When investigating clustering technology, embrace the Vision.
Last month, Vision Solutions announced the integration of its Vision Cluster1 solution with IBM's Systems Director Navigator.
Cluster1 is Vision Solutions' unique cluster management solution for System i, but many customers are still a bit confused about the differences between the use of clusters on the System i and the traditional use of clusters in the UNIX or Linux environment.
Thomas M. Stockwell, on behalf of MC Press Online, conducted an interview with Craig Johnson, Vice President for Research and Development at Vision Solutions, investigating how the Vision Cluster1 solution addresses these concerns.
Stockwell: Can you identify some of the differences (if any) between clustering for failover, load balancing, and other uses typically deployed on UNIX/Linux machines, and how System i clustering is used?
Johnson: The term "cluster" has different meanings.
In System i, it refers to high availability solutions built on top of IBM i5/OS cluster services technology. First shipped in V4R4 OS/400, cluster services is categorized as middleware infrastructure that provides high availability cluster technology for IBM i. IBM's technology provides a framework to help manage the availability of clustered IBM i systems or partitions as well as the resources that exist within that cluster.
Vision Cluster1 manages the customer's HA using the i5/OS cluster framework. It provides value for all customers interested in adopting the technology within their environment, but many are unclear of the benefits and the requirements for doing so.
Vision has the expertise in our engineering centers, our professional services team, our highly qualified business partners, and our customer care organization to not only help a customer understand and deploy a cluster-based solution, but also help the customer manage it and care for it once implemented.
Vision's team can guide customers through the process of deploying a cluster-based solution that will significantly increase the availability and resiliency of your critical business systems.
Vision Cluster1 will also work with any environment and any application.
Stockwell: How does this solution work for the benefit of System i customers?
Johnson: Today, we have customers that require a level of availability and resiliency beyond the typical two-system environment. Some of our customers not only require a local high availability solution to manage the availability of critical systems and applications for planned outages such as maintenance, application upgrades, and OS upgrades as examples, but also require disaster recovery options due to catastrophic failure at a regional site.
A majority of our customers have deployed a non-cluster-based HA or DR solution comprised of one backup system. It is a simple solution that provides a significant improvement in availability beyond a single-system environment.
Clustered environments consisting of three or more systems provide a level of resiliency well beyond what a typical two-system environment can provide. Even if a customer is required to roll over for a planned outage, a clustered solution will continue to provide DR support.
Stockwell: How does the use of Vision Cluster1 interface with Vision Solutions' previous offerings?
Johnson: All of our customers today can use Vision Cluster1 in conjunction with one of our enterprise Vision HA offerings. Vision Cluster1 also supports IBM technologies such as switched disk or switchable I/O pools as well as other SAN-based technologies.
Given the variety of data resilience technologies, customers continue to use our Vision HA products within their environments for a variety of reasons. It is the easiest solution to deploy, it requires little or no change to our customer's application, it provides the greatest level of flexibility from a deployment perspective, it is the most efficient in terms of system resource and communication requirements and provides the fastest recovery times from a switching perspective than any other resiliency technology offered on the platform.
Stockwell: How does Cluster1 help System i customers better approach the goal of continuous availability? Where is the benefit of using Cluster1 in an HA requirement? What about pure DR? Is there an advantage?
Johnson: Cluster1 allows customers to incorporate another level of resiliency into their environment with the addition of redundant systems or partitions. For instance, as the number of systems increases, so does the complexity of monitoring and managing that environment. Vision Cluster1 is designed to make it very easy to monitor and manage those environments. Vision Cluster1 gives you visibility to the health not only of the systems that comprise the cluster but the applications, the devices, and the critical business data.
For instance, Vision Cluster1 has built-in audits that continually check the health of your environment and report out any issues that require your attention. In many cases, built-in healing support will automatically check and repair issues that can negatively impact the availability of your environment.
Stockwell: IBM Systems Director Navigator is IBM's way to Web-enable a Web console interface for System i administration. How does Vision Cluster1 integration complement this IBM product?
Johnson: The integration of Vision Cluster1 directly into IBM's Web-based console shows you how close our relationship to IBM is and how important that relationship is to both companies. The integration of Vision Cluster1 allows customers that are utilizing IBM's Web-based console to not only monitor and manage the IBM i systems, but to use it to also monitor the availability of those systems within a single user interface.
An integrated solution helps minimize the need to use multiple interfaces to monitor the overall state of your IBM i systems, including the availability aspects. Vision also provides separate interfaces that extend the capabilities and features beyond what is currently available in the integrated support.
Stockwell: Where do you see IBM's focus on System i IASPs for replication prompting Vision Solutions' strategic clustering technologies in the months and years ahead?
Johnson: For many customers that are interested in segmenting their applications into IASPs--either for specific business reasons, such as allowing multiple footprints of the same application on a single partition, or taking advantage of hardware mirroring technologies for disaster recovery purposes, or as part of their backup strategy--Vision Cluster1 not only supports those technologies but can be set up to help monitor and manage those technologies within a single interface.
One example of a production environment that one of our customers has deployed is a scenario where a customer has adopted switchable I/O pool shared by two partitions locally within their data center to improve their system resiliency. This implementation requires the application to be converted to use IASPs.
This customer also utilizes a Vision HA product for replication to a remote site to increase their overall business resiliency.
Vision Cluster1 provides a solution that can help monitor and manage a mix of technologies that is not offered by anyone else.
With the integration of IBM i and IBM AIX on Power, many customers are beginning to consider a strategy of deploying or utilizing their SAN infrastructure for IBM i. Many can leverage that infrastructure today using a Vision Cluster1 and HA solution with virtually no changes to their application environment required.
Vision is committed to embracing all the technologies being utilized by Power customers and leveraging those technologies to take our customers' business resiliency to the next level. In addition to our availability solutions on IBM i on Power, Vision also offers products EchoCluster and EchoStream for customers on AIX. As a result, Vision provides availability solutions across IBM Power. We are very well-aligned with IBM's integration strategy on Power.