Veeam 100 Summit 2026: The Green Content Recap

From 26 to 28 October 2026, I was in Lisbon for the Veeam 100 Summit, Veeam’s yearly gathering of its top community members. It was billed as the biggest Veeam 100 Summit ever: Vanguards, Legends, Veeam MVPs, and User Group leaders were joined for the first time by Rising Stars and Global Systems Integrators.

Three days of deep-dive sessions with Product Management, R&D, and field CTOs is a lot to digest, so this post captures what stuck with me.

Everything in this post comes from the Green Content kit Veeam sent to attendees after the event. If a session is not covered here, it is because it was Red or Yellow, not because it was not worth talking about

The agenda ran four parallel rooms over two full days, plus a welcome evening and departures. These are the sessions whose content is Green, in the order they ran:

  • Mon 26 Oct – Welcome Reception.
  • Tue 27 Oct – Opening General Session, with sponsor sessions from ExaGrid and Scality.
  • Tue 27 Oct – Zero Trust Data Resilience.
  • Tue 27 Oct – SureBackup Without Borders.
  • Tue 27 Oct – Veeam Intelligent ResOps: Your M365 Data is Smarter Than You Think.
  • Tue 27 Oct – Community awards, the traditional toast and the group photo.
  • Wed 28 Oct – Public Clouds: what’s next for self-managed and backup as a service.
  • Wed 28 Oct – An Expert’s Guide to the Veeam Forums.
  • Wed 28 Oct – Veeam 100 Summit Celebration.

The Veeam 100 is a community before it is an event, and the opening session made that clear from the first slide. It was also the biggest Veeam 100 Summit so far, with two groups joining for the first time.

Who was in the room?

Veeam Vanguards: influencers who share on their own external platforms: blogs, podcasts, videos and talks

  • Veeam Legends: the most active people on Veeam’s own properties: the R&D Forums, the Community Hub and the User Groups.
  • Veeam MVPs: Veeam employees in customer-facing technical roles who share and engage well beyond their job description.
  • Veeam User Group leaders: the members who organize local VUG meetups around the world.
  • Rising Stars: New group in 2026.
  • Global Systems Integrators: New group in 2026.

Veeam calls the first three its “three-fold community strategy”.

What I like is that it covers all people sharing knowledge: on their own channels and blogs, on Veeam’s channels, and inside the companies.

  • 07 new or reactivated Veeam User Groups.
  • 84 VUG events: 36 virtual and 48 in-person, 12 more in-person events than the year before.
  • 19 community activations at VeeamON Tours and Veeam Unplugged events across Europe, Africa, ANZ, North and South America.
  • EUR 2,231 donated to WWF by the community through the SWAG store.
  • 05 years of the Veeam Legends program.

Your Feedback Actually Ships

The message from Product Management was simple: you are not annoying.

They want feedback because it motivates the R&D teams and not just the product managers.

To prove the point, they showed feature requests that came from people in this room and were shipped into the platform during 2026:

  • A web UI for anyone who struggles with the Windows console.
  • Storage expansion on the Veeam Software Appliance and Veeam Infrastructure Appliance.
  • GFS (weekly, monthly, yearly) immutability on the Hardened Repository.
  • Sharing one Hardened Repository server between several backup servers.

It is a good reminder that posting a well-written feature request on the Veeam R&D Forums is not shouting into the void.

The Pay It Forward Challenge

The community team also gave us homework for the week:

  1. Be brave. Meet someone you don’t know and break the ice.
  2. Make a real connection. Swap LinkedIn details and actually keep talking during and after the Summit.
  3. Pay it forward. Ask your new connection two questions: “What can I do to help you?” and “Is there a business or community project we could work on together?”

Simple, but it works. Some of the best conversations I had in Lisbon started exactly like that, between sessions, over coffee, lunch, or at dinner.

Moments that Make it Different

What separates the Veeam 100 Summit from a regular conference is the moments in between: the welcome reception on the first evening, the Brad Jervis Memorial Award, the traditional toast and the group photo at the end of the first full day, and the celebration night off-site on the last day.

You leave with technical notes, but also with friends across dozens of countries.

Want to Join?

Veeam Vanguard nominations open in December 2026. Follow the Veeam blog so you don’t miss the announcement.

If you are already active on the R&D Forums, the Community Hub, or a local User Group, the Veeam Legends program is another way in.

Two Veeam architects gave the first public look at Veeam’s new security best practices framework, now published at bp.veeam.com/security and written for Veeam Backup & Replication v13.

