Google Tried to Charge Me 3 600 yr I Built My Own Server Instead

This is a PC, but not just any PC. It's actually the biggest new upgrade to my office. A server. It's sleek, it's quiet, and it might just save me thousands of dollars per year on cloud subscription fees. Now, I've been using Google Drive for the past five years, but with all of the videos we publish now, our storage needs have gone up like crazy. Now, recently, I wanted to increase my Google Drive storage from 10 terabytes to 20 terabytes, and Google asked me for $300 per month for a measly 10 terabytes of data. You heard that right, $300 per month, or $3,600 per year. With that money, I could build a banging new PC every single year, so yeah, I've had enough. Instead of paying them, I figured, why not make my own server for much, much, much less? Now, I've watched a lot of server videos on YouTube, and they all make it look so easy. All you need is an old PC, some hard drives, and an internet connection. Simple enough, right? Well, not exactly, as you'll see soon, but is it worth it? Did I waste all my time building my own server only to go running back to Google Drive, or is this something I should have done many years ago? So, this isn't my first server, but it's my first real server, one where I'm actually learning whatever it is that I'm even doing. I've been hearing the word server for years now, but what the heck is it? After countless hours of research and video watching, I think I finally get it. A server is a computer, this thing, that other devices, such as PCs or phones, can use, too. And they connect to the server over a network, aka the internet, or your own local wired LAN connection. So, if your server stores files, your PC and phone can use those files, too. If your server runs programs, other devices can, too, without running it on their own device. And if you're anything like previous me, you probably think this is too hard and it's meant for IT nerds who wanna make everything too complicated and just not practical at all. Yeah, I thought that, too. But after actually building and using my new server, I think everyone should do it. You can store your old photos, music, movies, family videos, documents, anything. It's freedom and control at its finest. Maybe it's time to finally be unshackled from the confines of clouds like iCloud, Google Drive, or Dropbox. It's awesome, and it's a lot easier and more affordable than you think it is. While pre-built servers exist, I highly recommend setting one up yourself. You'll have more control and be able to troubleshoot if problems come up. And trust me, they'll come up. Sure, I've been tempted to just keep paying Google. You can just keep throwing money at the problem and someone else will solve it for you. It's pretty appealing. But I had conquered swimming recently, and Linux, and flying with our little toddler for the very first time. I was on a hot streak of doing hard things. So why not build a server too? So the goal with this new server I'm building is to backup all of our old videos that currently exist in a Google Drive in a folder aptly named old switch and click videos and are currently active projects in another folder in case something happens to Google Drive, like if it gets hacked or has a data problem. And these two folders currently use between six and seven terabytes of data. So basically the goal is to build one big glorified off-cloud storage, like a big hard drive, but better. And this is called a NAS, which is network attached storage. It just stores files for you. So you don't have to store it in your own PC and it's attached via a network. I know you probably don't have huge video files like we do, but you probably do have files and data you want to store. And it's a great way to get your server started because then you can build off and do some really cool things. But I'm getting ahead of myself. Let's just get started. But first let me disclaim that I have zero IT background. And if I do anything wrong or something could be done better, please comment respectfully below so we can all learn. But let's just say I learned a whole lot from doing this. So this is an old PC that's just been sitting in the closet for a long time, about three-ish years. You can use any old PC or laptop or even a Raspberry Pi. It doesn't need to be high spec either. So maybe it's working, maybe not. We're going to test that since it's been about three years since it's been turned on. After attempting to turn it on, it got stuck on this automatic repair screen over and over. And I know this is a Windows problem. And since I just want the PC for parts, I don't actually need the Windows OS. So I installed a clean Linux Mint on it and it booted. So it's working. Yay. But there's one big problem. This PC case doesn't have much room for hard drives. I think it has two and I don't even know how to access it. Since it's going to be a NAS, it's going to need a lot of hard drive space. But which case do I get? I haven't exactly been in the market for building PCs. So my knowledge