One of the advantages of the HANA database is that capabilities like high availability automatically become available to the applications that run on top of the database – dependent of course on the underlying hardware.
So that means that SAP Business One version for HANA when paired with the right certified hardware automatically supports High Availability.
***Please note that the same also applies to SAP Business One Analytics powered by SAP HANA
So what exactly is High Availability for HANA about?
Well, a good place to start is with a definition of High Availability which I have taken from a White Paper on the subject written by Chaim Bendelac from the SAP HANA Development Team which you can find in the SAP Business One Technical Toolkit and you can also find it here – http://www.saphana.com/docs/DOC-2775 .
High Availability is a set of techniques, engineering practices and design principles for Business Continuity. This is achieved by eliminating single points of failure (fault tolerance), and providing the ability to rapidly resume operations after a system outage with minimal business loss (fault resilience).
So it is fundamentally about business continuity….a discussion around load balancing and performance optimization requires more detail so I am not going to cover it in this blog.
The area that I want to touch on here and is hardware redundancy as this seems to be the main area of concern for customers and partners that I talk with about the subject of up-time management and fault tolerance.
Of course, hardware redundancy is a well established subject and the basics include the provision of UPS based power protection, hot swappable and redundant power supply units, hot swappable network cards, RAID disk arrays and even hot swappable memory units.
So to ensure that you are maximizing your fault tolerance in a single server environment you need to consider each of these components when specifying your hardware…of course all of the HANA certified servers provision one or more of these options….you can also find details of these in the SAP Business One Technical Toolkit and in the Product Availability Matrix document for SAP Business One version for SAP HANA also in the toolkit.
You’ll need to ensure that at a bare minimum you have a RAID array, UPS power protection and redundant power supplies in your server to provide a minimal level of protection against failure and I would recommend RAID 10 (striping and mirroring) for your data and log storage and RAID 1 (mirroring) for your O/S.
Of course the next and more expensive option is to go with complete server redundancy and deploy multiple servers in a mirrored configuration where all data is automatically written to both HANA environments and the underlying hardware/software manages the fault detection and recovery.
HANA handles this scenario and manages the process automatically once configured correctly…the good news is that when your servers are configured in a live replication mode, your single HANA server license applies to both machines so you only need to license one instance to run two instances in a fault tolerant mode.
Here’s a graphic that summarizes the options quite well.
You’ll need to be on HANA SPS04 or higher to support these scenarios….but this should be your situation anyway as we only support SPS04 or higher with SAP Business One version for HANA.
Again, I refer you to the excellent introductory white paper written by Chaim Bendelec and I would like to express my thanks to him for providing a great intro to the topic and for allowing me to reproduce the graphic and description of Fault Tolerance from his white paper in this blog.