Wednesday, February 3, 2016

2057595 - FAQ: SAP HANA High Availability

Symptom
FAQ: SAP HANA High Availability


Environment
SAP HANA


Cause
You are interested in further details related to SAP HANA High Availability.


Resolution

1. What does SAP HANA Scale-Out mean?

In a SAP HANA Scale-Out system, multiple hardware servers are combined into one logical system, mainly to increase the total amount of memory. Thus scale-out refers to increase the available memory by adding additional hosts (in opposite to scale-up, where the memory within one host is increased).

2. What does SAP HANA High Availability mean?

High Availability increases the failure tolerance within one data center by providing a fast switch over to an operational state of the SAP HANA database.

3. What does SAP HANA Disaster Tolerance mean?

Disaster Tolerance refers to extended failover capabilities across datacenters.
This is typically achieved by replicating the data of the main SAP HANA system to another SAP HANA system in another datacenter at a different geographical location.

4. Which solutions for High Availability does SAP HANA offer?

High Availability within one Data Center:
  • Backup & Recovery
  • SAP HANA Host Auto-Failover  
    To make a SAP HANA database highly available, the Scale-Out system is extended by additional stand-by nodes, which can take over in case one or more regular nodes should become unavailable. If one host fails the standby host automatically takes over by gaining access to the data and log volumes of the failed host.
  • SAP HANA System Replication
    Within SAP HANA replicates the data and the redo log of every committed transaction to a secondary SAP HANA database with the same SID and instance number. In this setup the two databases (primary and secondary) are located close to each other.
  • SAP HANA Storage Replication
    On storage level replicates the data and the redo log of every committed transaction to a secondary site.
Disaster Tolerance between data centers:
  • SAP HANA System Replication
    Within SAP HANA replicates the data and the redo log of every committed transaction to a secondary SAP HANA database with the same SID and instance number.
  • SAP HANA Storage Replication
    On storage level replicates the data and the redo log of every committed transaction to a secondary site.

5. Where do I find more information about SAP HANA High Availability?

More information can be found in the following documentation:

6. Host Auto-Failover: Do SAP HANA application server user sessions stay in reconnect status?

A switch to the standby currently will lead to an abort of the transaction, if the SAP Application Server transaction is connecting to the database. If during the failover a transaction is accessing the database, NetWeaver notices the lost connection and performs a rollback of the ABAP transaction. If, however, the ABAP transaction is not active or does not access the database for the time of the failover, nothing happens. For more details on “Configuring Application Servers for Failover” please refer to the SAP HANA Admin Guide: http://help.sap.com/hana/SAP_HANA_Administration_Guide_en.pdf

7. Host Auto-Failover: What happens to active BW process chains (i.e. during data load) when a failover happens ?

Write transactions get canceled in case of a failover. Thus the transaction will be canceled and rolled back by the standby indexserver which will take over.
For BW process chains this means that this process step will be canceled and you will have to restart at this failed step and continue with the loading chain where it was interrupted.
If, however, the process chain was interrupted while there was no open write transaction, the loading will continue without problems.

8. Host Auto-Failover: Do you need to manually start an ex-failed node?

If the ex-failed node is operational again and restarted, it automatically is taken as new standby node in the Host Auto-Failover landscape. Thus, when a failover was performed and the failed host is available again, no automatic fallback will happen. A controlled failback can be performed by stopping or restarting the standby host that is currently in use. Automatic failback will only happen when the complete landscape is restarted, also see Host Start Order.


Keywords
SAP HANA High Availability
Scale-Out
System Replication   
Host Auto-Failover
Disaster Tolerance
Disaster Recovery
FAQ



Header Data

Released On 16.06.2015 12:17:15
Release Status Released to Customer
Component HAN-DB SAP HANA Database
Other Components
HAN-DB-HA SAP HANA High Availability
Priority Normal
Category How To
Posted by at

No comments:

Post a Comment

Subscribe to: Post Comments (Atom)