Back to Blog
    Homelab
    Networking
    Infrastructure
    Proxmox

    My Network Used to Look Like This: The Nostalgia and Reality Behind a Homelab Diagram

    March 13, 2026
    6 min read read
    # “My Network Used to Look Like This”: The Nostalgia and Reality Behind a Homelab Diagram Every homelab has a moment in its history that looks perfect on paper. Multiple servers humming along. High-speed networking. Carefully planned infrastructure. Everything connected exactly the way it should be. It’s the kind of setup you proudly turn into a diagram and share with other enthusiasts. That’s exactly what happened when one user shared an old network diagram they built using **ExcaliDraw**. The diagram itself was created a couple of years ago for a class assignment, back when their infrastructure was far more elaborate than it is today. Since then, life happened. Hardware got sold. Equipment failed. The network evolved. What was once a carefully designed homelab slowly transformed into something smaller and more practical. But the diagram still exists—and it tells a story that many homelab builders recognize immediately. ## When a Diagram Becomes a Snapshot of a Moment Network diagrams have a strange quality. They freeze infrastructure in time. The diagram shared in this discussion was built during a period when the user had access to much more powerful hardware. Their environment included enterprise equipment like **HP DL380 Gen9 servers** and **10-gigabit Mellanox networking gear**. For a homelab, that’s serious hardware. But maintaining setups like that can be expensive. Power consumption adds up. Replacement parts cost money. And sometimes the equipment itself simply fails. In this case, the user explained that one of the servers had to be sold, along with the Mellanox 10-gig networking cards and receivers. The distribution switch eventually died as well—it had already been a hand-me-down nearing the end of its life. Suddenly the once-complex network looked very different. The diagram remained as a reminder of what the setup used to be. ## Why People Still Love Drawing Their Infrastructure Even in an age of automated infrastructure tools, people still draw diagrams. There’s a reason for that. Diagrams turn abstract systems into something visible. Instead of thinking about dozens of connections in your head, you see the entire architecture laid out in front of you. That’s especially helpful in homelabs, where people experiment constantly. Containers move between nodes. Firewalls change. Storage pools expand. A good diagram helps you remember what connects to what. And sometimes, it helps explain your setup to other people who are curious about how everything works. That curiosity showed up quickly in the discussion. Several people asked how the diagram was created, assuming it required programming knowledge or specialized tools. The answer was surprisingly simple. ## The Tool Behind the Diagram: ExcaliDraw The diagrams were created using **ExcaliDraw**, a collaborative drawing tool known for its hand-drawn style. Compared to traditional diagram tools like Visio or Draw.io, ExcaliDraw feels intentionally loose. Lines look sketchy. Shapes aren’t perfectly rigid. The result resembles something drawn on a whiteboard rather than a corporate presentation slide. That style makes it especially popular among developers and system administrators. It feels informal. You can sketch ideas quickly without worrying about perfect alignment or strict formatting. For many people, that freedom makes it easier to brainstorm complex systems. The author mentioned that they were using ExcaliDraw long before the recent surge of AI tools and coding assistants. It was simply their preferred way to map out infrastructure ideas. Interestingly, the diagrams were originally built using JSX-style components, blending coding concepts with visual design. That approach isn’t common, but it reflects how many developers think about systems today: as combinations of components connected by flows. ## Learning Infrastructure the Hard Way Another interesting part of the conversation was how the diagram came to exist in the first place. It wasn’t originally built for a production homelab or a work project. It was created as part of a **security class assignment**. The diagrams illustrated communication flows between systems—things like senders, receivers, encryption, and digital signatures. Those flows could represent email delivery, encrypted messaging, or other secure communication processes. In other words, the diagrams were less about hardware and more about understanding how secure data travels between systems. That academic project turned into something larger. The student ended up receiving an A for the assignment, and the diagrams became a foundation for learning more about networking and infrastructure design. For many people in the homelab community, that path feels familiar. You start with a class project. Then the project becomes a hobby. Eventually the hobby turns into real infrastructure. ## The Messy Side of Building Diagrams While the diagrams looked impressive, the creator admitted the process wasn’t exactly smooth. Spacing elements in complex diagrams can be surprisingly difficult. Once a network diagram grows beyond a few nodes, everything becomes harder to organize. Components overlap. Lines cross awkwardly. Sections of the architecture start crowding each other. The user joked that the “React zones” in the diagram had become messy while trying to reorganize everything. Anyone who has built large diagrams knows this frustration well. You start with a clean layout. Then you add another server. Then another network segment. Before long the diagram resembles a plate of spaghetti. Still, those imperfections are part of the process. ## From Diagrams to Real Infrastructure The discussion eventually shifted toward what the user is running today. One commenter suspected that a setup like this probably involved containers or Docker-based workloads. They were right. The author revealed that they currently run a **four-node Portainer cluster**, which manages containerized services across multiple machines. That’s a far cry from a simple single-server homelab. Even though the original hardware has changed, the spirit of experimentation remains the same. Containers, clusters, networking flows—all the elements that appeared in the diagrams still exist in the actual infrastructure. The diagram simply captured an earlier stage of that journey. ## Why Old Diagrams Still Matter It’s easy to dismiss old infrastructure diagrams once the hardware changes. But they often reveal something important: how someone learned to think about systems. Every homelab evolves. Servers get replaced. Networks shrink or grow. Technology changes quickly. But the lessons learned while designing those diagrams stay with you. Understanding how data flows through a system—how firewalls interact with containers, how encryption works between sender and receiver—builds a mental model that carries forward into future projects. That’s why many engineers keep their old diagrams around. They’re not just documentation. They’re a record of how their understanding of systems evolved. ## The Quiet Pride Behind a Homelab If there’s one emotion that runs through conversations like this, it’s quiet pride. Not the flashy kind you see in polished tech marketing videos. Instead, it’s the pride of someone who built something themselves. Something imperfect, experimental, and constantly evolving. The user who shared the diagram even admitted they were still learning—especially when it comes to frontend tools like React. Networking and containers were their strengths, but visualizing everything in code was a challenge. And yet they kept working through it. That’s the real story behind most homelabs. They’re not static systems. They’re ongoing experiments.