Lift and shift VMware on-premises to IBM Cloud®

End of Marketing: As of 31 October 2025, new deployments of VMware Solutions offerings are no longer available for new customers. Existing customers can still use and expand their active VMware® workloads on IBM Cloud®. For more information, see End of Marketing for VMware on IBM Cloud.

Over the past two decades, many enterprises have developed the VMware Software-Defined Data Center (SDDC) as the dominant virtualization platform for hosting x86 production and business critical workloads in on-premises or colocated data center facilities. VMware has a 44.8% market share in x86 virtualization platform technologies, per Statista.

In parallel with maintaining the production platform on-premises, and as a society outcome of the global pandemic, businesses are requesting an acceleration of the digital transformation to a cloud native enterprise platform - in effect, a recognition that the compute platform became the business, with hosted digital products generating the revenue. For the business, the digital compute platform needs to be trusted, highly available, secure, scalable, with agile integration to the digital products user experience and feature set evolution.

This white paper describes the business requirements, solution considerations, and migration approaches to move the on-premises x86 VMware virtualization compute, storage, and network to IBM Cloud for VMware Solutions. The production workload environments (PROD, UAT, SIT, DEV) are moved to IBM Cloud for VMware Solutions by using the lift and shift method with no change to the business applications.

On-premises business challenges

Business challenges are leading clients to reevaluate the x86 VMware virtualization on-premises or colocated platform:

Licensing cost increases for VMware virtualization: Businesses are looking to minimize the impact of VMware Cloud Foundation (VCF) bundled pricing changes effective 1 May 2024 and mitigate potential future VCF subscription cost increases. As part of their acquisition of VMware, Broadcom introduced a new unified subscription based licensing model for VMware software. This new model that is resulted in businesses needing to change their then perpetual VMware software licenses into two bundles: VMware Cloud Foundation or VMware vSphere Foundation Essential.
High operational costs to maintain an on-premises or colocation data center facility: Significant operating expenditure OpEx is incurred to maintain compliance and security of the VMware infrastructure resources across the x86 bare metal servers, storage, and network devices. OpEx costs include:
- Managing costly regulatory, industry certifications and audits.
- Time consuming server and VMware Cloud Foundation component security patching updates.
High capital expenditure CapEx costs to grow or adapt the on-premises facility to adapt to business requirements include:
- Slow, multi-week, new VMware ESXi server provisioning for new capacity requirements.
- Slow, multi-week, increase in VMware ESXi server memory RAM or additional storage request execution.
- Shorter servers refresh cycles to avoid costly extended warranties.
Cybersecurity: Businesses are faced with growing cybersecurity threats and data breaches requiring their CISO to adopt the National Institute of Standards and Technology NIST cybersecurity best practices and apply adequate preventive and recovery security measures to their systems. There are many challenges in deploying the zero trust security model across hybrid cloud.
Deliver improved platform availability, resiliency, security, performance, and scaling business requirements due to the accelerated shift to a digital economy observed post pandemic. The business compute platform that once needed to be available 8 hours per day, 5 days per week, now are required to be available 24x7 with overall platform availability of 99.99%+. As a cloud provider, IBM can provide higher availability, security, performance, at lower costs than businesses can typically do on-premises.
Slow innovation and deployment of new modernized cloud native applications or services: Initial application modernization to the cloud native operating system Linux, and the container platform Kubernetes, is successful for web internet facing applications but has struggled particularly with the modernization of Windows.Net Framework applications, leading to delayed application modernization to cloud native development programs.
Delayed implementation of a data center exit business mandate.
Maintaining VMware automation assets while developing Infrastructure as Code assets across the hybrid cloud platform.

