At a recent analyst briefing an IBM manager implied that the Unified Resource Manager used by the hybrid zEnterprise to manage the zBX would converge with the Flex System Manager used to manage the new PureSystems. “We’re working on it,” he said earnestly. Someone at IBM might be working on it or at least thinking about it or arguing about it, but it is not going to happen in any foreseeable future.
That is the definitive word from Jeff Frey, IBM Fellow and CTO for System z: “Flex Manager and the Unified Resource Manager will not come together,” he told DancingDinosaur. That does not mean the zEnterprise/zBX and PureSystems won’t play nicely together, but they will do so higher up in the IT stack. “We will federate the management at a higher level,” he said. Today, that pretty much means organizations using both platforms, zEnterprise and PureSystems, will have to rely on Tivoli to tie the pieces together and manage them. At the lower levels in the stack where the hardware lives each platform will require its own management tooling.
This may be a disappointment to those who mistakenly thought IBM’s idealized vision of efficient hybrid computing meant one administrator working at a single console could manage everything happening across the various platforms, from a misbehaving partition to monitoring the resource consumption of a particular application. This was never the intent, Frey noted. You still can do it to some extent, but it won’t be through the Unified Resource Manager.
Instead, Tivoli will provide a federation layer to enable higher level, logical management across both systems. When you need to manage some physical aspect of the underlying hardware you still will need platform-specific tools like Systems Director.
This shouldn’t be a surprise; the zEnterprise/zBX and PureSystems start from two different views of control. The zEnterprise brings the mainframe tradition of disciplined centralized control, and that control point is the z. PureSystems views control from the distributed perspective. These are not compatible views, which are why it makes sense to federate as much management as possible at the high end of the stack, far above the nitty-gritty of the underlying hardware.
As IBM moves forward with the next advances to the zEnterprise/zBX and to PureSystems the situation may seem even more confusing unless you stay focused on the point of control issues. For example, expect some IBM improvements incorporated into PureSystems hardware to make it into the zBX, but that does not mean the Flex System Manager will be able to manage the zBX.
Similarly, IBM is planning to push zBX scalability beyond the 112 blades the box supports today as well as adding clustering capabilities. The blade count expansion combined with the technology enhancements brought over from PureSystems, Frey hopes, should make clear IBM’s long term commitment to the zBX and zEnterprise hybrid computing.
At the same time, IBM is enhancing PureSystems for the purpose of scaling it beyond its current four units. This will give it something more like the Ensemble approach used with the System z. An ensemble is a collection of one or more zEnterprise system nodes where each node is comprised of a z and its optionally attached zBX. As such an ensemble can consist of a single z with no zBX attached, or two to eight CPCs where at least one has a zBX attached. The resources of a zEnterprise ensemble are managed and virtualized as a single pool of resources integrating system and workload management across the resulting multi-system, multi-tier, multi-architecture environment.
In the end, both the zEnterprise/zBX and PureSystems are poised for scaling. That should become obvious with the next rev of the zEnterprise, which should happen in 2013 if IBM sticks to its historic 3-year mainframe rev cycle. The new rev will address some of the current gaps, like partition mobility and live image mobility as well as the usual bigger and faster story.
With two hybrid computing platforms the hybrid approach is here for real at IBM. Now the question is which workloads can most benefit and on which platform. Expect DancingDinosaur to explore this question in the coming months.