PROVIDER - Virtual workload backup
Veeam
Veeam builds on Veeam Backup & Replication and covers virtual machines, physical servers, cloud-native workloads, and SaaS data from one console. Instant Recovery and immutable backups are its most cited capabilities, at the cost of per-workload licensing.
4.3pts
Detailed Notes
Pros 5 items
Cons 4 items
Service Overview
Veeam is a US-headquartered backup and data protection software vendor, and its core product, Veeam Backup & Replication, belongs to the "installed on the customer's own infrastructure" camp within Backup & Migration. Customers deploy the software, backup files, snapshots, and metadata stay on storage they control, and the vendor supplies licences and support rather than a managed service.
The product line reads as three threads. Virtualization backup covers VMware and Hyper-V and extends to physical servers and cloud hosts through agents. SaaS backup is its own product, Veeam Backup for Microsoft 365, covering mail, OneDrive, SharePoint, and Teams data. Monitoring sits with Veeam ONE, which tracks job success, capacity trend, and alerting. All three share one recovery engine, so the semantics of backup, verification, and restore stay consistent across them.
Two technical points get discussed most. The first is Instant Recovery: the backup file is mounted over NFS or iSCSI and the VM boots straight from it, so the business comes back before a full restore finishes and recovery drops from hours to minutes. The second is immutability: each backup file is locked for a retention window, and inside that window not even an administrator can delete or rewrite it, which targets the "encrypt the backups first, then demand ransom" path.
Key Strengths
- Four workload classes, one console: VMs, physical and cloud servers, cloud-native workloads, and SaaS apps sit in the same interface, so cross-environment restores need no tool switching or per-platform scripting
- Minute-level Instant Recovery: booting directly from the backup store, combined with verifiable restores through SureBackup, lets restore drills run weekly instead of quarterly and turns RTO into a measured figure rather than a promise
- Immutability against ransomware: locked files cannot be deleted or rewritten, and paired with offline media and an offsite copy it covers the most common dual failure of production encrypted and backups encrypted
- Broad storage compatibility: block storage, NAS, S3-compatible object storage, and dedicated appliances all work as targets, so existing storage assets stay in place with no vendor-specific hardware added to the rack
- Recovery verification and reporting: Veeam ONE reports success rate, capacity, and compliance, and SureBackup can boot verification VMs on a schedule, which gives auditors real restore records instead of configuration screenshots
Product Ecosystem
Platform capabilities
The core is Veeam Backup & Replication, which handles image-level backup, replication, Instant Recovery, and storage snapshot integration. Its job model rests on two objects: a job defines sources and scheduling, a repository defines where data lands, whether immutability is on, and whether synthetic fulls are produced. Synthetic fulls merge the incremental chain into a fresh full restore point, so a weekly full re-run is unnecessary and the backup window stays narrow.
Veeam ONE is the monitoring and reporting companion. It watches job state, repository capacity, VM backup coverage, and alerts. Operations teams tend to treat it as the referee for "is the backup actually running" rather than an optional add-on, because the most common failure in backup systems is not a failed job but a job that quietly stopped two weeks ago.
Veeam Backup for Microsoft 365 is licensed separately and covers the SaaS side: mail, calendar, OneDrive, SharePoint, and Teams. SaaS backup fails differently from virtualization backup. The former is about accidental deletion and retention expiry, the latter about host and storage failure, so its policies lean toward frequent incrementals with long retention.
Integrations
- Hypervisors: native vSphere and Hyper-V integration with storage-level snapshots (VSS and VSS for SMB) and changed block tracking
- Cloud and containers: AWS, Azure, and Google Cloud as backup targets, with Kubernetes workloads covered through Kasten
- Storage and media: block storage, NAS, S3-compatible object storage, and tape libraries for long-term archive
- Operations: a PowerShell module and REST API cover job control, status queries, and report export, so the platform plugs into existing monitoring and ticketing
- Alerting: SNMP, email, and Syslog outlets that aggregate into Zabbix or Prometheus-style alerting
Typical scenarios
Three patterns dominate: image-level backup and fast rollback for on-premises virtualization clusters; an offsite room or public cloud as the second copy location to satisfy 3-2-1; and long-term retention of SaaS mailboxes and documents for offboarding and audit. What all three share is a demand for restores the team can verify itself, not just a dashboard of successful jobs.