IBM Cloud for VMware Solutions business benefits

Moving business critical VMware workload environments to IBM Cloud for VMware Solutions enables the following benefits:

Enterprise-grade cloud infrastructure: Uses IBM Cloud robust and secure infrastructure that is designed to meet the stringent requirements of business critical workloads.
Flexibility: Access to a wide variety of hardware profiles including the latest Intel Xeon processors that are 4th generation, access to VMware compute accelerators including smartNics, enabling higher virtual instance densification, and higher CPU allocation to virtual servers. Scaling VMWare compute on IBM VMware offerings include the capability for new capacity expansion within the hour.
Opportunity to move VMware managed services and VMWare automation maintenance to IBM Cloud® for VMware Cloud Foundation as a Service. A disaster recovery as a service (DRaaS) is an optional add-on.
Advanced Cloud Services: Access to latest GPUs such as Nvidia and Intel for AI client solutions along with IBM watsonx services gen AI, inferencing, and machine learning to fast-track innovation into business processes. Additional performance benefits by using latest Intel Sapphire Rapids accelerator features are possible for vSAN storage, which is configured with Express Storage Architecture ESA, using the CPU offloading capability.
Optimize x86 workloads: In moving to IBM Cloud, there is further opportunity to optimize the x86 workload to alternative IBM Cloud enabled virtualization platforms such as KVM, Xen, PowerVM, or LinuxONE on Z platform.
Cost savings: Avoid unnecessary expenses associated with repurchasing licenses for new infrastructure when adopting the new Broadcom and IBM Cloud for VMware Solutions subscription. IBM Cloud for VMware Solutions allows businesses to address Broadcom Licensing immediately and generate a better near-term Return on Investment (ROI), which can be redirected toward future rehosting, replatforming, or application modernization to microservices initiatives.

Generating Return on Investment with IBM Cloud VMware Solutions

IBM Cloud is the first VMware partner to offer an automated and full featured VMware Cloud Foundation as a Service uniquely providing complete entitlement and control. Enterprises can take advantage of thousands of customizable options, built-in security and controls, bring familiar best-of-breed tools from VMWare ecosystem, and get up to 201% return on investment over three years with IBM Cloud. The ROI is based on the Forrester Total Impact study that is commissioned by IBM.

Architecture overview

The solution architecture proposes to retain the VMware platform, by using seamless migration standard tools such as VMware HCX to lift and shift the hosted business workloads into IBM Cloud for VMware Solutions.

Lift and shift workloads — Lift-and-Shift VMware workloads to IBM Cloud VMware Solutions architecture overview

Compute solutioning

Business requirements for compute solutioning

Provide a low risk migration to cloud of business critical on-premises VMware virtualized x86 application and management environments.
Provide a migration capability to move environments by using a lift and shift method from on-premises to cloud without requiring application changes.
Fast track data center exit to reduce costs.
Reduce the impact of VMWare licensing changes by properly sizing virtual machine cores to business usage.
Provide high performance x86 CPU and network bandwidth for large enterprise applications such as SAP and databases.
For business critical workloads on VMWare, deliver 99.99% availability and RPO near zero.
Optimize VMware operational and capital expenditure costs.

Compute solutioning considerations

The following table summarizes the compute design considerations when solutioning lift and shift VMware on-premises workloads to IBM Cloud for VMware Solutions.

