in的MOLAP模式下存在如下问题： 历史数据每日刷新，失去了增量的意义。 每日回溯历史数据量大，10亿+的历史数据回溯。 数据计算耗时3小时+，存储1TB+，消耗大量计算存储资源，同时严重影响SLA的稳定性。 预计算的大量历史数据实际使用率低下，实际工作中对历史的回溯80%集中在近1个月左 右，但为了应对所有需求场景，业务要求计算近半年以上的历史。 不支持明细数据的查询。 解决方案：引入MPP引擎，数据现用现算

栈更加专精，⼤⼤降低了数据中台的开发、运维和使用的门槛，⼤幅提 升了数据应用开发的效率； Ø PieCloudDB具备完善的商业化技术服务，提供本地化技术支持，技术响应更加及时，有效提升服务的SLA； Ø PieCloudDB采用简洁的技术架构，没有复杂的组件叠加，让集群运维管理门槛更低，帮助集团企业⼤幅降低后期的 运维管理成本； Ø PieCloudDB是云原⽣的存算分离架构，可以⽆缝

·如果客户的实际应用还存在大量较高并发的小 IO 操作，比如随机 查询，可以考虑 SSD+SAS+ 表空间的方式，并将随机 IO 类应用对 应的表放在 SSD 设备上，从而有效的隔离底层 IO，达到更好 SLA。 某电信用户存在大量对 400 个字段以上大宽表的随机查询场景，之 前由于 IO 资源争用，在批量作业调起时随机查询响应时间显著增加， 通过这种方式改造后，达到了很好的效果。 ·预留一台硬件服务

interrupted during the planned maintenance time, and that the maintenance window can fit into your SLA requirements, there are some settings you must specify when putting a host into maintenance mode:

interrupted during the planned maintenance time, and that the maintenance window can fit into your SLA requirements, there are some settings you must specify when putting a host into maintenance mode:

interrupted during the planned maintenance time, and that the maintenance window can fit into your SLA requirements, there are some settings you must specify when putting a host into maintenance mode:

interrupted during the planned maintenance time, and that the maintenance window can fit into your SLA requirements, there are some settings you must specify when putting a host into maintenance mode:

interrupted during the planned maintenance time, and that the maintenance window can fit into your SLA requirements, there are some settings you must specify when putting a host into maintenance mode:

interrupted during the planned maintenance time, and that the maintenance window can fit into your SLA requirements, there are some settings you must specify when putting a host into maintenance mode: