Skip to Content

In this blog series, I will share the database perspective to achieve Business Continuity with SAP SYBASE solutions. Business continuity is really about a) Ensuring High Availability of business and operational systems and b) Disaster Recovery planning. High Availability deals with avoiding disruption of business activity and Disaster Recovery is aimed at restoring operations after an interruption. I will focus on High Availability architectures and solutions for prominent SAP database platforms including SAP Sybase ASE, Replication Server, SAP Sybase IQ and SAP HANA.

High Availability has become a must-have requirement for today’s business systems. During my consulting career, I have developed High Availability solutions for a number of customers in a variety of industries such as media and entertainment, newspaper publishing and printing, online reservation systems, financial and trading applications, industrial process management systems, and highway traffic warning systems. Even though the industry segments are different, the common theme is that the customers all run mission-critical applications that:

– rely on critical data and information systems

– must be up and running during specific timeframe in the day or more and more and more often 24/7.

If their applications fail, they are unable to broadcast radio or TV programs, produce/sell newspapers, book reservations, enter deals, run their plants etc. In the end they will lose business, creditbility, clients, and money. Unavailability of a highway traffic warning sign application can have very serious consequences.

As SLA requirements are specific to applications, activities and customer context, there is no one High Availability solution that fits all cases. That’s why, SAP has developed a set of business continuity solutions.  The range of solutions is depicted in an “Availability” pyramid as shown below. Each level of the pyramid represents an availability requirement. The highest level of the pyramid provides the highest level of availability and ensures minimal downtime.

HA_pyramid.png

Each level is associated to a set of SAP preferred solutions that covers one or more failure types (server failure, data failure, site location failure) and has its own characteristics in term of downtime, recovery automation and application transparency. I will detail each level from the bottom to the top.

1.    Server Recovery

This is the first level of security. There are several possible cold standby setups:  hot spare, synchronous transfer of transaction data (LVM or storage replication) and log shipping.  For instance, in log shipping method data is backuped on the production site, transferred and restored on a standby site.

➕ easy to implement

➖ failure impact client applications

2.    Server Availability

Server availability means that this level addresses the failure of the database host. This level requires extra investment for clustering facilities (redundant hardware and cluster software). SAP products sit on top of O/S cluster and are considered as application resources or database services. Database services are driven by cluster layer, they are started/stopped/monitored and ultimately failed over in case of failure. SAP products are compatible with common proprietary cluster solutions like IBM PowerHA (formely known as HACMP), HP Serviceguard, ORACLE SunCluster and generic solution Symantec VCS.

centralized management of all resources (HW and software) on O/S cluster level

➖ requires O/S Cluster skills

➖ failure impact client applications

downtime due to databases recovery (including tempdbs)

3.    Site Availability

This level addresses a failure of a computer room, or an entire site location. Two types of business continuity solutions are available for site availability : SAP SYBASE ASE with HA option and SAP SYBASE ASE Cluster Edition. These solutions also cover server availability.

SAP SYBASE ASE with HA option.

SAP SYBASE ASE with HA option (also known as HA companion) relies on a O/S cluster layer. This solution clusters two dataservers together (also known as companion dataservers). At a given time a database (or a set of databases) is tied to a dataserver. In case of failure, database resources are exported, imported and restarted on the surviving host/dataserver. A database recovery takes place.

in case of failure companion dataserver is already up and running to accept failedover databases

➖ requires O/S Cluster skills

limited to 2 dataservers

downtime due to databases recovery (not tempdbs)

SAP SYBASE ASE Cluster Edition.

SAP SYBASE ASE Cluster Edition is a shared disk cluster solution. It does not require any additional O/S cluster software (in contrast to the SAP SYBASE ASE with HA option). In this solution a set of dataserver instances is gathered into a logical cluster. Clients connect to the logical cluster (not to a specific database instance). In case of a failure of an instance of the logical cluster, its clients automatically fail over the other instances without disruption. No database recovery takes place and downtime is minimal for this solution.

➕ minimal downtime

automated failover of clients

4.    Multi site High Availability with Asynchronous Replication.

       Multi site High Availability with Asynchronous Replication ensures that databases are synchronized on the network via SAP SYBASE Replication Server. Data failure is handled by data redundancy.  This solution is based upon Change-Data-Capture technology from SAP SYBASE (SAP SYBASE Replication Server). Transactions from source dataserver are forwarded to another dataserver hosted on a remote location providing high level of data security and flexibility.

       ➕ site redundancy handles server and site failure

       ➕ data redundancy handles data failure

5.    Continuous Multi site High Availability

This business continuity solution is a comprehensive solution and offers a maximum of availability. It combines SAP SYBASE ASE Cluster Edition and Replication Server. Data failure is handled by data redundancy and server failure is handled by SAP SYBASE ASE Cluster Edition. 

➕ site redundancy handles server and site failures

➕ data redundancy handles data failure

➕ minimal downtime

In my next post, I will elaborate on the above business continuity solutions.

Stay in the conversation by following SAP Services on SCN

Follow along throughout the event on Twitter at  @SAPServices

To report this post you need to login first.

1 Comment

You must be Logged on to comment or reply to a post.

Leave a Reply