Table 1: Compute considerations move VMware to IBM Cloud
Compute areas	Description
Retain VMware virtualization	Retaining VMware as the x86 virtualization platform allows for the same VMware components and operational skills and automation assets to be reused in IBM Cloud. No application changes are needed when moving to IBM Cloud for VMware Solutions by using the lift and shift migration method for VMware clusters. Other options include: Rehosting to an alternative hypervisor such as KVM or Microsoft Hyper-V, which can require application changes that are related to the infrastructure configuration such as IP addresses. Also, automation assets for the virtualized platform stack such as security patching for the hypervisor layer must be developed. Replatform the VMWare x86 virtual servers to KubeVirt containers coupling the move to cloud with the application modernization program. This introduces the risk of virtual instances remaining which can delay data center exit.
Choosing IBM Cloud for VMware Solutions offering	The choice of the IBM Cloud for VMware Solutions is based on the following criteria: The level of flexibility the customer requires in the VMWare Cloud Foundation landing zone architecture and deployment automation. The level of transfer VMware Cloud Foundation skills and automation to IBM Cloud for the Day 2 management. The level of network performance and bandwidth that is required with the VMware Cloud Foundation on VPC offering the highest level of performance and integration with IBM Cloud native Virtual Private Cloud.
Sizing the VMware Cloud Foundation license cores	It is important to properly size the new server environment to minimize the cost of VMware licensing changes. Start with an RVTools report from the current on-premises VMware server environment. On-premises infrastructure is often sized to handle peak workloads. The report shows the number of vCPU, RAM, and storage assigned to each virtual machine. For RAM and storage, both the amount that is provisioned for the virtual machine and the amount in use when the report was run is displayed. The amount that's needed in the cloud should be closer to the in-use numbers. Consider the model numbers of the VMware ESXi servers on-premises and note that the server models in IBM Cloud likely are newer, with faster processors and more cores. Typically, fewer vCPU are needed in the new environment, which lowers the VMware licensing cost. Cloud compute, memory, and storage can be scaled rapidly so sizing can be aggressive initially in the cloud landing zone and scaled up as needed. Scaling or adding additional bare metal servers by using the VMware Cloud Foundation for VPC or IBM Cloud for VMware Cloud Foundation as a Service offerings can be done within an hour. Many VMware environments, such as development and test workloads, don't need to run all day. It is recommended to shift nonproduction to hourly consumption to IBM Cloud VMware Cloud Foundation as a Service for 20-30% price decrease in assets and licenses. Many businesses are conservative when establishing a ratio of CPU to vCPU with ratios like 2:1 or 4:1 being common. With the more powerful Intel Sapphire Rapids, Generation 4, processors available on IBM Cloud, these ratios can be increased dramatically to 6:1 or even 12:1 sometimes.
Optimize VMware operational costs	IBM Cloud for VMware Solutions include several automation features that can help lower your operational costs. In the VMware Cloud Foundation as a Service offering, IBM handles the configuration, capacity management, monitoring, patching, upgrades, and security of the underlying VMware infrastructure, through a highly available management plane. This can greatly reduce labor costs.
Additional tooling to right size on-premises x86	Many business critical x86 environments, which have an ecosystem of platforms including VMware, Bare Metal, and custom appliances can benefit from a Total Cost of Ownership (TCO) or FinOps review. One IBM tool that can be used to right-size your VMware environment is Turbonomic. Turbonomic employs a dynamic approach to rightsizing resources, assuring optimal performance while minimizing cost. Turbonomic can size your virtual machines’ vCPU and vMemory to match observed demand and peaks and can automate rightsizing in real time to meet demand. IBM also offers a Total Cost of Ownership (TCO) Analysis for businesses. This analysis uses FinOps software from Akasia to compare the costs of running VMWare workloads on-premises and major cloud providers including IBM.
Migration tooling	The virtual machine migration method depends on the VMware on IBM Cloud flavor, the size of the environment and the functions needed. When migrating to VMware Cloud Foundation as a Service on IBM Cloud, the recommended method is to use VMware Cloud Director Availability (VCDA). VCDA enables migration from traditional on-premises VMware environments or from other VMware Cloud Foundation as a Service on IBM Cloud VMware Cloud Director environments. For all other VMware Cloud Foundation on IBM Cloud flavors, the main options are VMware HCX, Zerto, Veeam, and PrimaryIO. VMware HCX allows the preservation of the migrated virtual machines IP addresses, create easily reversible migration waves, and extend the network from on-premises to the cloud, which allows for progressive migrations. Zerto and Veeam are more traditional replication solutions and simpler to implement. PrimaryIO provides a DRaaS service that can also be used to migrate VMware workloads from on-premises to VCF on classic or on VPC in IBM Cloud.