of cases is pretty outdated. And a quick Reddit search quickly fixed that though. It seems like the Fractal R series cases are perfect server PCs. And after more research, the Fractal Define R5 was perfect. It's all black, so super simple look. There's no tempered glass, no clear see-through anything. So cable management isn't a huge deal if I don't nail it. And let's be honest, I'm not going to nail it. And it can hold up to eight hard drives. That's more drives than I've ever seen in a PC. Wait, do I even have enough SATA ports for eight drives? In the meantime, I took apart all of my old PC parts and got them ready to transfer over to the new case. And how hard could rebuilding a PC that I built many, many years ago be? We'll see about that. I've built three PCs in total in my whole life, but thankfully there's full video guides on YouTube that I can just follow along step-by-step, which I did until I got to the pre-build test, when you test a PC with the minimal amount of parts required before actually putting in the case. So I did that and nothing happened. No fans started spinning, no lights lit up, just nothing. What had I done wrong? I followed these steps exactly. And the PC was working before I took it apart. So after trying that over and over, I gave up. And the box of parts just sat in the corner in the office for weeks, maybe even months. I thought maybe someday that the thing would just fix itself but that's just wishful thinking. So I got our production assistant, AKA Jake's sister, to lug the box of parts to Micro Center to get it diagnosed. And surprisingly, it was only $20, which is not bad at all. They said the motherboard stopped working and to get the exact same motherboard was $180. So she bought it and brought it back. And with working parts, the build went pretty flawless. It took about four hours of building and it was done. I pressed the power button to test it and it worked. Things were finally going my way and I enjoyed the process too. This was actually fun. So at this point, we've got a working PC and I'll list the final parts here on the screen and in the PC part picker link below. It's currently running Linux Mint Mate and I've got two, soon to be six, eight terabyte hard drives along with an overkill amount of chases, fan, chassis, fans, fans, cooling fans, chassis fans. So if I were to rebuy this whole PC today, it would be about $1,800 for my old parts. So about 50% of the cost of Google Drive's 3,600 a year but I get more data. It's definitely a win and way, way within the budget. So the PC is done. You would think that that's the hard part, but nope. The next couple steps get a lot harder. So because this next step is a bit more permanent and fixing it becomes a pain in the butt, you basically have to start over. Yeah, it was a difficult decision and that's deciding what OS to run my NAS on. There are so many that it becomes overwhelming to pick. It's similar to Linux distros where there are a ton to pick from, but only a small amount will be a good fit for you. And as a total beginner, I wanted something simple but powerful, but also had a lot of community support where I could just look things up or ask someone else. And I narrowed it down to two, TrueNAS and Unraid, both of which are very popular options. The biggest difference, one is free, the other costs some money. Okay, there's actually a lot more differences, but I don't quite understand them yet. Since this is my very first DIY NAS, I figured it would be better to go with something more cost effective like TrueNAS. Plus I've heard a lot of good things. I've heard about it before without even being into servers, so it must be good, right? Setup was easy enough. All I had to do was load an ISO onto a boot drive and then boot it up into the newly built PC, just like how you would with any operating system. Now that was the easy part. What followed was a tad bit harder, a lot harder. Took me a long time. And since I have no IT background, I didn't understand most of what I was seeing on the screen. It asked me to go over menu items like using the web UI which I assume you want, I want. When it comes to problem solving, my strategy nowadays is to have another laptop nearby to look up what I should do either on a search engine or to ask chat GPT. A lot of people said to use AI when I encounter problems, so I'm using it. Honestly, at this point, I was pretty happy with myself. I built a PC, I loaded up TrueNAS, and I was well on my way to having a functional server. Boy, was I wrong. The next step was to set up my pools of data. Not only did I not know how to do that, but I didn't even know what half the stuff I was reading meant. VDEV, ZFS, RAIDs, what am I even looking at? So I went back and forth between chat GPT, the TrueNAS documentation, and the web UI to try to figure it all out one step at a time. So now that I've done it all, I can truly say that I was an idiot. Remember how I referenced PC building guides and followed them step-by-step to build my PC? Well, there are full setup guides on YouTube