Transcript Document

What HADR Option(s) Are
Right For You?
Where’s The AlwaysOn?
Levels of Protection
Instance
Database
Object
Availability Groups



FCI



Log Shipping



Mirroring



Replication (Merge)



Replication (P2P)



Options Per Version
2000
2005
2008/R2
2012
2014
Availability Groups





FCI





Log Shipping





Mirroring





Replication (Merge)





Replication (P2P)





SQL 2000 Options By Edition
Workgroup
Standard
Enterprise
FCI*



Log Shipping



Replication (Merge)



* 2 nodes only in Standard
SQL 2005 Options By Edition
Workgroup
Standard
Enterprise
FCI*



Log Shipping



Mirroring**



Replication (Merge)



Replication (P2P)



* 2 nodes only in Standard
** High safety (synchronous) only in Standard
SQL 2008 / 2008 R2 Options By Edition
Web
Standard
Enterprise
Datacenter
FCI*




Log Shipping




Mirroring**




Replication (Merge)***




Replication (P2P)




* 2 nodes only in Standard
** High safety (synchronous) only in Standard
*** Web only supported as subscriber
SQL 2012 / 2014 Options By Edition
Web
Standard
BI
Enterprise
























Availability Groups****
FCI*
Log Shipping
Mirroring**
Replication (Merge)***
Replication (P2P)
• * 2 nodes only in Standard/BI
** High safety (synchronous) only in Standard/BI
*** Web only supported as subscriber
• **** Up to 4 secondary replicas in 2012. 8 secondary replicas in 2014
Failover Cluster Instances (FCI)
 You might also might know it as Clustering
 Common Terms:
 Active/Active or Active/Passive (not correct use N or N+1)
 MSCS (Microsoft Clustering Services) – Windows NT4.0 to 2003
 WSFC (Windows Server Failover Cluster) – Windows 2008 and up
 Full copy of SQL installed on all servers (nodes)
 Transparent client redirection on failover*
 Jobs, logins, linked servers, etc… also failover
 Some special hardware requirements
 Shared storage or 3rd party hardware solution
 There are single points of failure
Log Shipping
 Not HA technology, but critical for business continuity
 Scheduled backups of transaction logs get moved and restored to
other servers
 Can delay applying of logs on other servers
 Perfect for those “OH @#$@()” moments
 Possible to use built in routines, or roll your own
 Great for remote DR
 Limited to databases only.
 Logins, jobs, connection strings, etc.. Have to be managed
 Client redirection manual or via DNS
Mirroring
 Deprecated, but doesn’t look like it’s going away any time soon
 May not live past SQL 2016 with AG support in Std Edition
 Two different modes
 High Safety (2 phase commit)
 High Performance (Enterprise Edition only)
 Provides automatic failover (with Witness)
 Transparent client redirection (use Failover Partner in conn string)
 Single database in mirror
 Have manage failover for multiple databases
 Have to manually manage logins, jobs, linked servers, etc…
 Can only have a single mirror
 Has no domain requirements
Merge Replication
 Object level, so very restrictive
 Requires conflict resolution which can impact performance
 Can develop custom resolvers
 Needs a GUID on each row, impacts storage
 Does not scale well to high transaction levels
 Not a good choice
Peer to Peer Replication
 Object level, so very restrictive
 Need to very carefully manage ranges at each location for peer writes
to prevent serious database issues
 Using multi-site, multi-write can allow for local scale
 Management can be very complicated
Availability Groups
 Built on WSFC (to manage quorum) so requires AD
 Sends transactions over to secondary replicas
 Synchronous mode (allows for automatic failover)
 Asynchronous mode (great for offsite DR)
 Allows reading from secondary replicas
 Incurs a 16-byte per row overhead on Insert/Update
 4 secondary replicas in 2012, 8 in 2014
 Multiple databases allowed in an AG
 Requires manual management of logins, jobs, linked servers, etc…
Availability Groups Are The New Big Shiny
 They can be a (reasonably) easy way to get HA going without
specialized hardware, but…
 Multiply the storage costs, as you’ll be paying per server
 Reading off a secondary? That’ll cost you
 Multi-subnet configuration can lead to management nightmares
 SQLPS, SSMS do not support the MultiSubnetFailover connection option




Sync commit mode can inhibit performance
No cross database or DTC transactions (coming in SQL 2016!)
Will only work within a single domain and WSFC
Lose the WSFC or AD and lose the AG
 Do NOT have AGs as your sole HADR option!
HADR With AGs and…?
 First option… Log Shipping
 Provides ability to delay logs application to secondary server
 Secondary can reside outside the WSFC
 Second option… FCI
 Provides initial instance level protection
 Cannot automatically failover to another node in AG
 Third option… Log shipping & FCI
 Best of both worlds, but…
 No automatic failover within the AG
Questions?