IBM Cloud VMware Solutions offering comparison chart

The following table describes the five flavors of IBM VMware Solutions either as customer self-managed VMware Cloud Foundation or IBM managed VMware Cloud Foundation as a Service.

IBM Cloud VMware Solutions decision tree

The decision tree proposes a prescriptive choice of IBM Cloud for VMware Solutions based on the environment.

Storage considerations

The following table summarizes the storage design requirements and considerations when solutioning lift-and-shift VMware on-premises workloads to IBM Cloud VMware Solutions.

Storage requirements

Replace the on-premises VMware storage solution.
Optimize VMware storage operational expense OpEx and capital expenditure CapEx costs.
For business critical workloads, deliver 99.99% availability and RPO near zero.
Provide a replacement for on-premises tape backup storage.
Provide a secure automated backup service.

Storage design considerations

Storage considerations move VMware to IBM Cloud
Storage areas	Description
IBM Cloud storage options	The options for storage for VMware Solutions on IBM Cloud are IBM Cloud File storage (based on NFS, Network File Storage) VMware vSAN, or Block storage(iSCSI). VMware vSAN is based on a hyperconverged architecture where the storage of every compute node is aggregated into a single virtual storage pool. It supports advanced capabilities such as deduplication, encryption in transit, configurable resiliency level, and can be replicated across availability zones. vSAN software and 1 TiB per compute core of vSAN storage are included with every VCF license. Scaling up a vSAN cluster requires adding extra compute nodes. In the case of VCF on VPC, vSAN is the only available storage option. IBM Cloud File Storage (NFS) and Block storage (iSCSI) offer scalable storage without having to scale the compute nodes. However, they do not have the same resiliency options, and cannot be stretched across availability zones. Block storage (iSCSI) is only available for the VCF for Classic on IBM Cloud offerings and requires additional manual configuration. For auxiliary storage (not for the vSphere data store), IBM Cloud Object Storage can be used. It is best suited for unstructured data and long-term or archival data storage.
Backup and disaster recovery	The available backup and disaster recovery methods depends on the VCF on IBM Cloud flavor. Regarding disaster recovery, for VCF as a Service on IBM Cloud only, VMWare Cloud Director Availability (VCDA) is available and automatically provisioned with every VCF as a Service on IBM Cloud instance. It allows to create a disaster recovery configuration between two VCF as a service instances or between a traditional on-premises VMware environment or another type of VCF on IBM Cloud instance and a VCF as a service instance. VMware Cloud Director Availability failover capabilities are bidirectional, allowing to failover from the primary site to the recovery site and subsequently to fall back from the recovery site to the primary site. VMware Cloud Director Availability is the only supported disaster recovery solution for VMware as a Service as it is not possible to directly access the vCenter server. For more information, see VCF as a Service overview. For other VCF on IBM Cloud flavors, other disaster recovery solutions are available. Zerto, Veeam, and ProtectIO are the main options. Zerto and Veeam are self-managed solutions integrating directly with vCenter server to offer replication capabilities with configurable recovery points(up to a near 0 RPO with continuous replication). In the case of VCF for Classic - Automated on IBM Cloud, Veeam, and Zerto can be automatically deployed from the IBM CloudVMware Solutions portal. The configuration and operation of these solutions remains the responsibility of the customer. For more information, see Veeam on VCF on IBM Cloud and Zerto for VCF on IBM Cloud. ProtectIO is a Disaster Recovery as a Service (DRaaS) offering from PrimaryIO that runs natively on IBM Cloud. ProtectIO also enables near zero Recovery Point Object (RPO) through continuous replication. Its agentless protection technique has no impact on the performance of the primary site. For more information, see ProtectIO. Regarding backup, Veeam Backup and Replication are available for all VCF on IBM Cloud offerings. It is preinstalled with VCF as a Service on IBM Cloud instances and can be automatically deployed with VCF for Classic - Automated, with different deployment types available such as VMware VM, VSI, and Bare Metal Server. However, alternative 3rd party backup solutions can also be used in the case of VCF as a Service on IBM Cloud, third-party backup solutions must be agent-based as it is not possible to directly access the vCenter server.
Optimize VMware storage operational and capital costs	When moving to IBM Cloud VMware Solutions operational expenses cost reductions can be achieved with vSAN through replacing or reducing manual tasks such as storage deployment, compliance configuration, scaling, and capacity expansion with storage policies defined within VCF. When using File storage or Block storage from IBM Cloud storage administration and operations are by IBM Cloud.

