我有一个包含 200gb 记录（22 亿数据）和强大硬件（16 cpu 32 gb ram）Oracle 的庞大数据库。我需要在这些维度上查询这些庞大的数据。每小时有 200.000 条新记录添加到我的事实表中
时间-->年、月、日、小时
业务1--> 4级
业务2--> 3级
用户--> 100万
business1 和 business2 在同一个最低级别匹配，称为 ad_variation
1 service_record 表（20 亿数据）
service_record 包括时间戳、订阅者 ID、ad_variation_id 中的时间
为了查询这些庞大的系统，我使用了 62 个物化视图，但系统有时会出现诸如12008: ORA-12008: error in materialized view refresh path.
我的系统
service_record（20 亿数据）
|
my_cube ( service_record group by time , ad_variation_id,subscriber ) (5亿数据)
|
business1_hour 所有级别（my_cube group by trunc(time,hour) , business1_levels ,subscriber）（2 亿 - 6 亿数据）
|
business2_hour_all level(my_cube group by trunc(time,hour) , business2_levels ,subscriber) (200million - 600 百万数据)
|
business2_day 所有级别（business1_hour group by trunc（time,day），business1_levels，subscriber）（3000万-5000万数据）
|
business2_day_all level(business2_hour group by trunc(time,day), business2_levels,subscriber) (3000万-5000万数据)
|
time_levels
有谁可以向我提供如何改进或解决刷新问题？
甲骨文版本：10g
服务记录表：
创建表 SERVICE_RECORD 并行 16 存储（初始 1M 下一个 10M）按范围分区（服务时间）（每日分区）作为 select s.id, s.servicetime, s.advariationId, s.blindId, s.action from tap_prod.service_Record s where s .servicetime > to_date('01/01/2012','dd/MM/yyyy');
我的主要数据组：
CREATE materialized VIEW tap_cube
parallel 16 storage (initial 1M next 10M)
partition by range( service_time )
(daily partition )
按需快速构建即时刷新
AS
SELECT TRUNC (sr.servicetime,'HH') as service_time,
sr.advariationid AS ad_variation_id,
sr.blindid ASblind_id,
COUNT (sr.blindid) AS total_impr,
SUM (sr.action) AS total_action , count(sr.action) as col1 ,
count(*) as col2
FROM service_record sr
group by TRUNC (sr.servicetime,'HH'), sr.advariationid, sr.blindid;
我的business1逻辑表：
业务一级：
创建物化视图 cube_ty_1_adv_1_time_4
并行 16 存储（初始 10M 下一个 100M）
按范围分区（service_time）
（daily_partition
）
按需快速构建即时刷新
SELECT
CAST
(mycube.service_time AS TIMESTAMP) AS service_time,
ADV.broker_id as entity_id,
blind_id,
SUM (total_impr) as total_impr,count(total_impr) col3 ,
SUM (total_action) as total_action , count(total_action) as col1 , count(*) as col2
FROM tap_cube mycube, dim_advertisement adv
WHERE mycube.ad_variation_id = adv.ad_variation_id
group by CAST ( mycube.service_time AS TIMESTAMP)，ADV.broker_id，blind_id；
业务1维级别：
CREATE table
dim_advertisement
AS
SELECT bro.id AS broker_id,
adver.id AS advertiser_id,
camp.id AS campaign_id,
adv.id AS Advertisement_id,
ad.id AS ad_variation_id
FROM
tap_prod.advertisement adv,
tap_prod.ad_variation ad,
tap_prod.campaign camp,
tap_prod.advertiser adver,
tap_prod.broker bro
WHERE
bro.id = adver.brokerid
and camp.advertiserid = adver.id
AND camp.id = adv.campaignid
and ad.advertisementid = adv.id;