MAINFRAME > Administrator > IMS

Leap Ahead With IMS

Continue the focus on resiliency with IBM Information Management System

Continue the focus on resiliency with IBM Information Management System
Illustration by Bob Scott

Resiliency, generally speaking, is the capability of something to recover from adversity. In the programming world, this means the capability of the software to recover from unexpected errors or failures, or detect potential errors before they happen in order to correct them autonomically before they surface to the user. In an ideal world, this capability would perform while preventing any noticeable effect to the program's end user. The IBM* Information Management System (IMS*) works to achieve these goals, because any loss of service can mean a significant cost to the customer.

With IMS, there are many different levels of possible unexpected errors, such as application errors, device errors and system failures. Also, every user could have different requirements for the level of resiliency from their system. Higher resiliency requirements require more investment in the infrastructure to support them. A user with a relatively low requirement can be satisfied with the resiliency characteristics associated with a single non-data sharing IMS. While at the other the other end of the spectrum, a user might need to take advantage of all the Parallel Sysplex* technologies - such as shared message queues, Fast Database Recovery (FDBR) and other z/OS* components and features such as Geographically Dispersed Parallel Sysplex* (GDPS*) - to meet their resiliency requirements.

Database Access Resiliency

IMS incorporates many capabilities to reduce the impact of different errors associated with database access. For example, I/O toleration allows the applications to continue even if a database I/O error occurs. Data sharing can improve availability by providing continuous access to your databases in the event of a system or a failure of a sharing partner.

IMS High Availability Large Database (HALDB) support addresses at least two other issues. First, it provides a method to surpass the database size limitations of 4 GB and 8 GB. HALDBs allow users to partition databases into up to 1,001 separate partitions of 4 GB each while allowing the application to treat all partitions as one database. In other words, this prevents potential application error due to a database full condition. Second, partitioning also allows the individual partitions to be treated independently, allowing users to reorganize or recover single partitions while others remain available. So, instead of a failure of any part of the database making it unusable until recovered, it's possible for an error to be encountered on a single partition and the rest of the database's partitions remain available for processing while the other partition is recovered.

System Resiliency

On the system side, IMS protects the system from application errors by isolating the failure to the dependent region and automatically resolving in-flight updates, thus limiting the impact of the failure. IMS also has capabilities that can be implemented to limit the impact of system failures (such as an IMS failure or a z/OS* image failure). For example, IMS implemented Extended Recovery Facility (XRF) to allow a backup system to rapidly take over the work of a failed system. More recently, shared message queues and FDBR provide similar capabilities to limit the impact of a single system failure. For site failures, IMS provides a Remote Site Recovery (RSR) feature that allows a backup site to shadow an active site system enabling a remote system to resume service in less than one hour in the event of a significant active site outage.

Greg Vance is a senior software engineer at IBM's Silicon Valley laboratory. He's the IMS DBRC and RSR development team lead.

comments powered by Disqus



2019 Solutions Edition

A Comprehensive Online Buyer's Guide to Solutions, Services and Education.

Bringing IMS and SOA Together With IMS Connect

New requirements for IMS Connect functionality could make implementing an SOA environment with IMS easier and more flexible.

Celebrating 40 Successful Years

IMS version 10 supports synchronous and asynchronous callout

IBM Systems Magazine Subscribe Box Read Now Link Subscribe Now Link iPad App Google Play Store
Mainframe News Sign Up Today! Past News Letters