VMware Live Site Recovery (SRM) after Broadcom: Site Recovery Manager’s new names and licensing

What Site Recovery Manager is called now and how it is sold

VMware Site Recovery Manager (SRM) has had two new names since Broadcom acquired VMware, but its licence is still sold under the old name. Broadcom’s 9.0 release notes (March 2024) say that “VMware Site Recovery Manager is rebranded to VMware Live Site Recovery”, as part of VMware Live Recovery. The VMware Live Recovery FAQ of January 2026 describes that product as Live Site Recovery plus VMware Live Cyber Recovery, formerly VMware Cloud Disaster Recovery. Broadcom’s 9.1 release notes, dated 12 May 2026, add that “VMware Live Recovery has been renamed and integrated into VMware Cloud Foundation (VCF) as Protection and Recovery”, one appliance that also contains vSphere Replication and vSAN snapshots and replication. As of October 2026, Broadcom sells the recovery orchestration as a standalone VMware Site Recovery Manager licence per protected VM, for VCF, vSphere Foundation (VVF) and older vSphere editions, or within VMware Advanced Cyber Compliance (ACC), a per-core add-on for VCF only. vSphere Replication has no licence of its own and comes with VCF, VVF, Enterprise Plus and Standard, within the limits below.

From SRM to Protection and Recovery: versions and appliances

Broadcom: VMware Live Site Recovery 9.0 and Protection and Recovery 9.1 release notes; 9.1 appliance page (10 September 2026); KB 408127, 414400, 426655 and 444864 (no date shown, read on 6 October 2026).

KB 444864 says that upgrading to vCenter 9.1 “requires transitioning disaster recovery environments to the unified VCF Protection and Recovery 9.1 appliance”. The separate 9.0.2 appliances cannot be upgraded to it in place, while 9.0.3 and later can. From vSphere Replication 9.0.3, enhanced replication is the only supported type, and the move to 9.1 “requires that all replications are reconfigured to enhanced replication”. During a staged upgrade, a 9.1 site pairs with vSphere Replication 9.0.5 or later at the other site, Broadcom’s N-1 rule. KB 390634 puts vSphere Replication and SRM before vCenter in the VCF 9.0 update order, as our guide to the end of vSphere 8 general support shows. Broadcom’s Product Interoperability Matrix lists every version, SRM included, under Protection and Recovery (KB 447296), so check your vCenter build there.

How SRM and Advanced Cyber Compliance are licensed

Broadcom’s VMware Site Recovery Manager programme document of May 2026 licenses the software “based on the number of Protected Virtual Machines”, powered on or off. Per the 9.1 licensing page (10 September 2026), the same licence is installed at both sites, so “one-way protection of 100 VMs from site A to site B, requires a 100 VMs license”. Protecting 100 VMs in each direction takes a 200-VM licence. The page offers the standalone licence for VVF and “pre-version 9 vSphere software”, and Broadcom’s VCF 9.1 FAQ of 3 September 2026 lists SRM as a VCF add-on too.

ACC, one of nine advanced services sold on top of VCF, is licensed per core with “a minimum licensing requirement of 16 Cores per Processor”, only as an add-on to VCF or VMware Telco Cloud Platform, per its programme document of May 2026. For its disaster and cyber recovery features, every core managed by a vCenter instance with protected VMs needs a licence, and with several vCenter instances, only those where VMs are protected are licensed (FAQ Q93). Besides on-premises DR orchestration, ACC includes cyber recovery workflows; a cloud target needs VMware Live Recovery Cloud “along with the associated cloud hosts” (FAQ Q94).

Broadcom: VCF 9.1 FAQ (3 September 2026), question numbers as shown; where Broadcom’s documents name different entitlements, the table gives both, so confirm yours in writing; TechDocs 9.1 licensing and replication setup pages (10 September 2026); SRM and ACC programme documents (May 2026).

Count protected VMs per direction for SRM and cores per vCenter for ACC and, on VCF, ask for both options in writing. What else VCF and VVF include is in our VVF vs VCF comparison, the tier and term in our guide to the VMware renewal tier.

Our VMware optimisation service includes an estate and licence audit of versions, subscriptions and actual resource usage, and matches editions to your workloads within Broadcom’s current licensing logic. Send us the protected VM count and the cores per vCenter through the form below.

vSphere Replication 9 and vSAN: RPO, retention and enhanced replication

Broadcom describes vSphere Replication as “a hypervisor-based virtual machine replication and recovery”. After an initial full synchronisation, the vSphere Replication agent “sends changed blocks in the virtual machine disks from the source site to the target site”. In the 9.1 procedure the RPO can be set from 1 minute to 24 hours, and vSphere Replication “does not adhere to a strict replication schedule”. Below 5 minutes, the same procedure requires a VMware Live Recovery, ACC or Site Recovery Manager subscription, while the VCF 9.1 FAQ says that enhanced vSphere Replication “with RPOs as low as one minute requires ACC entitlement” (Q81).

Point-in-time instances keep older recovery points, up to 24 per VM, or up to 200 with the Latest instances option, which “requires an Advanced Cyber Compliance (ACC) subscription”. At recovery they become snapshots of the recovered VM, which helps when the newest copy already contains corrupted data. To keep six instances a day, the RPO must not exceed four hours.

