      |INDIA      |1992       |163430796231 |
2022-10-20 01:41:11.469 - SQL16: select c_city, s_city, d_year, sum(lo_revenue) as lo_revenue from lineorder,dates,customer,supplier where lo_orderdate = d_datekey and lo_custkey = c_custkey and lo_suppkey = s_suppkey and  c_nation = 'UNITED STATES' and s_nation = 'UNITED STATES' and d_year >= 1992 and d_year <= 1997 group by c_city, s_city, d_year order by d_year asc, lo_revenue desc; ;
  Status: PASS, Elapsed: 3.478, Affected: 600
  Info: Job[589:RESTRICT, tasks:193, time:3,262, size: 264,531, [0|193|3,262] ]
Job[591:DIMENSION_JOIN, tasks:193, time:4, size: 264,531, [0|193|4] ]
Job[593:DIMENSION_JOIN, tasks:193, time:4, size: 264,531, [0|193|4] ]
Job[595:DIMENSION_JOIN, tasks:193, time:3, size: 264,531, [0|193|3] ]
Job[598:OUTPUT_HASH_GROUP_BY, tasks:3, time:189, size: 1,800, [0|3|189] ]
Job[600:MERGE_HASH_GROUPBY_PARTITION, tasks:7, time:5, size: 600, [0|7|5] ]
Job[602:PROJECT, tasks:7, time:1, size: 600, [0|7|1] ]
Job[604:RANGE_SORT, tasks:7, time:1, size: 600, [0|7|1] ]
Job[608:PARTITION_TABLET, tasks:7, time:0, size: 600, [0|7|0] ]
Job[610:MERGE_ORDERBY_RANGES, tasks:1, time:0, size: 100, [0|1|0] ]
total time: 3,475, record count: 600
restrict的时间/整体时间: 3262/3475
result:
C_CITY     |S_CITY     |D_YEAR     |LO_REVENUE |
- - - - - - - - - - - - - - - - - - - - - - - -
UNITED ST3 |UNITED ST4 |1992       |1915435842 |
UNITED ST8 |UNITED ST0 |1992       |1910327375 |
UNITED ST5 |UNITED ST0 |1992       |1893024189 |
2022-10-20 01:41:16.505 - SQL18: select c_city, s_city, d_year, sum(lo_revenue) as lo_revenue from lineorder,dates,customer,supplier where lo_orderdate = d_datekey and lo_custkey = c_custkey and lo_suppkey = s_suppkey and (c_city='UNITED KI1' or c_city='UNITED KI5') and (s_city='UNITED KI1' or s_city='UNITED KI5') and d_year >= 1992 and d_year <= 1997 group by c_city, s_city, d_year order by d_year asc, lo_revenue desc; ;
  Status: PASS, Elapsed: 1.588, Affected: 24
  Info: Job[679:RESTRICT, tasks:193, time:1,525, size: 10,616, [0|193|1,525] ]
Job[681:DIMENSION_JOIN, tasks:193, time:6, size: 10,616, [0|193|6] ]
Job[683:DIMENSION_JOIN, tasks:193, time:4, size: 10,616, [0|193|4] ]
Job[685:DIMENSION_JOIN, tasks:193, time:3, size: 10,616, [0|193|3] ]
Job[688:OUTPUT_HASH_GROUP_BY, tasks:1, time:35, size: 24, [0|1|35] ]
Job[690:MERGE_HASH_GROUPBY_PARTITION, tasks:7, time:2, size: 24, [0|7|2] ]
Job[692:PROJECT, tasks:7, time:0, size: 24, [0|7|0] ]
Job[694:RANGE_SORT, tasks:7, time:1, size: 24, [0|7|1] ]
Job[698:PARTITION_TABLET, tasks:7, time:0, size: 24, [0|7|0] ]
Job[700:MERGE_ORDERBY_RANGES, tasks:1, time:1, size: 3, [0|1|1] ]
total time: 1,585, record count: 24
restrict的时间/整体时间: 1525/1585
result:
C_CITY     |S_CITY     |D_YEAR     |LO_REVENUE |
- - - - - - - - - - - - - - - - - - - - - - - -
UNITED KI1 |UNITED KI1 |1992       |1786080690 |
UNITED KI5 |UNITED KI1 |1992       |1705128984 |
UNITED KI1 |UNITED KI5 |1992       |1620054330 |
2022-10-20 01:41:19.166 - SQL20: select c_city, s_city, d_year, sum(lo_revenue) as lo_revenue from lineorder,dates,customer,supplier where lo_orderdate = d_datekey and lo_custkey = c_custkey and lo_suppkey = s_suppkey and (c_city='UNITED KI1' or c_city='UNITED KI5') and (s_city='UNITED KI1' or s_city='UNITED KI5') and d_yearmonth  = 'Dec1997' group by c_city, s_city, d_year order by d_year asc, lo_revenue desc; ;
  Status: PASS, Elapsed: 1.046, Affected: 4
  Info: Job[775:RESTRICT, tasks:193, time:1,012, size: 151, [0|193|1,012] ]
Job[777:DIMENSION_JOIN, tasks:109, time:4, size: 151, [0|109|4] ]
Job[779:DIMENSION_JOIN, tasks:109, time:4, size: 151, [0|109|4] ]
Job[781:DIMENSION_JOIN, tasks:109, time:3, size: 151, [0|109|3] ]
Job[784:OUTPUT_HASH_GROUP_BY, tasks:1, time:13, size: 4, [0|1|13] ]
Job[786:MERGE_HASH_GROUPBY_PARTITION, tasks:7, time:1, size: 4, [0|7|1] ]
