The railway had more than 5,000 virtual machines. That sounds like a number you could put in a spreadsheet, which was precisely the danger. A machine has a name, a processor, perhaps a license. None of those fields tells you whether moving it will quietly break a timetable, a payroll run, or some service nobody has touched since the last infrastructure team left. In CloudSphere's account of a North American railway assessment, its software found more than 15,571 commercial and custom applications running across the estate. The server count had been the visible problem. The web of dependencies was the real one.
The short version
- CloudSphere's Illuminate360 discovers infrastructure and maps applications by the business services they support.
- It sells to enterprise IT and security teams, and to service providers planning cloud migrations for clients.
- Its sharpest use is deciding what can move together, what is past end of life, and which cost scenarios deserve scrutiny.
- The company emerged from HyperGrid and iQuate in 2020; a $12 million Series B in 2024 pushed its visibility pitch into security operations.
A list of machines is not a map
Consider OneNeck IT Solutions, a managed-services company that assesses customers before moving them to Azure, managed Azure, or private cloud. Its team had been interviewing application owners, building spreadsheets, and mapping hybrid estates manually. CloudSphere's published case study says those efforts ran over budget and schedule. Other discovery tools OneNeck evaluated could find virtual machines, but did not provide the service context it needed. Some required agents on every customer server, with their own installation and removal work.
“Most of the discovery tools we evaluated would only discover virtual machines, with no application service context.”Dan Kepplinger, OneNeck IT Solutions
OneNeck wanted a view from the top down: start with the service a customer recognizes, then trace the applications and infrastructure underneath it. CloudSphere's agentless discovery supplied that view. OneNeck says it now uses the platform in assessments and for repeat snapshots of a customer's hybrid estate. The distinction sounds fussy until a move group is assembled. If two apparently separate applications share one database, that shared edge is more useful than either application's server count.
The trick is context: a server becomes useful when its role in a service is visible.
The arithmetic of an unseen estate
CloudSphere reports that its railway assessment exposed 563 operating systems past end of life. It also calculated a 39% reduction in Microsoft SQL Enterprise core counts and a possible 57% annual total-cost reduction for an AWS scenario using right-sized virtual machines and three-year reserved instances. Those figures are assessment outputs, rather than a public accounting of money saved after migration. They still change the conversation. The railway can price a specific plan, identify unsupported software, and decide which dependencies make a move risky.
An unnamed global pharmaceutical company produced a similar lesson on a shorter clock. CloudSphere says one week of discovery identified more than 1,000 applications, 57 unsupported operating systems, and a potential 55% annual cost reduction under a three-year reserved-instance and right-sizing plan. The interesting number is not simply the percentage. It is the time between asking “what have we got?” and having a workable picture. Any buyer should test that speed against its own network access, estate complexity, licensing terms, and the cost of the target cloud plan.
Two old businesses, one stubborn question
CloudSphere was announced in 2020 when HyperGrid, a cloud-management company, combined with iQuate, an agentless discovery and application-mapping company. The joining matters more than the naming exercise. One side had tools for governing cloud cost, security, and migration. The other could discover the older estate that had to be governed or moved. CloudSphere's application-centric approach is the argument that these should be one conversation. It groups resources by application and business service instead of making the server or cloud account the final unit of analysis.
The product now called Illuminate360 is a SaaS platform for finding assets across on-premises, hybrid, and multicloud environments, mapping their relationships, and layering outside data over service maps. IT teams can use it for move groups, cloud suitability, cost and license reviews, and end-of-life analysis. Security teams can use the same map to ask which business service is exposed when an alert arrives. CloudSphere calls the resulting network of assets and relationships a Knowledge Graph. That phrase can sound decorative; a useful graph answers a plain question: if this component goes down or moves, what else follows?
Its 2024 $12 million Series B, led by Atlantic Bridge with participation from ST Engineering and other existing investors, was explicitly tied to expanding into security operations. In 2025, CloudSphere announced an AWS strategic collaboration focused on VMware migration pathways, application service paths, total-cost work, and modernization recommendations. It had earlier been selected for Microsoft's Azure solution-assessment program. These relationships place CloudSphere in the planning layer beside cloud providers and services firms, where the migration is designed before it is executed.
Who buys a better question?
The direct buyer is usually an enterprise trying to make sense of a large inherited estate. A CIO may want a migration sequence or cost forecast; a CISO may want to know where unsupported software sits; a service provider may need to assess many client estates without weeks of interviews. OneNeck is a named example. CloudSphere also publishes anonymized cases involving a railway, a bank, and a pharmaceutical company. The bank case reports 4,718 applications discovered and 439 end-of-life operating systems. That is hardly a glamorous sales pitch. It is a useful one.
CloudSphere sells enterprise software and works through partners; it does not publish a standard price. Its comparison set spans manual inventories and configuration databases, provider migration-assessment tools, and specialist service-mapping platforms. Its claim is a practical difference: agentless discovery joined to an application and business-service map that can be reused across migration, optimization, and security work. The test is whether that map is accurate enough in the buyer's own estate to change a decision.
The company has also followed visibility into AI. In June 2025 it brought former enterprise CISO Paul Haywood onto its board while promoting shadow-AI discovery. In January 2026, LatticeFlow AI acquired AI Sonar, a wholly owned CloudSphere subsidiary built around AI discovery, along with its engineering operations. CloudSphere’s LinkedIn page now describes both CloudSphere and AI Sonar as part of LatticeFlow AI; the January acquisition announcement specifically named AI Sonar. Enterprises keep adopting new technology faster than they can catalog it. CloudSphere's recurring proposition is that governance begins with an honest inventory, and an honest inventory must show relationships.
What can another company copy? Begin a migration assessment with a business service, identify the machines and applications beneath it, and verify the links with the people who run them. Calculate costs for named scenarios, then keep projections separate from realized savings. Repeat the map as systems change. If the estate is small, stable, and already well documented, a dedicated discovery platform may buy little. If access is restricted or the environment contains unusual legacy systems, any automated map will need validation. In a tangled estate, though, moving first and understanding later is an expensive way to learn geography.