You're probably reading this post because you want to start a Minecraft server, or perhaps because you've started one and you want to expand. The short answer is "maybe not" in my opinion, but I'll lay down some facts and opinion. Maybe you'll think differently after you read this post, or perhaps you'll be more determined after reading this post. I'm fine with either.
My first Minecraft server (circa mid-2011) was hosted on a VPS with Linode. I remember when I used the /give command to give myself lots of diamond blocks and made goofy builds. Afterwards, I started a small friends-only Towny server using a Pentium 4 with only 768MB of memory hosted at home. Later, I had a failed public prison server (with horrible builds) on a GSP. I later managed to keep a near-vanilla survival server (Plain Jane) running for approximately nine months (before focusing on The Chunk). Today I operate crafting.today, another near-vanilla survival server with some people I know personally from Minecraft.
But my most significant experience was The Chunk, which I joined in August 2013. It had potential, and everyone there was hopeful we'd be the next big server. We believed in it, and we worked hard towards it. We had our crises:
- SkyDoesMinecraft crashing one of our game servers.
- Network downtime due to skid hack.
- Server reboots caused by bad hardware and slow support from OVH.
- Having to pull an all-nighter to rush-deploy an in-testing minigame because a YouTuber recorded and posted a video of it.
- Many more that I don't have the time to enumerate.
- Merging with CubeCraft went less smoothly than we'd wanted.
But that's only the small steps towards our goal. We ultimately failed, choosing to merge with a larger Minecraft network with more resources than we did. But many well-intentioned server owners have thought:
I'm going to be the next Hypixel!
Let's be frank: you can't launch the next Hypixel, or even the next big Minecraft server, unless you have the resources and ideas to make it even a mild success. There are technical, logistical and structural factors at work, which I will break down.
First, one obvious thing: you're certainly not going to be able to become the next big Minecraft server if you're going to buy this "unlimited" Minecraft server hosting plan from one popular Minecraft host:
Now, for the meat of the problem. The Chunk was faced with the issue of being able to scale to meet the needs of 1,000 to 2,000 concurrent players. Scaling to this level required immense technical work, from a new server deployment system to new infrastructure that could absorb the increased load. This infrastructure is not easy to create, and it took approximately three months before we could switch to PlayPen to provision all our servers.
The Chunk had a core staff team of approximately five people, organized into a flat structure with only the owner being the clear leader of the team. A few dozen more were admins, moderators and helpers, who kept the server orderly. We were mostly located in North America, making off-peak issues difficult to handle.
Hypixel has over 30 developers working for them, with many dozens of helpers and moderators keeping the server orderly. Hypixel's team works from all across the globe. Hypixel also has more resources than The Chunk ever did.
On top of this, The Chunk had no player analytics of any kind until January 2015. Having player analytics is crucial as they allow you to see which games get played and what maps/features/kits get used by players. Without analytics, you are left in the dark.
The Chunk was founded in mid-2013. At that time, you could start a server with more ease than you can today. Mojang was not strictly enforcing their EULA at this time, and everything seemed hunky-dory.
The tide turns
Fast forward to August 2014. Wolverness (a Bukkit contributor) discovered that the Bukkit project was owned by Mojang since early 2012. Because he did not know he was apparently contributing free work to a corporation without his knowledge, Wolverness issued a DMCA against the CraftBukkit distribution on bukkit.org and on GitHub, effectively shuttering CraftBukkit as Wolverness had contributed to major parts of the code. While other Minecraft server software projects also popped up (the most notable of which is Sponge) Spigot (which restarted development of CraftBukkit past 1.7.10) retained most server owners.
In September 2014, Microsoft bought Mojang. A year later, in February 2016, Mojang began to fully enforce the EULA against server owners. Two months later, Mojang added blacklisting functionality. Server owners developed techniques to work around the blacklist such as SquidHQ and automated SRV record creation.
Minecraft servers today
Today, your server is easy to miss, as many servers with the same concepts exist. While most Minecraft servers today are run by new server owners, others are run by experienced server owners who aim to get as much money as they can before starting yet another server to repeat the cycle ad infinitum.
One other trend is monopolization, especially minigame servers. Approximately three Minecraft servers have broken 30,000 concurrent players: Mineplex, Hypixel, and CubeCraft. Today, only Hypixel is the most popular. Even Hypixel has a bad prognosis for Minecraft, as it is working on a standalone game (apparently called Hytale).
So you still want to launch the next big Minecraft server?
Go right ahead. I'm not going to stop you. I hope you've learned something from my experience as a Minecraft server owner and developer. I will always wish you the best of luck with your server.
I originally wrote this for the server list I operate. I felt it was worth sharing to a wider audience, so I have reposted it here and removed references to my server list.