Job[788:PROJECT, tasks:7, time:1, size: 4, [0|7|1] ]
Job[790:RANGE_SORT, tasks:2, time:0, size: 4, [0|2|0] ]
Job[794:PARTITION_TABLET, tasks:2, time:0, size: 4, [0|2|0] ]
Job[796:MERGE_ORDERBY_RANGES, tasks:1, time:1, size: 2, [0|1|1] ]
total time: 1,044, record count: 4
restrict的时间/整体时间: 1012/1044
result:
C_CITY     |S_CITY     |D_YEAR     |LO_REVENUE |
- - - - - - - - - - - - - - - - - - - - - - - -
UNITED KI5 |UNITED KI1 |1997       |168840628  |
UNITED KI1 |UNITED KI5 |1997       |140264663  |
UNITED KI1 |UNITED KI1 |1997       |135684305  |
2022-10-20 01:41:29.956 - SQL22: select d_year, c_nation, sum(lo_revenue) - sum(lo_supplycost) as profit from lineorder,dates,customer,supplier,part where lo_orderdate = d_datekey and lo_custkey = c_custkey and lo_suppkey = s_suppkey and lo_partkey = p_partkey and c_region = 'AMERICA' and s_region = 'AMERICA' and (p_mfgr = 'MFGR#1' or p_mfgr = 'MFGR#2') group by d_year, c_nation order by d_year, c_nation; ;
  Status: PASS, Elapsed: 2.833, Affected: 35
  Info: Job[878:RESTRICT, tasks:193, time:2,137, size: 2,882,137, [0|193|2,138] ]
Job[880:DIMENSION_JOIN, tasks:193, time:8, size: 2,882,137, [0|193|8] ]
Job[882:DIMENSION_JOIN, tasks:193, time:3, size: 2,882,137, [0|193|3] ]
Job[884:DIMENSION_JOIN, tasks:193, time:2, size: 2,882,137, [0|193|2] ]
Job[886:DIMENSION_JOIN, tasks:193, time:3, size: 2,882,137, [0|193|2] ]
Job[889:OUTPUT_HASH_GROUP_BY, tasks:6, time:663, size: 210, [0|6|662] ]
Job[891:MERGE_HASH_GROUPBY_PARTITION, tasks:7, time:1, size: 35, [0|7|1] ]
Job[893:PROJECT, tasks:7, time:1, size: 35, [0|7|1] ]
Job[895:PROJECT, tasks:7, time:0, size: 35, [0|7|0] ]
Job[897:RANGE_SORT, tasks:7, time:1, size: 35, [0|7|1] ]
Job[901:PARTITION_TABLET, tasks:7, time:0, size: 35, [0|7|0] ]
Job[903:MERGE_ORDERBY_RANGES, tasks:1, time:0, size: 1, [0|1|0] ]
total time: 2,830, record count: 35
restrict的时间/整体时间: 2137/2830
result:
D_YEAR     |C_NATION      |PROFIT       |
- - - - - - - - - - - - - - - - - - - -
1992       |ARGENTINA     |312585625436 |
1992       |BRAZIL        |312719709853 |
1992       |CANADA        |307040911677 |
2022-10-20 01:41:37.244 - SQL24: select d_year, s_nation, p_category, sum(lo_revenue) - sum(lo_supplycost) as profit from lineorder,dates,customer,supplier,part where lo_orderdate = d_datekey and lo_custkey = c_custkey and lo_suppkey = s_suppkey and lo_partkey = p_partkey and c_region = 'AMERICA'and s_region = 'AMERICA' and (d_year = 1997 or d_year = 1998) and (p_mfgr = 'MFGR#1' or p_mfgr = 'MFGR#2') group by d_year, s_nation, p_category order by d_year, s_nation, p_category; ;
  Status: PASS, Elapsed: 4.462, Affected: 100
  Info: Job[1005:RESTRICT, tasks:193, time:4,155, size: 694,402, [0|193|4,155] ]
Job[1007:DIMENSION_JOIN, tasks:193, time:4, size: 694,402, [0|193|4] ]
Job[1009:DIMENSION_JOIN, tasks:193, time:3, size: 694,402, [0|193|3] ]
Job[1011:DIMENSION_JOIN, tasks:193, time:3, size: 694,402, [0|193|3] ]
Job[1013:DIMENSION_JOIN, tasks:193, time:3, size: 694,402, [0|193|2] ]
Job[1016:OUTPUT_HASH_GROUP_BY, tasks:6, time:272, size: 600, [0|6|271] ]
Job[1018:MERGE_HASH_GROUPBY_PARTITION, tasks:7, time:2, size: 100, [0|7|2] ]
Job[1020:PROJECT, tasks:7, time:0, size: 100, [0|7|0] ]
Job[1022:PROJECT, tasks:7, time:1, size: 100, [0|7|0] ]
Job[1024:RANGE_SORT, tasks:7, time:2, size: 100, [0|7|1] ]
Job[1028:PARTITION_TABLET, tasks:7, time:0, size: 100, [0|7|0] ]
Job[1030:MERGE_ORDERBY_RANGES, tasks:1, time:0, size: 10, [0|1|0] ]
total time: 4,460, record count: 100
restrict的时间/整体时间: 4155/4460
result:
D_YEAR     |S_NATION      |P_CATEGORY |PROFIT      |
- - - - - - - - - - - - - - - - - - - - - - - - - -
1997       |ARGENTINA     |MFGR#11    |31668757023 |
1997       |ARGENTINA     |MFGR#12    |31315629143 |
1997       |ARGENTINA     |MFGR#13    |31899989093 |


