Oracle优化:千万级大表逻辑判断的累赘-成都创新互联网站建设

关于创新互联

多方位宣传企业产品与服务 突出企业形象

公司简介 公司的服务 荣誉资质 新闻动态 联系我们

Oracle优化:千万级大表逻辑判断的累赘

insertinto pntmall_point_detail(PNTMALL_PNT_ID,PNTMALL_PNT_DT,PNTMALL_VALIDITY,PNTMALL_LASTUPDATEDT,

在达州等地区,都构建了全面的区域性战略布局,加强发展的系统性、市场前瞻性、产品创新能力,以专注、极致的服务理念,为客户提供成都网站设计、成都网站建设、外贸网站建设 网站设计制作按需网站策划,公司网站建设,企业网站建设,品牌网站制作,营销型网站,外贸网站制作,达州网站建设费用合理。

                           PNTMALL_RPTPNT_ID,PNTMALL_HAIERUID,

                   PNTMALL_OPTYPE_ID,PNTMALL_OPTYPE_DESC,

                         PNTMALL_NUM,PNTMALL_EXPDT,HPT_REDEMPT_POINT,

                         HPT_LEFT_POINT,HPT_FULLREDEMPT_STATUS)

  SELECT PNTMALL_PNT_ID,PNTMALL_PNT_DT,PNTMALL_VALIDITY,PNTMALL_LASTUPDATEDT,

                           PNTMALL_RPTPNT_ID,PNTMALL_HAIERUID,

                   PNTMALL_OPTYPE_ID,PNTMALL_OPTYPE_DESC,

                         PNTMALL_NUM,PNTMALL_EXPDT,0 as HPT_REDEMPT_POINT,PNTMALL_NUM,0

    FROM pntmall_point_detail_tmp a

          WHERE NOT EXISTS (SELECT b.PNTMALL_PNT_ID
            FROM pntmall_point_detail b
           WHERE a.PNTMALL_PNT_ID = b.PNTMALL_PNT_ID);

PNTMALL_POINT_DETAIL包含3800万条数据,cost 6 hours。

优化后

 

delete from pntmall_point_detail_tmp a where exists (select 1 from pntmall_point_detail b where a.PNTMALL_PNT_ID = b.PNTMALL_PNT_ID);

insertinto pntmall_point_detail(PNTMALL_PNT_ID,PNTMALL_PNT_DT,PNTMALL_VALIDITY,PNTMALL_LASTUPDATEDT,

                           PNTMALL_RPTPNT_ID,PNTMALL_HAIERUID,

                   PNTMALL_OPTYPE_ID,PNTMALL_OPTYPE_DESC,

                         PNTMALL_NUM,PNTMALL_EXPDT,HPT_REDEMPT_POINT,

                         HPT_LEFT_POINT,HPT_FULLREDEMPT_STATUS)

  SELECT PNTMALL_PNT_ID,PNTMALL_PNT_DT,PNTMALL_VALIDITY,PNTMALL_LASTUPDATEDT,

                           PNTMALL_RPTPNT_ID,PNTMALL_HAIERUID,

                   PNTMALL_OPTYPE_ID,PNTMALL_OPTYPE_DESC,

                         PNTMALL_NUM,PNTMALL_EXPDT,0 as HPT_REDEMPT_POINT,PNTMALL_NUM,0

    FROM pntmall_point_detail_tmp a;

cost 5 minutes。

其实还可以进一步优化

drop indexBER.INDEX_POD_PNTMALL_HAIERUID;

drop indexBER.PNTMALL_POINT_ID_HAIERUID;

delete frompntmall_point_detail_tmp a where exists (select 1frompntmall_point_detailb wherea.PNTMALL_PNT_ID =b.PNTMALL_PNT_ID);

insert/*+append*/into pntmall_point_detail(PNTMALL_PNT_ID,PNTMALL_PNT_DT,PNTMALL_VALIDITY,PNTMALL_LASTUPDATEDT,

                           PNTMALL_RPTPNT_ID,PNTMALL_HAIERUID,

                   PNTMALL_OPTYPE_ID,PNTMALL_OPTYPE_DESC,

                         PNTMALL_NUM,PNTMALL_EXPDT,HPT_REDEMPT_POINT,

                         HPT_LEFT_POINT,HPT_FULLREDEMPT_STATUS)

  SELECT PNTMALL_PNT_ID,PNTMALL_PNT_DT,PNTMALL_VALIDITY,PNTMALL_LASTUPDATEDT,

                           PNTMALL_RPTPNT_ID,PNTMALL_HAIERUID,

                   PNTMALL_OPTYPE_ID,PNTMALL_OPTYPE_DESC,

                         PNTMALL_NUM,PNTMALL_EXPDT,0 as HPT_REDEMPT_POINT,PNTMALL_NUM,0

    FROM pntmall_point_detail_tmp a; 

commit;

WHERE NOT EXISTS (SELECT b.PNTMALL_PNT_ID

            FROMpntmall_point_detail b

           WHEREa.PNTMALL_PNT_ID = b.PNTMALL_PNT_ID);

create index BER.INDEX_POD_PNTMALL_HAIERUIDon BER.PNTMALL_POINT_DETAIL (PNTMALL_HAIERUID) online nologing;

create unique indexBER.PNTMALL_POINT_ID_HAIERUID on BER.PNTMALL_POINT_DETAIL (PNTMALL_PNT_ID) online nologing;

总体优化思路,不要在insert中加入过多的判断语句,删索引,append,重建索引,如果是归档模式,alter table nologing;append 只适用于insert select 这种方式,而且insert后要加commit,否则无法进行其他DML操作。

实测 append 1600万条数据,cost 8s


标题名称:Oracle优化:千万级大表逻辑判断的累赘
网站网址:http://kswsj.cn/article/jggpso.html

其他资讯