|
 Optimizing continuity strategies helps maximize service uptime
Today’s enterprise-class IT services and applications span heterogeneous systems and draw resources from multiple sources in the fulfillment of business goals. Given such a sophisticated architecture, maintaining service and application continuity is a challenging proposition. Many points of possible failure translate into the daunting possibility of outages.
When such outages occur, what are the consequences? Among others, they are likely to include diminished end user productivity, reduced service availability to external clients and customers, revenue loss and brand impact—all undesirable to the host organization. Yet, a manual approach to this problem is suboptimal. Manually assigning services to failover systems, recovering data or systems when necessary and restarting applications and services can be time-consuming and operationally expensive, while also introducing the possibility of inadvertent error.
A superior approach would deliver on the promise of automation by tracking failed systems, applications and services; transferring them to failover systems; executing disaster recovery if needed; and thus coordinating the general process of maintaining IT service continuity as guided by industry best practices and a tight integration with various solutions in place throughout the IT infrastructure.
Two new IBM offerings drive service/application continuity through automation
 IBM empowers just such an approach through two new offerings: IBM Tivoli System Automation for Multiplatforms 3.1 and IBM Tivoli System Automation Application Manager 3.1.
Together, these two products help organizations achieve maximum continuity for both traditional IT services and composite business applications, thus delivering an end-to-end solution to this mission-critical business need. They empower organizations to detect failing components proactively (in some cases, before business consequences occur as a result of failure), manage planned outages more optimally, decrease costs, increase efficiency and simplify management via a single point of control.
How do the IBM offerings accomplish these many goals? Consider the case of IBM Tivoli System Automation for Multiplatforms 3.1, which targets traditional IT applications and services (as opposed to composite business applications). Organizations have often deployed these services in the data center by way of a cluster of systems; in the event of a primary system failure, the service can be transferred manually to the designated failover system.
Such an approach, while certainly an improvement on an architecture with no failover, can still involve unwanted complications and inefficiencies. For instance, the process of designating one system active and another inactive, then reversing these designations in the event of system failure, is typically manual and therefore requires extra attention from IT staff, as it increases the downtime of the failed service.
IBM Tivoli System Automation for Multiplatforms 3.1, on the other hand, provides a solution which is both faster and more resilient. As guided by pre-created policies, IBM Tivoli System Automation for Multiplatforms 3.1 responds to system failure by allowing applications and resources to be restarted on any server in a cluster. Furthermore, after the failed system is recovered, it is automatically added back to the pool of available systems, thus helping to maximize the uptime of all applications and services supported by the solution.
How much more efficient is this approach? One way of measuring the improvement, of course, comes from tracking actual recovery times against target times determined by Recovery Point Objectives (RPOs). IBM Tivoli System Automation for Multiplatforms 3.1 helps by allowing administrators to track starting, stopping, restarting and failover times objectively—which is useful for both real-world failure and preliminary trials— to determine target RPOs in the first place.
A second significant benefit of IBM Tivoli System Automation for Multiplatforms 3.1 is that it can reduce the total hardware required for both active and failover service. Consider a conventional clustering solution in which there is a 1:1 relationship between active and failover systems; this means that two systems are required at all times for a given service, one active and one idle. IBM Tivoli System Automation for Multiplatforms 3.1 improves this situation by supporting failover consolidation. That is, multiple active systems providing a service can be failed over to a smaller number, even a single physical host, instead of the same number of hosts. This approach represents a significant optimization of IT hardware, allowing the organization to realize the maximum return on investment and minimize the idle time of deployed systems.
Improved policies through graphic editing and the OPAL archive
One exciting new feature of IBM Tivoli System Automation for Multiplatforms 3.1 is its graphic policy editor. Prior versions of the solution required manual editing of policies via an XML document, but in version 3.1, policies can instead be edited in a graphic interface for enhanced ease of use and policy visibility. For instance, policy elements such as constructs and relationships can now be visualized directly onscreen, helping to ensure that they correspond accurately to operational resources and procedural goals. The new editor is also linked with the Integrated Solutions Console, provides filtering capabilities and includes a policy checker to detect and eliminate semantic errors, ensuring policies are as accurate as possible.