Instead of a closing Q&A, they used the last 15 minutes as an open discussion to collect feedback.

Their opening line set the tone: many of the ideas may sound obvious, but putting them all together into one coherent framework is not.

Why Zero Trust, and why “Data Resilience”

They picked Zero Trust because it is vendor-neutral, not a marketing trend, and backed by the wider security community.

The history is long: coined at Forrester Research in 2010, proven at scale by Google’s BeyondCorp in 2014, and formalized by NIST SP 800-207 in 2018.

The gap they point out is that Zero Trust governs who can reach a resource in production. It says nothing about what happens to that data once it is backed up, replicated, or kept for years.

That is exactly where modern ransomware goes first: remove the ability to recover, and the victim pays.

The attack, step by step

The key insight is that stages 1 to 4 happen quietly, often days before encryption, and each is a chance to stop the attack.

The four pillars:

  1. System Resilience: secure by design, a hardened control plane, immutable and diverse copies.
  2. Least Privilege Access: granular roles, Four-Eyes approval, MFA everywhere.
  3. Assume Breach: micro-segmentation, end-to-end encryption, trained people.
  4. Proactive Validation: visibility and baselines, threat detection, SOC integration.

Each concept follows the same path: the vendor-neutral Zero Trust idea, what it means for backup, and the VBR feature or design choice that implements it.

On top of the pillars sit cross-cutting security principles, ideas that only become a principle once they appear independently in at least three places in the written content.

What I wrote down

  • Time-bound defense. Credentials, keys, and certificates should expire. With gMSA and Kerberos, a Kerberoasted hash is stale before it is cracked. The flip side is “harvest now, decrypt later”: a 10-year GFS backup gives an attacker 10 years to break its key. TLS certificate lifespans drop from 398 days today to 47 days by 2029.
  • Secure by default is not secure by design. The Veeam Software Appliance ships hardened (Just Enough OS, DISA STIG, built-in immutability, updates managed by Veeam), but it reflects a generic threat model. Start from that baseline, review every setting, and add your own controls.
  • Never depend on what you protect. Best: a management domain in a separate forest with a one-way trust. Good: a workgroup. Never: the production domain. With v13 there are no passwords between components, and the appliance never joins a domain.
  • Beyond immutability. Treat 3-2-1 as a heuristic, use different technologies per copy, prefer bare metal for repositories, and isolate copies with separate credentials, network paths, and management.
  • Least privilege in practice. Map a named account or group to Backup Administrator, then remove BUILTIN\Administrators. Use 15+ character passwords for users and 25+ for service accounts kept in a vault. Disable accounts rather than delete them.
  • No single identity can destroy your backups. Four-eyes authorization for damaging operations, the VSA Security Officer (never disable it in production: it is irreversible without a reinstall), and TOTP MFA, which also works in dark sites.
  • A zone model. Untrusted, DMZ, Management, Trusted, Restricted, and Audit zones with default deny. Air gap is segmentation taken to the extreme. v13.1 moves backup traffic to port 6162.
  • Encryption everywhere. TLS 1.3 with mutual TLS, AES-256 at rest, an external KMIP KMS, and in v13.1 an ML-KEM (FIPS 203) post-quantum TLS handshake.
  • People are a layer. Backup Admins and Security Officers are named targets. Run phishing simulations and tabletop exercises, and encourage blameless early reporting.
  • Detect at every stage. Inline entropy scan and guest index scan during backup, Scan Backup with Threat Hunter and YARA at rest, Secure Restore on the way back. Add the Incident API for EDR/XDR, honeypots, and SIEM integration.

The final slide replayed the same five-stage attack, this time with a control stopping each stage.

The attack can still happen; what changes is that it ends in a clean, proven recovery instead of a ransom payment.

If you design or run Veeam environments, read the available guides.

A green backup job proves one thing: the data landed.

It does not prove the VM boots, that services start, or that the application works.

That is what SureBackup is for, and this session, presented by two Veeam MVPs, asked a timely question: what happens to that proof when customers leave VMware for other platforms?

Thanks to the Veeam Universal License, backups move easily.

Auditors, however, still ask, “Can you prove it restores?”

The challenge they set themselves: deliver recovery verification on any platform using only the VBR REST APIs that ship today. No new product, no waiting for native support.

SureBackup’s four pillars, and the gaps

On VMware and Hyper-V, one click hides four moving parts: an isolated network sandbox (with a proxy appliance doing NAT so VMs keep their production IPs), a mount engine (vPower NFS running the VM straight from backup), an in-guest prober (heartbeat, ping, application tests, optional malware scan), and a deterministic cleanup.