Enhanced replication sends traffic “directly to the host at the target site”, bypassing the appliance. The source ESX hosts therefore need TCP port 32032 open to the hosts of the target cluster, and the management server needs direct access to the management networks of all participating hosts. The 9.1 procedure names vCenter and ESX 8.0 Update 2 on an on-premises target as the minimum for enhanced replication, but 9.1 itself is compatible with vSphere 8.0 Update 3 and later, which sets the minimum at both sites. Enhanced replication in 9.1 also requires network encryption. A newly deployed appliance “can manage a maximum of 5000 replications”. At a 5-minute RPO, Broadcom’s figure is 500 VMs, depending on bandwidth and change rate per disk.

In 9.1, vSAN snapshots and replication copies VMs from vSAN ESA, VMFS or NFS clusters on vSphere 8.0 Update 3 or later to a cluster running vSAN ESA 9.0 or later. Local snapshots are in the VCF core entitlement, while replication needs “an additional entitlement”. The VCF 9.1 FAQ names ACC or standalone SRM for any replication to a vSAN ESA datastore “irrespective of the RPO” (Q36), and ACC for vSAN ESA snapshot replication to a remote site (Q81).

Without SRM or ACC, recovery is manual, since in Broadcom’s 9.1 procedure you recover “one virtual machine at a time”, with its network devices disconnected. Recovery plans start an application in a set order with new IP addresses at the recovery site.

Array-based replication through SRAs

With array-based replication, the arrays replicate the datastores and Protection and Recovery orchestrates the recovery through a storage replication adapter (SRA). Broadcom’s 9.1 documentation says SRAs “are not part of a protection and recovery release”; the array vendor develops and supports them. One SRA per array type is installed on the Protection and Recovery Server host, and the RPO “depends on the Storage Replication Adapters (SRAs) provided by the vendors”. Check the Site Recovery Manager Storage Partner Compatibility Guide for your SRA on 9.1 before the upgrade.

A VM is protected only if its replication is configured correctly, “even if the virtual machines reside on a protected datastore”. Storage DRS and Storage vMotion work with protected VMs only on arrays with consistency groups. Release 9.1 does not support NFSv4.1 datastores for array-based replication and is “the last release that supports vSphere Virtual Volumes”; Broadcom recommends moving vVols replications to array-based replication.

Recovery plans, test recovery, reprotect and failback

In Broadcom’s 9.1 documentation, a protection group is “a collection of virtual machines that protection and recovery protects together”. A recovery plan “is like an automated run book” that sets the order in which VMs power on and off and the network addresses they use after recovery. A plan goes through these operations in order:

1. Test, which starts the VMs on a test network at the recovery site; vSphere Replication keeps synchronising through redo logs, and array-based replication tests on an array snapshot.
2. Cleanup, which removes the redo logs or snapshots; Broadcom requires it “before running a failover or another test”.
3. Planned migration, or disaster recovery after “an unforeseen event that might result in data loss”.
4. Reprotect, once the recovery finishes without errors and the original site is back, which reverses replication and reconfigures protection groups and recovery plans for the opposite direction; after a disaster recovery, Broadcom’s 9.1 documentation first completes it with a planned migration.
5. Failback, a planned migration back to the original site after the reprotect, then a second reprotect so that replication runs in the original direction again (KB 417721).

A test “has no lasting effects” on either site and does not disrupt replication, except that plans which suspend local virtual machines do so in tests too. Our guide to failover and failback steps covers the day itself, from the decision to DNS and user access.

Live Site Recovery, Veeam replication or DRaaS compared

Broadcom TechDocs Protection and Recovery 9.1 (September 2026) and VCF 9.1 FAQ (3 September 2026); Veeam Backup & Replication 13 User Guide and Veeam Cloud Connect Guide (read on 6 October 2026); the Fits column is our reading.

Both approaches recover VMs from replicas in a set order, Broadcom with recovery plans, Veeam with failover plans in which “you define the order in which VMs must be processed”. With Veeam Cloud Connect, a service provider offers the recovery site as DRaaS, with full and partial site failover for its tenants. Which Veeam job protects against what is in our comparison of Veeam replication and backup copy.

Our disaster recovery service runs your recovery site in Baltneta’s Tier-3 data centres in Lithuania. Tell us which systems you would protect first and how they replicate today.

What we do

We hold the contract, with engineering by our partner Vixen.UNO. Under VMware optimisation, Vixen.UNO audits the estate, matches editions and subscriptions to your workloads and carries out changes, site recovery included, in agreed maintenance windows with a rollback plan at every stage. Our disaster recovery service runs a recovery site in Baltneta’s Tier-3 data centres in Lithuania (ISO 27001, PCI DSS), with Veeam Cloud Connect replication over an encrypted TLS/SSL channel from a 15-minute interval and scheduled failover tests in an isolated environment, each with a report. The service’s standard targets are RPO from 15 minutes and RTO of 1 to 2 hours for critical systems, and your targets are defined per system tier and fixed in the SLA. The first call is free of charge; the price of the technical assessment is fixed before work begins.

FAQ