IBM Turbonomic

IBM Turbonomic automates application resource management across hybrid and multicloud infrastructure, continuously optimising compute, storage and network resources in real time while enforcing policies. The framing is performance assurance rather than cost reduction, and that distinction is the reason to consider it over a cost management tool.

Most cloud optimisation software identifies waste and recommends cuts, which puts the infrastructure team in an uncomfortable position: every recommendation is a bet that performance will hold, and the person who accepts it owns the outage if it does not. Turbonomic inverts that by treating the objective as assuring application performance, with cost reduction as the consequence of removing resources that were never needed rather than as the goal being pursued.

Continuous real-time optimisation rather than periodic reporting is what makes that credible. Workload demand varies by hour and by season, so a rightsizing recommendation generated from last month's data is already describing a different system. Acting continuously means resources track actual demand rather than a historical snapshot of it.

Policy enforcement is what makes automation acceptable to the teams responsible. An organisation will not let software resize production infrastructure unsupervised without constraints on what it may do, when, and to which systems, and expressing those constraints as policy is what converts a recommendation engine into an automation platform.

Full-stack visibility spans application through to infrastructure, which is necessary for the approach to work at all: deciding whether a virtual machine is oversized requires knowing what the application on it actually needs, not merely what the operating system reports.

The product targets hybrid and multicloud complexity specifically. A free trial is offered, with case studies, integrations and resources published. Pricing is not published as plan figures.