Meet vHyperTrace: a read-only appliance that tells you what your VMware estate can become
One OVA in your own VMware environment. It reads the estate, validates the hardware, scores fourteen ways forward and writes the brief, the plan and the runbook. No agents, no internet, nothing phones home.
In this post
Every VMware estate we have worked on had two things in common. The team already knew they had to do something about it, and nobody could agree on what they actually had.
That gap is where modernization programmes stall: not in the cutover, but in the first month, when the spreadsheet disagrees with vCenter and both disagree with what the application owners remember. vHyperTrace is the instrument we built to close it.
What it is
A virtual appliance. One OVA, deployed into your own VMware environment, pointed at vCenter with a read-only account. It needs no internet connection, installs nothing in any guest, and never sends anything anywhere, not even to check its licence. When you power it off, the estate is exactly as it was.
- vCenter: hosts, clusters, VMs, datastores, networking, licences, certificates, alarms
- NSX Manager, SDDC Manager and Aria Suite Lifecycle, where you have them
- Guests, through VMware Tools or opt-in probes over SSH, WinRM, database and HTTP
- Server management controllers over Redfish, so firmware comes from the hardware
- Network flow records (NetFlow, IPFIX, sFlow) for dependencies, if you send them
- An RVTools workbook, if a live discovery is not yet approved
- Hardware validated against the Broadcom Compatibility Guide to component level
- Every build number checked against the interoperability matrix
- Fourteen upgrade and migration routes scored per workload
- Owner actions: the one list of what must change before the chosen path can proceed
- A schedule with a critical path, exportable to MS Project
- The Executive Brief, the Delivery Plan and the Transformation Runbook, in your branding
How a reading becomes a decision
Work is organised as Project → Environment → Run. The project is the engagement; the environment is one vCenter or one whole VCF estate; a run is one reading of it. Each reading moves through five stages.
- 01 Discovery Reads the management plane, the guests you allow and the hardware. Any phase you did not ask for is skipped and says so.
- 02 Analysis Compatibility, interoperability, lifecycle, capacity, security posture, dependencies. Every verdict stamped with the dataset it was judged against.
- 03 Recommendation Fourteen workstreams scored side by side. Readiness checks on the cloud landing zones you shortlist. The decision is recorded with its evidence.
- 04 Plan & Deliver Owner actions, the Planner, runbooks and reports: one handover kit for the sponsor, the programme manager and the administrators.
- 05 Optimize What you would change about the estate you already have, whether or not you move.
Fourteen ways forward
The product does not assume you are leaving VMware, and it does not assume you are staying. It scores every route that is open to the estate and shows what blocks each one.
From application groups to a plan to a handover
Scoring routes is half the job. The other half is turning a reading into work that a programme can run, and three parts of the product do that.
- 01 Application grouping Agentless, with no guest operating-system login. Flow records alone (NetFlow, IPFIX or sFlow from your switches and routers) are enough to produce accurate application groups: every conversation between two addresses becomes an edge in the dependency graph, and workloads that talk to each other are grouped so coupled systems move together in one wave.
- 02 Planning The locked decision seeds the Planner: streams, gates, dependencies and a critical path, with owner actions attached to the gates they block. Export it as MS Project XML when the PMO wants it in their tool, and tailor the default runbook to your change process in Runbook Studio.
- 03 Reporting Reporting writes the Executive Brief for the sponsor, the Delivery Plan for the programme manager and the Transformation Runbook for the administrators, each in its own voice and stamped with the datasets its verdicts used. Word, PowerPoint, Excel, HTML and a delivery kit that bundles the handover. Never PDF, never a price.
Application grouping without agents is the part people ask about first. Agent-based tools map dependencies by installing something in every guest and watching its sockets, which means a login to every operating system and a change to production before the assessment has started. vHyperTrace does not log in to any guest for this. Point your network devices at its flow collector, and the graph builds itself from what the network already sees. Anything the flows did not show is marked as not observed, never guessed.
Who uses it
Consultancies and systems integrators use it to scope a VMware decision in the first week of an engagement instead of the third, and to hand the client documents their own platform team can open and check. Enterprise infrastructure teams use it to walk into a VCF upgrade or a cloud migration with the gates measured, and to re-read the estate before every change window.
vHyperTrace is available now, licensed per appliance, with a short trial on request for organisations with a real VMware decision ahead of them. The button below goes to a short form and then to the founders.
Over the next few weeks we will publish what we learned building it: what actually gates a VCF 9 upgrade, why an RVTools export is not an assessment, how nine cloud routes score differently, and why firmware should come from the hardware rather than the hypervisor. Subscribe below if that is useful.