Network considerations

The following table summarizes the network design requirements and considerations when solutioning lift-and-shift VMware on-premises workloads to IBM Cloud VMware Solutions.

Network requirements

Provide secure access to web applications from internet.
For Business critical workloads, deliver 99.99% Availability and RPO near zero.
Provide secure connectivity between customer remaining on-premises and VMware Cloud Foundation on IBM Cloud.
Provide global load balancer capability for Web applications.
For regulated and business critical workloads, provide DMZ.
Optional: Provide integration with on-premises Microsoft Active Directory.

Design considerations

Table 3: Network considerations move VMware to IBM Cloud
Network areas	Description
Choosing the IBM Cloud region	Decide whether the VCF instance will be deployed in a multizone region (MZR) or in a single zone region (SZR). Using an MZR enables to achieve a higher level of availability and resiliency through the deployment of at least two VCF instances in two different availability zones. Note that if vSAN is used, cross-availability zone replication is possible but requires a witness running in a third availability zone. The vSan witness can consist of a vSan witness appliance, which can be run on any VCF on IBM Cloud host including on a VMware Cloud Foundation as a Service multitenant instance.
Provide secure connectivity between a customer remaining on-premises and VMware Cloud Foundation on IBM Cloud	To securely connect from the remaining on-premises environment to the IBM Cloud environment, IBM Cloud Direct Link or VPN connectivity can be used. IBM Cloud Direct Link is a high capacity private connection while VPN connectivity consists in an encrypted site-to-site or client-to-site tunnel over the public internet. On the IBM Cloud side, the termination point for both options depends on the level of isolation that is required inside the IBM Cloud environment and on the VCF deployment flavor. For more information, see IBM Cloud Direct Link, IBM Cloud VPNaaS and IBM Cloud Classic VPN options.
Provide secure and available access to web applications from the Internet	To secure or filter traffic entering the environment, capabilities such as DDoS mitigation or web application firewall (WAF) should be considered. IBM Cloud Internet Services (CIS) includes these capabilities as well as the capability to act as a global load balancer (GLB), allowing to increase the availability and resiliency of the workloads running on VCF on IBM Cloud, if at least two different instances are deployed in at least two different availability zones. Alternatively, third party GLBs can also be used. For more information, see IBM CIS.
Firewall	When migrating from on-premises to VCF on IBM Cloud, the firewall constructs available within the VMware environment remain unchanged, however particular attention should be given to the firewalling at the edge of the VMware environment. The recommended approach is to use a similar type of firewall appliance in IBM Cloud as the one used on-premises to limit the operational changes, use the skills of existing operations team, and minimize the risk of disruption. Another point to consider is the firewall availability and resiliency requirements, especially if IBM Cloud VPC as the high availability modes that are supported will vary based on the firewall appliance type. Several firewall virtual or physical options are available for IBM Cloud VPC: Juniper vSRX, Fortinet vFSA, F5 Big-IP VE, Palo Alto VM-Series. Several firewall virtual or physical options are available for IBM Cloud Classic: Juniper vSRX, Fortinet FSA and vFSA, Vyatta Virtual Router Appliance. It is also possible to bring your own gateway as a virtual appliance on top of a dedicated IBM Cloud Classic bare metal or as an IBM Cloud VPC VSI deployed from a custom image.
Interconnectivity within IBM Cloud	To reach the various IBM Cloud from the VCF on IBM Cloud environment, Virtual Private Endpoints (VPE), or Cloud Service Endpoints (CSE) can be used. VPEs are available for IBM Cloud VPC VCF instances while CSEs are available for IBM Cloud Classic VCF instances. Both VPEs and CSEs allow to privately reach the IBM Cloud services, keeping the traffic within IBM Cloud and avoiding any egress traffic charges. For more information, see IBM Cloud VPE and IBM Cloud CSE. Integration of the different IBM Cloud IaaS, such as VPC, classic, and PowerVS, is achieved by using the IBM Cloud Transit Gateway, which is an as a service BGP (Border Gateway Protocol) router that can also be used as the termination point for direct-link connectivity. IBM Cloud transit gateway can be local, limited to a specific region or global, spanning multiple regions. For more information on IBM Cloud Transit Gateway, see: IBM Cloud Transit Gateway