我这边主要做了一下内容：

  1. 将数据库执行结果通过Linux脚本下载下来

cat ../conf/ssb_test.sql | ./cplus.sh > ssb30_record_result.txt
  1. 将数据库日志中的 job 信息通过Linux脚本下载下来

./cplus.sh <<EOF
desc history;
EOF
  1. 编写python脚本提取 RESTRICT 时间，并组合前两步骤的结果
a = """
 {具体的 job 信息}
"""
b = """
{具体的 查询结果 信息}
"""
if __name__ == '__main__':
    result_a = []
    # print(a.split("\n\n"))
    for item in a.split("\n\n"):
        result_item = '\nrestrict的时间/整体时间: '
        flag = False
        for item_item in item.split("\n"):
            if "RESTRICT" in item_item:
                flag = True
                result_item += item_item.split("time:")[1].split(", ")[0] + "/"
            elif "total time" in item_item:
                result_item += item_item.split("total time: ")[1].split(", ")[0] + "\n"
        if not flag:
            continue
        result_item = result_item.replace(",", "")
        result_a.append(item + result_item)
    result_a.reverse()
    # for item in result_a:
    #     print(item)
    b_1 = ["\n".join(item.split("\n")[1:6]) for item in b.split("\nSQL")]
    result_b = []
    for item in b_1:
        if "|" in item:
            if "\nSelects" in item:
                result_b.append("result: \n" + "\n".join(item.split("\n")[0:-2]) + "\n\n")
            else:
                result_b.append("result: \n" + item + "\n")
    if len(result_a) != 26 or len(result_b) != 26:
        raise Exception("result'len should =26 and result_a'len should =26")
    result = ["\n".join(item) for item in list(zip(result_a, result_b))]
    result.insert(0, "d")
    result = result[0:-1:2][1:]
    print("\n".join(result))


