News from Pulse: Fytte the First

My up-front impression from the Tivoli’s Pulse conference in Las Vegas is that Tivoli is now a lot more than just a collection of Operations tools (in fact, it probably has been, for some time). It’s becoming a new business automation development environment around orchestrating Cloud and other services (which might be externally sourced or written yourself), to deliver business outcomes – although it does automation and optimisation of operations processes and infrastructure too, of course.

This “new Tivoli” was exemplified by Neil Roberts, CIO of Yarra Trams (from Melbourne, Australia), describing the smart infrastructure it uses to manage its passenger experience around its mantra (in my words) of “One Team: doing zero harm; thinking like a passenger; and adopting continuous improvement”. Basically, the outcome seems to come from effective, smarter, enterprise asset management using Maximo, with a mobile interface (the tramTRACKER app). This means that customers can query the status of “their” instrumented asset (the tram) and work out alternative strategies for getting to work if it is broken; while the asset itself is fixed (together with any follow-up maintenance required) and got back into service as quickly as possible by Yarra Trams. Smarter infrastructure with smart asset management becomes a better passenger outcome.

The hard news from Pulse? Well, Danny Sabbah has now moved on again and is a sort of “CTO without portfolio” as CTO and GM of Next Generation Platform. This isn’t, to my mind, a very well-defined role yet, but reflects the fact that IBM very sincerely believes that we are at an inflection point, where smarter infrastructures operating with real-time decision support; the instrumentation of everything in an “Internet of Things”; and orchestration of cloud services more or less on demand; radically changes the way the business understands and uses its technology platforms. Sabbah has a demonstrated ability with innovative technology and has also set up IBM businesses in the past; his new role (as I see it) is to support a unified point of view across IBM’s somewhat silo’d (still) divisions and to make sure that IBM can support emerging technologies even if they are seen as disruptive by established players inside IBM – and possibly even more disruptive by their customers. If Tivoli has now moved one step further along the path that seems to have been developing in Sabbah’s mind ever since I first met him at Rational (when I was asking why Rational and Tivoli weren’t part of the same thing), the fact remains that IBM has to manage its brands and the expectations of its less advanced customers even while it sees an imperative need to be ready for a ‘Next Generation Platform’. Sabbah’s new role and his mandate to work across divisions and technologies is a healthy sign that it is managing this journey well.

Deepak Advani replaces him as General Manager of Tivoli Software. He has a rich technology background and tons of vision and enthusiasm – Tivoli will be worth watching in the next few months, I think (even a name change to reflect its greater scope could be on the cards).

Announcements at Pulse include:

• Open-standards-based cloud offerings around IBM SmartCloud, with flexible deployment options including running on IBM’s Pure Systems platform with accelerated deployment on these optimised infrastructures; including IBM SmartCloud Control Desk and open betas of IBM SmartCloud Monitoring (with betas of Application Insight, Workload Analytics and enhanced Capacity Optimization) and IBM SmartCloud Orchestrator (built on TOSCA, OpenStack and OSLC standards; and supporting DevOps-style deployment of hybrid and public clouds). There is a new IBMSmartCloud Storage Access self-service portal and IBM Tivoli Storage Manager Ops Centre (for root cause problem analysis with an admin dashboard) is in open beta. Note the emphasis on analytics and associated operational decision support; and the more accessible beta programs.
• Security was a Pulse theme, with an impressive ‘whole systems’ solution that copes with one-off custom attacks. I just hope that placing all this in the Security Division doesn’t lead to developers thinking that security is “someone else’s problem”; although that’s not how IBM sees it). IBM Endpoint Manager has been enhanced with simpler life-cycle management, patching for virtual machines offline, before they can be accessed, and export of Endpoint metrics for QRadar security intelligence – this last is part of the vital integration of security analytics with Big Data.
• The DevOps story continues strong, with enhanced visibility, control and automation across silos. The Maximo Health, Safety and Environment Manager is enhanced with audits and surveys for better risk visibility and preparation and certification facilities. The IBM SmartCloud Control Desk is enhanced with social collaboration features and there’s a new DataCentre Infrastructure Management partnership with Emerson.
• Predictive Asset Optimisation is a new Signature Solution which brings Big Data, predictive analytics and asset management together in a framework. Perhaps this fits conceptually with enhancements to Maximo and Tririga, to optimise the deployment of a mobile workforce (with field deployment newly extended to Android).

Finally, we got a demonstration of Oasis’ TOSCA-based cloud interoperability “working”: press a button to export a SAP cloud app to TOSCA; press a button to import from TOSCA to IBM. OK, that raises lots of questions and SAP is not actually supporting TOSCA in its products yet (and may never do so; there are politics and customer demand issues, although it is on the standards committee), but it’s a start. IBM and CA Technologies, plus other big names, are behind TOSCA – although I noted that Amazon, for instance, doesn’t support TOSCA yet. TOSCA could be incredibly important as, apart from anything else, it shows that IBM and SAP (at least) actually realise that their customers probably won’t accept lockin to a single vendor as they build hybrid (public/private) cloud environments. Even if IBM is good for your private cloud, you might well want to use, say, Amazon (pity about TOSCA) for your public cloud apps… Most customers like preserving choices!

The elephant in the room, as usual, is interoperability testing (and despite talking to IBMers at the conf, this still isn’t entirely clear to me): who manages this (open source “plugfests” were mentioned; someone has to cover administration costs for these); how cheap and accessible will be the testing suites; will the big players using TOSCA try to freeze out innovative small newcomers. Interoperability testing suites are expensive to build; if access to them is expensive, or if you have to pay a new fee every time you fail the test and need to find new issues, or if invites to “plugfests” are restricted, interoperability testing can be used to give established players effective control over a genuinely open standard.

Nevertheless, it’s early days for (to give it its full name) OASIS Topology and Orchestration Specification for Cloud Applications (TOSCA) and I’m still optimistic. Cross-vendor cloud interoperability (or the lack of it) remains a major issue for any sort of cloud vision today.

So, that’s about it for now. Off to breakfast. I’m sure there’ll be more to come.