From a homelab to OmniStage

It started as a homelab passion project.

One of the presenters moved his lab to Proxmox, missed SureBackup, and built pmx-lab: VXLAN SDN zones mirror production subnets, a gateway VM impersonates the real gateway, and health checks run through the QEMU guest agent (exit code 0 means healthy), with console screenshots as proof.

It proved the pattern, but it was one three-node cluster, one CLI, and still a manual restore click.

The next step was the cloud. Using the VBR v13.1 REST API (x-api-version 1.3-rev2), the flow is just four calls: find the newest restore point, start Instant Recovery to Azure with verifyVMBoot, poll until the VM is Ready (about 8.5 minutes), then unmount and discard every change. Capacity is rented only for the minutes a test runs.

That became OmniStage: first a Go CLI for Azure, then a full web app with multi-user support, a Live Run Viewer, alerting, SMTP reports, webhooks and Entra ID SSO, and finally a multi-platform tool with pmx-lab folded in as the Proxmox provider.

A run resolves dependencies, boots stages in order (domain controllers and databases first), runs role-based probes (TCP 88/389/445 for a DC, 1433 for SQL Server, an HTTP 200 for a web server), then reports and tears down. If something fails, we can hold the VMs for live triage. The live demo is backed up on Proxmox and verified in Azure.

Six lessons for anyone building on VBR

  1. Network isolation is key: nothing gets out of the sandbox, but the right systems can still reach in.
  2. Public APIs are the superpower. Private APIs are a fallback, not a plan.
  3. Port checks travel to any platform; guest scripts go deeper but are not portable.
  4. AI coding needs a sharp brief: Claude delivered best when scope and direction were nailed down first.
  5. Scope tighter than you think: they cut down to one on-prem and one cloud platform.

Their wishlist for the Veeam Data Platform: public Instant Recovery APIs on every hypervisor, SureBackup for UHAPI platforms, VRO support for any SureBackup platform, and SureBackup for Cloud Connect providers.

The OmniStage community roadmap includes a pre-boot malware scan, custom verification scripts, and VSPC connections. A great example of what the community can build on top of the product.

The Veeam Data Cloud team showed where Veeam is heading after joining forces with Securiti AI. Their framing: we moved from a Backup & Recovery era (“assume restore”, 1980 to 2017), to a Cyber Resilience era (“assume breach”, driven by ransomware), and now into an Agentic era: “assume drift”, with autonomous agents acting on data at machine speed.

The tagline was “There is no AI without Data Security. There is no Trust in AI without Data Resilience.”

Veeam Intelligent ResOps (iResOps) is an add-on for Veeam Data Cloud (VDC) for Microsoft 365 that brings that thinking to SharePoint, OneDrive, Teams, and Exchange. It is built around four use cases:

  1. Protection intelligence: protect what matters most, reduce redundant, obsolete, and trivial (ROT) data, and prioritize sensitive data.
  2. Recovery intelligence: not just “can I recover it?” but “should I?”, so you do not reintroduce sensitive or risky data.
  3. Safe AI: Copilot and agents are only as safe as the data they can reach.
  4. Investigation and impact analysis: find what was touched and the blast radius quickly.

What it does

  • Intelligence Agent: ask questions in natural language over production and backup data, for example, “Which files with sensitive data are accessible by AI agents?”
  • Backup intelligence: check whether a system is protected, prioritize those holding sensitive data, and restore precisely instead of in bulk.
  • Data minimization: find duplicates and old files that are a retention risk, then quarantine or archive them automatically. Less ROT means smaller backups, shorter windows, and cleaner data for Copilot.
  • Access intelligence: identify human and non-human identities and spot over-permissioned ones, such as sites with PHI open to every hospital employee.
  • Classification and labeling: AI-based classification with graph context to find “toxic combinations”, and automatic labels in SharePoint Online and OneDrive through Microsoft Information Protection.
  • Undo AI mistakes: an agent activity log shows what each agent read, wrote, or deleted, and with what permissions. Because it sits on VDC for Microsoft 365 backups, fixing a bad agent action is a file-level rollback rather than a rebuild.

The Purview story was neat: sensitivity labels are imported from Microsoft Purview, iResOps classifies a file as PII, the matching “Top Secret” label is applied in Microsoft 365, and Copilot can no longer open the file. Veeam positions iResOps as complementing Purview and Agent 365 rather than replacing them: Agent 365 governs the agents; iResOps makes sure the data they reach is accurate and protected.

