Where we keep you updated on the latest developments and trends in the field of virtualization. Virtualization is the process of creating a software-based representation of something, such as a server, a network, or an application, that can run on different hardware platforms. Virtualization can improve efficiency, flexibility, scalability, and security of IT systems. In this edition, we will discuss some of the recent news and announcements related to virtualization, such as:

• Cisco’s decision to end HyperFlex development and partner with Nutanix
• IBM’s plans to deliver mainframe AI services and support
• MacStadium’s VMware virtualization replacement now available at AWS
• EU’s approval of Broadcom’s acquisition of VMware with conditions
• Verge.io’s addition of GPU virtualization to its HCI software

Cisco snuffs HyperFlex development, hands HCI future to Nutanix

Cisco has announced that it will stop developing its own hyperconverged infrastructure (HCI) product, HyperFlex, and instead partner with Nutanix to offer its customers a choice of HCI solutions. HyperFlex was Cisco’s attempt to compete with market leaders such as VMware, Dell EMC, and Nutanix in the HCI space, which combines compute, storage, and networking in a single appliance. However, HyperFlex failed to gain significant market share and faced technical challenges. Cisco will continue to support existing HyperFlex customers until 2028, but will not release any new features or hardware for the product1.

• Cisco and Nutanix have signed a strategic partnership agreement that will enable Cisco customers to run Nutanix’s HCI software on Cisco’s Unified Computing System (UCS) servers1.
• Nutanix’s HCI software supports multiple hypervisors, including VMware ESXi, Microsoft Hyper-V, and Nutanix’s own AHV1.
• Cisco and Nutanix will also collaborate on joint go-to-market activities, technical integration, and customer support1.
• Cisco’s move is seen as a recognition of Nutanix’s leadership and innovation in the HCI market1.

IBM set to deliver mainframe AI services, support

IBM is preparing to launch a new set of services and support for its mainframe customers who want to leverage artificial intelligence (AI) capabilities on their systems. IBM’s mainframes are powerful computers that run mission-critical applications for large enterprises in various industries. IBM claims that its mainframes can handle AI workloads better than cloud-based solutions due to their high performance, reliability, security, and data privacy2.

• IBM will offer a range of AI services for mainframe customers, such as AI strategy and roadmap development, AI model development and deployment, AI governance and management, and AI skills training2.
• IBM will also provide AI support for mainframe customers, such as AI troubleshooting, AI optimization, AI monitoring, and AI security2.
• IBM will leverage its own AI technologies, such as Watson and AutoAI, as well as open-source frameworks, such as TensorFlow and PyTorch, to deliver AI solutions for mainframe customers2.
• IBM’s mainframe AI services and support will be available in the first quarter of 20232.

MacStadium’s VMware virtualization replacement now available at AWS

MacStadium, a company that provides Mac infrastructure as a service for developers who need to run macOS applications or test iOS apps on real devices, has launched a new product that replaces VMware’s vSphere environment for virtualized Macs. VMware’s vSphere is a popular platform for managing virtual machines (VMs) on different types of hardware. However, VMware announced in August 2021 that it would stop supporting vSphere on Apple hardware due to Apple’s transition from Intel to ARM processors3.

• MacStadium’s new product is called Orka Cloud Services (OCS), and it is now available on Amazon Web Services (AWS)3.
• OCS allows developers to create and manage VMs running macOS on AWS bare metal instances using MacStadium’s own orchestration layer called Orka3.
• OCS supports both Intel-based and ARM-based Macs running macOS Big Sur or Monterey3.
• OCS is compatible with various tools and platforms that developers use for continuous integration and delivery (CI/CD), such as Jenkins, GitLab, GitHub Actions, Bitbucket Pipelines, and Xcode Cloud3.

EU OKs Broadcom acquisition of VMware with strings attached

Broadcom’s bid to buy up VMware has cleared a key regulatory hurdle with the conditional approval of EU authorities. Broadcom is a chipmaker that produces components for various devices and systems. VMware is a software company that specializes in virtualization, cloud computing, and networking. Broadcom announced its intention to acquire VMware for $61 billion in August 2021.

• The EU Commission, which is responsible for ensuring fair competition in the European market, has approved the deal on the condition that Broadcom will ensure interoperability and compatibility between its hardware products and VMware’s software products, as well as other software products from VMware’s competitors.
• The EU Commission was concerned that Broadcom could use its dominant position in the market for Fiber Channel host bus adapters (HBAs), which are devices that connect servers to storage networks, to harm competition and innovation in the market for virtualization and cloud software.
• The EU Commission’s approval is subject to a market test, which means that interested parties can provide feedback on the proposed commitments before they become legally binding.

Verge.io adds GPU virtualization to its HCI software

Verge.io, a startup that offers a software-defined hyperconverged infrastructure (SD-HCI) platform, has added GPU virtualization to its product. GPU virtualization is the process of sharing the processing power of graphics processing units (GPUs), which are specialized chips that can perform complex calculations faster than CPUs, among multiple VMs or containers. GPU virtualization can enable applications that require high-performance computing, such as AI, machine learning, gaming, and graphics rendering.

• Verge.io’s SD-HCI platform allows customers to create their own HCI systems using any x86-based hardware, such as servers, laptops, or desktops.
• Verge.io’s GPU virtualization feature supports both NVIDIA and AMD GPUs, and it can dynamically allocate GPU resources to VMs or containers based on their needs.
• Verge.io claims that its GPU virtualization feature can deliver up to 95% utilization of GPU resources, compared to 10% to 15% utilization achieved by traditional methods.
• Verge.io’s GPU virtualization feature is available as part of its SD-HCI platform, which costs $99 per node per month.

That’s all for this edition of the Virtualization Weekly newsletter. We hope you found it useful and informative. If you have any feedback, questions, or suggestions, Thank you for your time and attention. See you next week!