二、解决线程 bug


我们知道，在 windows 中，通过任务管理器可以看CPU信息，比如下面是我在windows上的CPU信息截图：

334.webp.jpg


现在问题来了，我的数据库系统到底是启用 8 线程好呢？还是 16线程好呢？

答案是 16 线程。以逻辑处理器为准！

数据库原先启用的是8线程，这就是问题所在。这块优化完毕之后，“Star Schema Benchmark 标准测试集优化”基本已终结。


三、测试结果


333.webp.jpg


左边是咱们数据库，右边是 Starrocks 数据库


文章标签：
Python
测试技术
SQL
Linux
Windows
算法
数据库
关键词：
测试优化
数据库优化
数据库测试
benchmark测试
数据库schema
七镜
目录
相关文章
萝卜带泥
|
15天前
|
人工智能 前端开发 测试技术
探索软件测试中的自动化框架选择与优化策略####
本文深入剖析了当前主流的自动化测试框架，通过对比分析各自的优势、局限性及适用场景，为读者提供了一套系统性的选择与优化指南。文章首先概述了自动化测试的重要性及其在软件开发生命周期中的位置，接着逐一探讨了Selenium、Appium、Cypress等热门框架的特点，并通过实际案例展示了如何根据项目需求灵活选用与配置框架，以提升测试效率和质量。最后，文章还分享了若干最佳实践和未来趋势预测，旨在帮助测试工程师更好地应对复杂多变的测试环境。 ####
萝卜带泥
40 4
游客762btuqu5wybw666
|
21天前
|
机器学习/深度学习 前端开发 测试技术
探索软件测试中的自动化测试框架选择与优化策略####
本文深入探讨了在当前软件开发生命周期中，自动化测试框架的选择对于提升测试效率、保障产品质量的重要性。通过分析市场上主流的自动化测试工具，如Selenium、Appium、Jest等，结合具体项目需求，提出了一套系统化的选型与优化策略。文章首先概述了自动化测试的基本原理及其在现代软件开发中的角色变迁，随后详细对比了各主流框架的功能特点、适用场景及优缺点，最后基于实际案例，阐述了如何根据项目特性量身定制自动化测试解决方案，并给出了持续集成/持续部署（CI/CD）环境下的最佳实践建议。 --- ####
游客762btuqu5wybw666
51 4
丰宝宝
|
8天前
|
人工智能 监控 测试技术
探索软件测试中的自动化框架选择与优化策略####
【10月更文挑战第21天】 本文深入剖析了软件测试领域面临的挑战，聚焦于自动化测试框架的选择与优化这一核心议题。不同于传统摘要的概述方式，本文将以一个虚拟案例“X项目”为线索，通过该项目从手动测试困境到自动化转型的成功历程，生动展现如何根据项目特性精准匹配自动化工具（如Selenium、Appium等），并结合CI/CD流程进行深度集成与持续优化，最终实现测试效率与质量的双重飞跃。读者将跟随“X项目”团队的视角，直观感受自动化框架选型的策略性思考及实践中的优化技巧，获得可借鉴的实战经验。 ####
丰宝宝
24 0
游客nmbcre4jd7shs
|
6天前
|
机器学习/深度学习 人工智能 Java
探索软件测试中的自动化框架选择与优化策略####
本文深入探讨了在软件测试领域，面对众多自动化测试框架时，如何根据项目特性、团队技能及长远规划做出最佳选择，并进一步阐述了优化这些框架以提升测试效率与质量的策略。通过对比分析主流自动化测试框架的优劣，结合具体案例，本文旨在为测试团队提供一套实用的框架选型与优化指南。 ####
游客nmbcre4jd7shs
17 1
游客sjjs6jgz4gp7m1
|
19天前
|
存储 NoSQL 分布式数据库
微服务架构下的数据库设计与优化策略####
本文深入探讨了在微服务架构下，如何进行高效的数据库设计与优化，以确保系统的可扩展性、低延迟与高并发处理能力。不同于传统单一数据库模式，微服务架构要求更细粒度的服务划分，这对数据库设计提出了新的挑战。本文将从数据库分片、复制、事务管理及性能调优等方面阐述最佳实践，旨在为开发者提供一套系统性的解决方案框架。 ####
游客sjjs6jgz4gp7m1
41 8
mrq4nk6ni2neg
|
20天前
|
存储 SQL 数据库
深入浅出后端开发之数据库优化实战
【10月更文挑战第35天】在软件开发的世界里，数据库性能直接关系到应用的响应速度和用户体验。本文将带你了解如何通过合理的索引设计、查询优化以及恰当的数据存储策略来提升数据库性能。我们将一起探索这些技巧背后的原理，并通过实际案例感受优化带来的显著效果。
mrq4nk6ni2neg
36 4
软件求生
|
21天前
|
缓存 监控 测试技术
全网最全压测指南！教你如何测试和优化系统极限性能
大家好，我是小米。本文将介绍如何在实际项目中进行性能压测和优化，包括单台服务器和集群压测、使用JMeter、监控CPU和内存使用率、优化Tomcat和数据库配置等方面的内容，帮助你在高并发场景下提升系统性能。希望这些实战经验能助你一臂之力！
软件求生
37 3
爱专研的技术土狗
|
22天前
|
SQL druid 数据库
如何进行数据库连接池的参数优化？
数据库连接池参数优化包括：1) 确定合适的初始连接数，考虑数据库规模和应用需求；2) 调整最大连接数，依据并发量和资源状况；3) 设置最小空闲连接数，平衡资源利用和响应速度；4) 优化连接超时时间，确保系统响应和资源利用合理；5) 配置连接有效性检测，定期检查连接状态；6) 调整空闲连接回收时间，适应访问模式并配合数据库超时设置。
爱专研的技术土狗
35 1
sunrr
|
26天前
|
SQL 缓存 监控
数据库优化
【10月更文挑战第29天】数据库优化
sunrr
31 1
mao29955
|
26天前
|
缓存 关系型数据库 MySQL
如何优化 MySQL 数据库的性能？
【10月更文挑战第28天】
mao29955
52 1

