积极答复者
SQLSERVER的伸缩性以及高可用性

问题
-
如果数据量非常大,但是又非常重要,必须要求高度的事务完整性,当机器硬件已经达到极限的情况下,那么,我觉得要处理的话,必须利用分布式方案了,如:分布式分区视图,可以横向扩展,添加更多的机器进来,必然在伸缩性上能够解决这个问题,但是微软官方上说“分布式事务以及跨数据库的事务”是不能进行镜像或者Always on的,那么,如何解决呢?
对等复制不能达到高度的事务完整性要求,所以这个方案肯定是不行的!再顺便说说ORACLE的RAC,RAC是多台机器共用一套存储设备,加入就把它叫做硬盘,对于海量数据的时候,瓶颈都在硬盘上,多套CPU去操作同一个硬盘,这RAC的性能能够有很好的提升吗?
答案
-
If you need query multiple dbs on same server, always on is the way to go for DR because sql will failover group of dbs together.
- 已建议为答案 Amy PengMicrosoft employee, Moderator 2012年11月22日 5:30
- 已标记为答案 Amy PengMicrosoft employee, Moderator 2012年11月30日 1:36
-
或许 Windows Azure SQL Database Federation 能满足跟你scale out 跟 HA 的双重需求。
scale out :它的partition能提供。
HA:这个是Windows Azure提供的基本功能。
http://msdn.microsoft.com/en-us/library/windowsazure/hh597452.aspx
Please click the Mark as Answer button if a post solves your problem!- 已编辑 Michael CS 2012年11月21日 8:43
- 已建议为答案 Amy PengMicrosoft employee, Moderator 2012年11月22日 5:30
- 已标记为答案 Amy PengMicrosoft employee, Moderator 2012年11月30日 1:36
全部回复
-
If you need query multiple dbs on same server, always on is the way to go for DR because sql will failover group of dbs together.
- 已建议为答案 Amy PengMicrosoft employee, Moderator 2012年11月22日 5:30
- 已标记为答案 Amy PengMicrosoft employee, Moderator 2012年11月30日 1:36
-
或许 Windows Azure SQL Database Federation 能满足跟你scale out 跟 HA 的双重需求。
scale out :它的partition能提供。
HA:这个是Windows Azure提供的基本功能。
http://msdn.microsoft.com/en-us/library/windowsazure/hh597452.aspx
Please click the Mark as Answer button if a post solves your problem!- 已编辑 Michael CS 2012年11月21日 8:43
- 已建议为答案 Amy PengMicrosoft employee, Moderator 2012年11月22日 5:30
- 已标记为答案 Amy PengMicrosoft employee, Moderator 2012年11月30日 1:36