Security considerations

The following table summarizes the security design requirements and considerations when solutioning IBM Cloud VMware Solutions for lift and shift VMware on-premises Workload into IBM Cloud.

Security requirements

Adopt a zero trust security strategy
Provide single sign-on for Enterprise administrators to access IBM Cloud account and resources by using the enterprise identity and password.
Provide a relevant industry compliant IBM Cloud VMware solution platform
Provide a Key Management System for data encryption
Provide for client encryption key retention solution.
Provide privileged access management
Optional: Provide single sign-on with Integration of on-premises LDAP service

Security considerations

Table 4: Security considerations move VMware to IBM Cloud
Security areas	Description
Adopting zero trust	Zero trust is a security strategy for cloud and hybrid cloud networks. It replaces the traditional security approach of network perimeter protection by using firewalls and other security controls. With zero trust the operating principle is “never trust, always verify” moving the focus away from the network perimeter placing security around individual resources. Follow the principle of duty segmentation by using Role Based Access Control (RBAC). Segmentation of duty helps ensure network, compute, and storage administrators use separately defined RBACs. Use Access Groups to map the enterprise LDAP VMware administrators such as NSX or Cluster Admins to Access Groups. Developers do not have access to production. Implement the principle of least privilege by assigning only required access permissions. A time-based conditional access can be granted to limit access to the time frame you specify reducing the opportunity for attack if a security breach occurs. Access to and changes to resources that include applications, compute, network, storage are secured, audited, and verified. For more information, see Best Practice for organizing resources and assigning access Enhance security by applying Context-based restrictions Protecting Virtual Private Cloud with Context-based restrictions Trusted profiles as foundation for secure cloud
Single sign-on	For over the past two decades, LDAP Directory Services have been the standard for the enterprise user identity provider (IdP). Review the ability to federate your corporate identity provider (IdP) with IBM Cloud IAM to allow your enterprise users access to IBM Cloud with the company username and password SSO. Enterprise Federation (EF) is the authentication model that allows an enterprise’s IdP (Identity Provider) to authenticate users by using their enterprise credentials to IBM Cloud using IBMid. Once LDAP directory service federation is established between IBMid and the customer LDAP directory (or IdP) users from (SAML-based) LDAP attributes are evaluated at login and if they meet all of the conditions for a trusted profile, they are prompted to apply one or more trusted profiles. Trusted profiles grant users the level of access that they need to complete specialized and specific tasks in a limited time-period, for example, 1 - 4 hours. Time-based access allows frequent authentication checks for reduced security risks. For more information, see IBMid Enterprise Federation Introduction How IBM Cloud IAM works Using Trusted profiles for IAM to IBM Cloud Enabling authentication from an external identity provider
Compliance	Regulatory compliance: Help ensure the IBM Cloud for VMware Solutions adheres to relevant industry regulations and compliance standards, such as GDPR, HIPAA, or any other applicable requirements. For regulated industries implementing a continuous security and compliance posture on IBM Cloud platform can be done by using Security and Compliance Center. Secure boot helps ensure VMware ESXi servers start with trusted software by verifying the signatures for all code in the boot process. For more information, see IBM Cloud Compliance Programs Understanding Compliance in IBM Cloud Getting started with Security and Compliance Center Implementing secure boot on VPC bare metal.
Data encryption	Protecting data against unauthorized disclosure, modification or destruction throughout the data lifecycle is of paramount importance when moving VMware on-premises production critical workloads to IBM Cloud. VMware provides two areas to consider encryption: storage or vSAN encryption and Virtual Instances (or vSphere encryption) by using KMIP (Key Management integration Protocol) for key management system (KMS) integration Customers can use IBM Key Protect for IBM Cloud keys to create key management interoperability protocol (KMIP) adapters for use with their VMware Solutions for vSAN and vSphere encryption. For more information, see Using Key Protect with KMIP. For environments or systems having a high security categorization, or regulated environments customers can use IBM Hyper Protect Crypto Services to manage encryption keys. Hyper Protect Crypto Services is a single tenant key management service and dedicated hardware security module (HSM) partition. Hyper Protect Crypto Services is the only service in the cloud industry that's built on FIPS 140-2 Level 4-certified hardware. For more information, see Configuring Hyper Protect Crypto Services for VMware Solutions.
Privileged access management	Administrative access to the IBM Cloud for VMware Solutions is restricted, audited, and logged through a bastion host solution.

