Recent research shows the mainframe is alive and well. In a qualified industry survey, we polled 600 customers and 80 percent expect their MIPS usage to hold firm, with more than 56 percent projecting MIPS increases, and the majority of those bringing new applications onto the platform as a component of that growth. Not only is their usage growing, but they are investing as well, with 30 percent of the respondents expecting to spend more on software for the platform in the next 12 to 24 months. Lest one think this growth is coming only on the back-end of increased transaction volumes or organic growth in legacy systems, we also see increasing penetration of MQSeries and z/OS-based WebSphere, with respondents who reported MIPS growth plans also showing a significantly higher adoption in those two technologies.
So what is driving this growth in usage as well as in adoption of new technologies and types of workload? The answer is quite simple: The mainframe remains one of the highest performing, most reliable platforms in the world, and by some accounts, it already handles more than 70 percent of the planet’s transaction volume each day. IT sites are surrounding the platform with a variety of distributed front-end systems, application servers, and other types of devices, yet the platform remains the “information and transaction hub” of their business-critical environments.
This evolution has resulted in increasingly complex and distributed systems with the mainframe holding a central role for consolidating and processing data. As the machine in the center of these networks, the performance and availability attributes of the mainframe become of utmost concern in the context of a much wider ranging system. This also drives the need for open and reliable mechanisms for collaborative computing, resulting in a broader adoption of MQSeries and WebSphere.
While one may consider the core platform quite reliable and robust, the problems IT faces now span the entire system, and the mainframe must be managed in this broader context. The problems raised by the survey respondents focus on issues that are becoming exposed by this massive increase in the overall complexity of the environment.
From our research, we exposed four clear areas of concern that repeatedly surfaced:
Root cause analysis/end-to-end diagnosis: Respondents report they have a very good handle on managing subsystems and components. But, as applications and services increasingly span wider and more complex computing networks, discovering the source of an issue first seen in terms of the service or application context is becoming increasingly difficult. Being able to correlate health and performance metrics from the mainframe to overall service or application status and then quickly traverse between the two is a key need in the market. Service-Oriented Architectures (SOAs) also add to the complexity of this type of diagnosis.
Managing dynamic SQL: The rise of distributed Webbased applications as well as ad hoc reporting and query tools have driven up the amount of dynamic SQL running in the environment. Dynamic SQL is inherently more difficult to manage and troubleshoot, resulting in a widespread need for tools to better manage this problem. And, as companies run more packaged applications and software developed offshore, their control over changes in the application environment diminishes.
Change, configuration, and process compliance: The mainframe is operating as a component within the broader Web of the applications’ environment, and as a result, change and configuration management can’t be done in isolation. IT shops must address this need as well as do it under increasing pressure from regulatory issues such as Sarbanes-Oxley compliance or internal process improvement needs.
Security and privacy: As the center of most large IT organizations data and transaction hubs, the mainframe is housing critical data that must be protected, all the while granting more access and connections to the outside world. Our research has shown this topic to be of concern in a wide variety of organizations, with the use and manipulation of production data in test of particular interest.
Summary The mainframe is alive and well with customers investing in new applications. Contrary to the notion that the mainframe is a mature and stable legacy platform, we see it taking on new and exciting roles in the ever-more complex and demanding IT environment.