If you’ve spent years mastering the intricacies of vSphere, vCenter, NSX, and the broader VMware ecosystem, you know the power and stability it brings to the data center. But the tech landscape is always evolving, and perhaps you’re exploring alternatives like Platform9 Private Cloud Director (PCD) – maybe for cost reasons, flexibility, or a desire to embrace open standards.
Making that mental shift can feel daunting. New names, different architectures, open source – where do you even start? That’s exactly why we’ve put together a series of blog posts designed specifically for experienced VMware administrators like you. Our goal is to bridge the gap, translating familiar VMware concepts into the world of Platform9 PCD.
Why this series? Connecting the dots
Platform9 PCD offers a powerful platform for managing both traditional virtual machines (using KVM) and modern Kubernetes clusters. Under the hood, it leverages robust open-source technologies like KVM, ClusterAPI, and Open vSwitch.
While the end goals – high availability, efficient resource use, secure networking, persistent storage – are often the same ones you strive for in vSphere, the way Platform9 PCD achieves them can differ in implementation and terminology. This series aims to demystify those differences and highlight the similarities, making your exploration of Platform9 PCD much smoother.
A sneak peek at the topics we’ve covered:
Over the next few reads (or if you want to jump around!), we dive into specific comparisons:
- Keeping apps online (HA & Load Balancing): How does PCD’s built-in VM High Availability stack up against VMware HA? We compare the mechanisms for automatic VM restarts. We also look at load balancing, contrasting NSX LB concepts with PCD’s integrated LBaaS (Load Balancer as a Service), which notably uses an efficient OVN-based approach without requiring dedicated load balancer VMs.
- Smart resource allocation (DRS): Remember DRS automatically balancing VM workloads? We explore Platform9’s equivalent, Dynamic Resource Rebalancing (DRR), examining how it monitors cluster resources and uses live migration to optimize VM placement.
- Storage deep dive (Datastores, vVols, Thin Provisioning): Storage is critical! We break down how PCD handles storage compared to vSphere datastores. We look at PCD’s two main types: Ephemeral (like a temporary disk tied to the VM) and persistent Block Storage. We discuss how PCD Volumes, Backends, and Volume Types provide capabilities analogous to vVols and how Thin Provisioning works (including the benefits and crucial monitoring needs) compared to VMware’s implementation.
- Untangling the network (vSwitches, NSX): Networking can be complex. We compare the familiar vSphere Standard Switch (vSS) and Distributed Switch (vDS) concepts to PCD’s use of Open Virtual Network (OVN). We highlight how PCD provides vDS-like centralized management and forms the foundation for services like virtual routing and security groups (PCD’s answer to micro-segmentation, similar conceptually to NSX Distributed Firewall).
- Running modern apps (Kubernetes & Tanzu): If you’re looking at Tanzu for integrating Kubernetes with vSphere, you’ll want to see how Platform9 PCD compares. We explore PCD’s built-in Kubernetes management, focusing on its unique hosted control plane architecture (think EKS/GKE but private), its flexibility in deploying workers on VMs or bare metal, and how it leverages Cluster API for lifecycle management, auto-healing, and autoscaling.
Thinking of migrating? Potential concerns addressed
Moving from a familiar ecosystem always raises questions. Here are a few points VMware admins often consider, which our series touches upon:
- New names, familiar goals: Yes, there’s new terminology (Volumes, OVN, KVM, DRR, LBaaS, Security Groups). But as the blogs show, these often map directly to concepts you already know (Datastores/VMDK, Networking/SDN, Hypervisor, DRS, Load Balancer, Firewall rules). The goal is usually the same, just achieved with different (often open-source) tools.
- Control plane philosophy: Instead of vCenter and potentially NSX Managers, Platform9 uses its own Management Plane (available as SaaS or Self-Hosted), orchestrating open-source control services like ClusterAPI and OpenTofu. The Kubernetes control plane is also handled differently, often centrally managed for you rather than requiring dedicated VMs per cluster.
- Embracing Open Source: PCD is built on KVM for VMs and Cluster API for Kubernetes. This means less vendor lock-in, more flexibility, and leveraging huge communities. It also means interacting with a different ecosystem than the VMware-centric one.
- Feature nuances: While PCD aims for parity on core enterprise features, the exact implementation might differ. For example, the specific load balancing algorithms available right now in PCD’s LBaaS might differ from NSX’s full list, or the specific knobs available for DRR might differ from DRS. The blogs highlight these functional comparisons.
- Management experience: PCD offers a unified interface and API for managing both KVM VMs and Kubernetes clusters side-by-side, which can be a significant advantage.
Ready to dive deeper?
We hope this introduction provides a useful starting point. Each blog post in this series goes into much more detail on its specific topic, complete with comparison tables to help solidify your understanding.
Explore our eight learning modules and become a Private Cloud Director expert.
Exploring alternatives like Platform9 Private Cloud Director doesn’t mean discarding your years of virtualization expertise – it means applying that knowledge to a new, powerful, and open platform. We encourage you to read through the topics that interest you most!