Migration program

The move of a business critical x86 VMware production platform to IBM Cloud is supported by the migration program, which we propose to divide into 5 phases: Discovery, Assessment, Minimum Viable Product (MVP), Migration, and Operate.

Move to IBM Cloud Phased Approach — VMware move to IBM Cloud phased migration approach

Discovery - Phase 1: Data is collected across technical, sales, and business teams to understand the motivation and internal factors driving the move to cloud. Identify priorities such as scalability, solving functional issues, lack of in-house skills, reducing infrastructure, and IT expenses. Identify and document current infrastructure, applications, relationships, and dependencies including drift from the expected business required state. The key to data collection for the VMware on-premises or colocated platform is to an extraction tool such as RVtools.
Assessment - Phase 2: Data that is collected in the discovery phase is used to assess the migration to cloud viability to and choosing the right target platform solution. Prioritize the lift and shift migration of the VMWare SDDC to IBM Cloud for VMware Cloud Foundation so that current VMware ISV support, VMWare processes, and automation are maintained. Avoid adding repackaging (applications) or refactor to microservices patterns to decouple the modernize from move to IBM Cloud program. Assess the supply-chain of workload validating the utilization and usage of VMware components. Right-size the target VCF landing zones consolidating VMware clusters using higher core density servers. Assess the software supply chain and data for the required encryption level and business key integration. The outcome of the assessment phase is a high-level architecture of the cloud environments with an associated bill of materials.
Minimum Viable Product (MVP) - Phase 3: Provides the confidence in the viability of the proposed solution. It is important to clearly define and articulate upfront the success criteria of the MVP as well as to identify the approving stakeholders. The successful MVP builds trust and confidence in the solution and provides the go ahead for the migration of the production workload. The MVP familiarizes the operating team with the cloud platform and guides toward a smooth adoption.
Migration - Phase 4: Execution of workload environment migration. The duration of the migration to IBM Cloud phase depends on the complexity and criticality of the production environment. IBM Technology Expert Labs includes accelerator services to migrate workloads to IBM Cloud.
Operate (run) - Phase 5: Refers to tasks and processes to maintain the workload operational after the workload has been migrated to the target IBM Cloud VMware platform. Key processes include backup, observability, and maintaining compliance.