Limitations
- Licensing is per workload, so cost climbs linearly as virtualization grows and dense-VM environments need the unit economics modelled up front
- Backup & Replication, Microsoft 365 backup, and Veeam ONE are installed and licensed separately, which commonly produces either "backups run but nobody monitors them" or a missed SaaS deployment
- It leans on the hypervisor snapshot mechanism, and long or orphaned snapshot chains slow the backup window, requiring a separate snapshot governance process
- Teams in mainland China must supply storage and media themselves and lean on resellers for support, so complex incidents take longer to diagnose than a first-party response
Use Cases
- Organizations already running VMware or Hyper-V (★★★★★): image-level backup and Instant Recovery fit the virtualization management model, with restore granularity down to a single VM or up to a full-cluster drill
- Mid-size and large teams with explicit ransomware requirements (★★★★☆): immutable backups plus an offline copy form a complete line of defence, provided lock windows and retention are designed first, otherwise locked old copies fill the repository
- Microsoft 365 subscribers (★★★★☆): SaaS backup is licensed separately and suits organizations whose mail and documents must outlive the vendor's default retention, at the price of an additional per-user line item
- Hybrid cloud and container environments (★★★☆): on-premises and public cloud targets combine freely, though Kubernetes needs Kasten on top and configuration complexity rises
- Small teams needing only single-server file backup (★★): licensing and deployment cost outweigh the need, and lightweight tools or SaaS-only options are cheaper
Pricing
| Item | Price | Notes |
|---|---|---|
| Veeam Backup & Replication subscription | Annual per-workload subscription, tiered editions | A workload is one VM, one physical server, or one cloud instance; unit price varies by edition and channel, check the official quote |
| Perpetual licence | One-time purchase plus annual support | Version updates and support require ongoing renewal, so fold a 3-year-plus holding cost into the comparison |
| Veeam Backup for Microsoft 365 | Annual per-user subscription | Licensed separately from virtualization backup and billed per protected user |
| Veeam ONE | Annual per-workload subscription | Can be bundled with backup licences or bought alone |
| Community Edition | $0 | Feature-limited, suitable for evaluation and small trials, without formal support |
The table describes licence structure only and excludes storage media, bandwidth, and implementation services. Real budgets should include backup storage capacity: immutable copies and long retention multiply storage demand, and that line often costs more than the software licence.
FAQ
How much faster is Instant Recovery than a full restore?
The difference is "run first, move later". Instant Recovery mounts the backup file to the hypervisor and boots the VM, so the service returns, and a Storage vMotion to production storage follows on a schedule, which puts recovery in the minute range. A full restore writes the virtual disks back first, and time scales with data volume.
Can immutable backups stop ransomware on their own?
No. Immutability only guarantees that files cannot be deleted or rewritten inside the lock window. If an attacker reaches a period outside the lock, or production and backup share the same credentials, the path stays open. Pair immutability with offline media, a separate management network, and least-privilege accounts, and treat it as one layer of several.
What does Veeam offer over an entirely SaaS backup service?
Data and control. Veeam keeps backup files on customer storage, so restores do not depend on the vendor's service availability or subscription cycle. In exchange, the customer maintains storage, verifies restores, and handles capacity growth. A SaaS service absorbs all three but holds the data on its own side.
What should teams in mainland China plan for?
Media and support. Backup storage and object storage must be sourced separately, cross-border bandwidth affects offsite copy transfer time, and local support runs through authorized resellers, so confirm response targets and scope at purchase time.
Code Examples
Connect with PowerShell and start a job
# Connect to Veeam Backup & Replication; credentials come from environment variables
Import-Module Veeam.Backup.PowerShell -DisableNameChecking
$secure = ConvertTo-SecureString $env:VBR_PASSWORD -AsPlainText -Force
$cred = New-Object System.Management.Automation.PSCredential($env:VBR_USER, $secure)
Connect-VBRServer -Server $env:VBR_HOST -Credential $cred
# Start the named job and read back the latest session result
Get-VBRJob -Name "Daily-VM-Backup" | Start-VBRJob
Get-VBRBackupSession |
Where-Object { $_.JobName -eq "Daily-VM-Backup" } |
Select-Object JobName, Result, CreationTime -Last 1
Disconnect-VBRServer
A Success session only means the job raised no error. Restore readiness still needs a verification job such as SureBackup to boot a VM for real.
Query job state through the REST API
#!/usr/bin/env bash
set -euo pipefail
# Log in for a token; the account comes from environment variables, 9419 is the default REST port
TOKEN=$(curl -s -X POST "https://${VBR_HOST}:9419/api/v3/login" \
-H "Content-Type: application/x-www-form-urlencoded" \
--data-urlencode "grant_type=password" \
--data-urlencode "username=${VBR_USER}" \
--data-urlencode "password=${VBR_PASSWORD}" | jq -r '.access_token')
# Pull the 5 latest sessions and print name, result, and creation time for alerting
curl -s "https://${VBR_HOST}:9419/api/v3/sessions?limit=5&orderColumn=CreationTime&orderAsc=false" \
-H "Authorization: Bearer ${TOKEN}" \
| jq -r '.data[] | [.jobName, .result.result, .creationTime] | @tsv'
The REST service uses a self-signed certificate by default. Replace it with a trusted one in production instead of passing -k to skip validation.
Community & Resources
- Veeam official site — Product lines, edition comparison, and licensing
- Veeam Help Center — Full administrator guides and parameter reference per product
- Veeam Best Practice Guides — Official practice for repository design, retention, and immutability
- Veeam Community Forums — Troubleshooting, upgrade notes, and compatibility threads