What’s more, it’s also possible to obtain pre-created policies based on industry best practices to optimize recovery times and minimize application and service outages. Such policies can be obtained from the IBM-hosted Open Process Automation Library (OPAL), and they allow IBM Tivoli System Automation for Multiplatforms 3.1 to support both IBM and third-party solutions, including IBM DB2, SAP, IBM Tivoli Monitoring, IBM Tivoli Change and Configuration Management Database, Oracle databases, IBM WebSphere Application Server and others.
Another major feature of IBM Tivoli System Automation for Multiplatforms 3.1 is Linux continuity via IBM System z. Customers with IBM System z mainframes, known for exceptionally high robustness and continuity, can leverage that investment using an optional xDR component. This coordinates with z/OS and IBM Geographically Dispersed Parallel Sysplex (GDPS)—IBM’s enterprise-class disaster recovery solution—to provide business continuity for Linux through disk error detection, coordinated site takeover, sanity checks and other features.
Improved continuity for composite applications
Of course, while conventional IT services and applications represent one continuity challenge, a second comes from composite business applications. Maximizing continuity of these applications, which draw from multiple data resources and utilize logical links between different systems in the infrastructure, can be particularly difficult, and requires a dedicated solution with optimized features.
Just such a dedicated solution is IBM Tivoli System Automation Application Manager 3.1, a new offering created to help organizations optimize composite application continuity from end to end—at every point in the often-complex, multi-tiered chain of logic execution and data sourcing. Entire composite applications can be automatically started, stopped, restarted and failed over when required, improving IT efficiency and minimizing service downtime and operational costs.
How does IBM Tivoli System Automation Application Manager 3.1 accomplish this impressive feat? Part of the answer is found in the way the solution reduces configuration through intelligent analysis. Special adapters are used to collect information from IBM and non-IBM clustering technology, such as IBM System Automation for z/OS, Microsoft Cluster Server, Veritas Cluster Server or HACMP, about composite applications and application status. This means that administrators are only required to enter information about groups, resources and dependencies, and makes it possible to achieve higher continuity through advanced functions, such as shifting applications between clusters on demand, or, in some cases, across platforms.
Centralized management is now enabled via a Web-based single point of control for truly holistic oversight of application continuity, even though applications may be deployed on different operating systems such as Linux, AIX, z/OS, Windows and Solaris. This empowers IT to treat composite applications as a logical whole, starting or stopping them wherever needed regardless of the host platform. Furthermore, through integration with IBM Change and Configuration Management Database, the solution can utilize a discovery library adapter to map an application’s resources and relationships and store it for future reference and exploitation by other solutions in the fulfillment of service management strategies.
Like IBM Tivoli System Automation for Multiplatforms 3.1, IBM Tivoli System Automation Application Manager 3.1 includes a new GUI-based policy editor and tracks recovery/failover times quantitatively against pre-created RPOs. Also paralleling the multiplatform solution is its integration with System z mainframes for disaster recovery functions.
In the case of IBM Tivoli System Automation Application Manager 3.1, however, this is achieved in a slightly different way. First, due to its extraordinary reliability and uptime, System z is a very logical choice of host platform for IBM Tivoli System Automation Application Manager 3.1; organizations certainly want to ensure that their continuity solution is itself as continuous as possible. Second, the solution utilizes a feature known as shoulder-tapping to link itself to the System z-based GDPS (where available). When IBM Tivoli System Automation Application Manager 3.1 determines that application and resource outages exist, it can automatically trigger disaster recovery actions by notifying GDPS.
Part of a comprehensive, Tivoli-driven continuity strategy
Finally, both of the new IBM offerings can be conceived of as only part of a total high availability and continuity solution. By integrating with other members of the best-in-class Tivoli system management portfolio, the continuity benefits described here can be linked and extended to other IT service domains and operational functions.
For instance, IBM Tivoli Monitoring delivers key, real-time status information about different resource dynamics throughout the IT infrastructure. This information can be utilized by the two new offerings to spur high availability for services and improve, automate and simplify overall administration. In the event available storage drops below a critical level for a particular service, IBM Tivoli Monitoring can convey that information to the System Automation offerings, helping to ensure the service continues. Similar benefits are conveyed through integration with IBM Change and Configuration Management Database, IBM DB2 database platforms and others, driving business value throughout IT by maximizing application and service uptime to the fullest possible extent.
|