Licensing is per user per year in two tiers: Foundation (core intelligence) and Advanced (adds access intelligence, classification, and labeling), and the whole M365 tenant must be licensed.

The Demonstration

The demonstration walked through the product end to end: an M365 Intelligence Overview dashboard (data elements by classification, agent-accessible data volume, sites shared externally with sensitive data), a Graph Explorer, the Universal Data Explorer across OneDrive, Teams, SharePoint, Outlook and VDC for M365, and policies for data labeling, file insights, file quarantine and file ROT.

One ROT policy flagged sensitive files older than six years.

What stuck with me was the Agent Activity Timeline, which shows exactly which files an AI agent accessed or modified, minute by minute, next to the available restore points.

Product Management covered what is coming for Veeam’s self-managed public cloud backup and for Veeam Data Cloud.

Public cloud management in the VBR Web UI

The reason is simple: customers do not like too many consoles. VBR 13.2 brings public cloud backup management for AWS, Azure, and Google Cloud into the VBR Web UI:

  • Appliances: deploy a new backup appliance from VBR in a few clicks, or connect to an existing one, and manage them all in one list.
  • Policies: view and filter backup policies across every connected cloud in one grid, and create SLA-based or schedule-based policies per workload with an inline wizard.
  • Protected data: a single flat table of every cloud backup, with a one-click inline restore.
  • Repositories: manage storage for all appliances, including Veeam Data Cloud Vault, in an admin-only section.

Later versions will add deeper Backups and Repositories pages, plus new Workers and Policy Templates pages.

What is new per cloud

The AWS malware detection flow is worth a closer look. You enable it with a toggle in the backup policy. The EC2 file system is indexed during backup, and the index is stored in S3.

VBR then reads it through an external repository and runs the scan with Veeam Data Analyzer, looking for suspicious files and extensions, mass deletions, and extension changes.

Direct Restore to OCI

This one tells a nice story.

With the VMware exodus, OCI has become a major migration target, and many customers choose it for cost.

Until now, restoring an agent backup to OCI meant seven steps across three tools: export a VMDK, upload it, create a custom image, remember to enable UEFI boot (skip it and the machine will not boot), launch the instance, attach block volumes, then restore data disks.

With Direct Restore, it becomes three steps in one tool, with the same wizard as for AWS, Azure, and Google Cloud.

Veeam Data Cloud for Azure

Planned for this year: custom SLA templates for Azure SQL and Azure Files, a test of Azure SQL credentials before you save a policy, ad-hoc snapshots and backups with their own retention, ransomware detection for VMs, VNet backup, and backup copy jobs to another region.

Two Product Management analysts gave a short, practical session on getting the most out of forums.veeam.com. The forums are for product design questions, announcements, feature requests, and “how can I accomplish this?” topics. They are not a technical support forum.

My notes:

  • Post in the right product forum and search first. Product Managers usually follow only their own product’s forum, and your question has probably been asked before.
  • Keep feature requests short and human. One request per topic, the right terminology, what needs to change, and a brief use case. Long AI-written requests are hard to read and hard to search. Read a thread before adding your +1.
  • Errors go to Support, not the forums. Topics about technical issues are deleted. Post only after a case is closed if you have questions about the outcome. If you are unhappy with case handling, use the Talk to a Manager button (KB2320).
  • CVE questions go through Customer Support. Public topics risk spreading incomplete or outdated information. Veeam also runs a program on HackerOne.
  • Licensing and pricing belong with your Veeam sales rep. Product Management only enforces licensing rules in code; it does not set them.

They also explained what happens to a feature request. PMs discuss it to ensure they understand it, log it in an internal tracker that groups customers with the same need, track demand and effort, and prioritize approved items for a target release. So yes, someone is listening.

Three tips I am adopting: use search.veeam.com for full-text search across the KBs and documentation, and subscribe to the Weekly Forum Digest.

If a doc page is missing something, the “Submit Feedback” link on every User Guide page and KB article goes straight to the technical writers.

If I had to sum up these days in one idea, it would be that proof matters more than promises. A successful job is not a clean restore.

Immutability alone is not resilience. An AI agent with access to your data needs an undo button. Every Green session, from Zero Trust to SureBackup on any platform to iResOps, came back to proving that you can recover, and recover clean.

The other theme was the community itself. Two Veeam MVPs built OmniStage in their spare time, Product Management shipped features this group asked for, and the forums remain a direct line to the people who build the product.

A big thank you to the Veeam community team, every speaker, and the sponsors for an incredible time.

Useful links


Discover more from CloudnRoll

Subscribe to get the latest posts sent to your email.

You may also like...