Optimize x86 on-premises into IBM Cloud

Typically the on-premises VMWare x86 virtualization platform is associated to an ecosystem of compute platforms such as bare metal x86 Windows or Linux servers hosting stand-alone or clustered applications. The move to cloud assessment can bring out additional business requirements that are linked to the full x86 ecosystem and proposed as follows:

Optimize x86 business requirements

Provide Cloud native virtual instances deployable through Continuous Integration Continuous Delivery (CI/CD) and declarative language scripts such as Terraform.
Enable AI inferencing in production business processes. Provide encoder decoder capabilities for developing gen AI business process innovation.
Provide a cost efficient, flexible, and scalable, mission and business critical Kubernetes platform for modernized container cloud native applications.
Provide an alternative CPU platform to x86 for large enterprise workloads such as SAP or Oracle.

Compute considerations

Table 5: Optimize x86 on IBM Cloud
Compute areas	Description
Replatform x86 virtual machines to IBM cloud virtual server instances	Consider the level of performance in new virtual instances in IBM Cloud when choosing the workload to replatform. Required changes to Network configuration, including potential IP address change, can lead to complexity and delays in replatforming. Deployment, scaling, patching automation for application and operating system that is built with VMware API will need to be redeveloped with Terraform and Ansible scripts. Additional IBM Cloud services can be required to provide comparable VMware capabilities. IBM Cloud provides migration tooling from on-premises VMware to IBM Cloud Virtual Server Instance, the RackWare Management Module (RMM) migration solution provides a seamless virtual-to-virtual replatforming for VMware virtual machine (VM) to IBM Cloud® virtual server instance migration. It allows the adoption of existing capabilities of IBM Cloud. For more information, see About virtual server instances for VPC and VSI deployment pattern in VPC
Enable gen AI inferencing and model training in production processes	Clients are faced with significant capital expenditure to acquire and implement gen AI accelerator clusters in on-premises data center to enable gen AI capability such as inferencing in production processes. For more information, see Managing GPUs. IBM Cloud provides GPU-enabled compute profiles on demand with cost-effective access to NVIDIA GPUs helping to accelerate compute time for compute intensive workloads such as inferencing, machine learning or model training. Many watsonx ai services and capabilities such as catalog services or deployable patterns to accelerate gen AI into business processes. For more information, see watsonx.ai SaaS with Assistant and Governance and AI summarization by using highly resilient serverless architecture pattern.
Providing flexible, scalable, and secure mission production critical Kubernetes platform	Kubernetes is the leading cross industry container orchestration platform for hosting modernized cloud native applications. Clients are challenged to deliver two platforms, VMware and Kubernetes, to meet required business availability, scalability and efficiency. IBM Cloud provides a secure business critical enterprise Kubernetes platform offering by using the leading industry Red Hat OpenShift platform. Multi architecture compute for Red Hat OpenShift Clusters now allows for mixed cluster compute nodes across x86, Power, and LinuxONE on IBM Z systems. Deploying Red Hat OpenShift on IBM Cloud Bare Metal servers provides elimination of Hypervisor costs. For more information, see Getting started with Red Hat OpenShift on IBM Cloud.
Providing an alternative to x86 enterprise scale compute platform	During the assessment phase for move to cloud, some of the large x86 workloads that are either hosted on the VMware hypervisor or on x86 bare metal, can be identified to benefit from a change to a more cost efficient and performant platform such as PowerVS on IBM Cloud. IBM manages the PowerVS Hypervisor, which uses firmware based virtualization that minimizes hypervisor compute overhead. It provides the capability to activate capacity when required and deactivate when it is no longer needed. Power VS provides support for dynamic relocation of CPU, RAM, storage, and I/O without impacting the workload. It offers a secure stack from virtual instance to application including guest runtime security and support for Oracle hardware partitioning.

The decision tree for optimizing the x86 workload is based on recommendations from the migrate to cloud assessment phase.