热门文章

最新文章

  • 1
    阿里文娱测试实战：机器学习+基于热度链路推荐的引流，让对比测试更精准
  • 2
    .Net单元测试业务实践
  • 3
    测试自动化和准时交付直接的关联
  • 4
    cartographer环境建立以及建图测试（详细级）
  • 5
    float position的測试案例
  • 6
    Spring框架下的单元测试方法
  • 7
    关于友好性测试中鼠标样式的修改
  • 8
    性能测试工具集锦(上)
  • 9
    Nightwatch.js – 轻松实现浏览器的自动测试
  • 10
    透析QTP自动化测试框架SAFFRON
  • 1
    新一代数据库技术：融合人工智能与分布式系统的未来前景
    346
  • 2
    SQLite数据库介绍以及安装
    96
  • 3
    sprinboot+vue集成neo4j图数据库
    164
  • 4
    neo4j图数据库下载安装配置
    142
  • 5
    几种经典的支持数据库的表单设计器参考
    35
  • 6
    springboot数据库回滚失败原因
    53
  • 7
    spark用于分析数据并将数据保存到数据库中是
    72
  • 8
    生成数据库结构文档
    38
  • 9
    【服务器】python通过JDBC连接到位于Linux远程服务器上的Oracle数据库
    90
  • 10
    数据库SQL语言实战（六）
    49

    • 相关课程

    更多
  • 数据库的前世今生
  • 数据库核心概念
  • 从传统数据库到云数据库演进
  • 数据库常见问题排查
  • 数据库及SQL/MySQL基础
  • 高校精品课-西安交通大学 -数据库理论与技术

    • 相关电子书

    更多
  • DTCC 2022大会集锦《云原生一站式数据库技术与实践》
  • 阿里云瑶池数据库精要2022版
  • 2022 DTCC-阿里云一站式数据库上云最佳实践

    • 相关实验场景

    更多
  • 使用DAS实现数据库SQL优化
  • 自建数据库迁移到云数据库
  • PolarDB NL2SQL：自然语言驱动的数据智能
  • MySQL引擎及架构优化
  • PolarDB for AI：在数据库中通过SQL实现AI能力
  • 云原生HTAP数据库，让你的交易和分析一库搞定
    • 下一篇
    阿里云无影云电脑免费试用，最长可试用3个月