that will walk you through step-by-step on how to set up your TrueNAS server. But I didn't realize that until I had already done it myself. So here's some basic lingo, if you're doing this yourself. A pool is basically a big container that holds all your data. The big warehouse building, if you will. Then VDEVs, aka virtual devices, are subsets of the pool. This is how the warehouse is laid out in different sections. Then inside VDEVs are datasets. These are the boxes that are going to hold all of your data. It's much more organized than having everything just spilled out all over the warehouse floor with no shelves or labels. Now that that's out of the way, I created my first pool named Old Switch & Click Videos, which is way too long of a name. I should have made that way shorter. It would have made my life way easier, but I didn't. Then I had to make my VDEV, which was harder than I thought. So inside that menu, it gave me a ton of options of what type of VDEV you want. That's where Stripes, Mirrors, and Raids come in and not the ones you do in games. Basically a Stripe is like putting your files into the device with no redundancy, no backups, no copies. This is the most risky option. If it fails, that's it. You have to start all over again. Then a Mirror is having two exact copies of each other. You need two drives to do this. If one fails, you've got the other one. Then there's higher levels of protection and redundancy called Raids. This part was a bit more confusing, but the higher the Raid number, the more drives you need. For example, Raid 5 uses two usable drives with one Parity drive. Parity drive, not like the comedy parody, but Parity. A Parity drive is what you use to rebuild a drive when it fails, but it can still take hours or even days. So that's nice, right? You get some protection in case of drive failure, which would be nice. But unfortunately, I only have two identical eight terabyte drives, so I'm doing Mirrors, which isn't so bad because I have two copies. So this is only level zero of my server. In the future, I do want Parity. I want remote access and I wanna do a lot more with it, but I can't upgrade something that doesn't even exist yet. So we've gotta start somewhere. So I've got my pool and my VDEV with eight terabyte drives in it. And I didn't realize at the time, but if one of your VDEV goes down, the entire pool is done for. And this is why redundancies are so important. For now, I've got some type of safety net and it feels good. We're making progress, a lot of progress. So the next challenge arose. How do I transfer all my files from my Google Drive to my NAS? Do I have to download all seven terabytes of data and then re-upload it? That would take me days and days of waiting. Well, fortunately, ChatGBT has all the answers, yay. There's something called a cloud sync where you can sync data from your cloud storage, like Google Drive, Dropbox, or whatever the heck else you use. Not Apple iCloud though, they don't let you do that. You can sync your drive to your NAS and it was easy and straightforward. I went through the menu options, selected what folder I wanted to sync, when I wanted to do it, and I pressed save. And it took basically an entire day and my internet was super slow because of it, but eventually it finished and I had done it. I had actually done it. I had backed up my Google Drive folder. Well, I think I did because I have to make sure that it actually worked and I didn't know how to do that. The true NAS web UI doesn't have a file browser, so you can't just check if everything copied in, but you can set up this thing called SMB, which stands for server message block. Knowing that isn't exactly helpful though, but it's a way for your computers to interact with your server as long as you're on the same network. So using SMB, I can access the files with my Mac, my laptop, Jake's computer, or any other computer in the office. It sounds great. What could go wrong? Nothing, surely, but surprise. I didn't know about datasets when I set this up, so I transferred the data directly into my VDEV root folder. Oh, how naive I was. Well, sharing an entire pool is a bit tougher than sharing a dataset. It's not impossible, but it's not recommended. It can cause a lot of problems later on, problems that I don't know about yet, but I've read about them and they don't sound so good. Stuff like permission headaches and not being able to move or manage your data. But for now, I think I can live with a couple things breaking here and there. I sort of have to live with it because I tried moving all my data into a dataset and it didn't work. So I accept my errors. I'm gonna move forward and set up SMB sharing. So I created new users, myself and Jake. And once it worked though, it was awesome. The most awesome thing is to be able to browse through your files, click in and open it. And it felt fast, really fast. I could browse old B-roll, look at thumbnails. I could even upload files into the network instantly. So at that moment, I knew that what I was working on was worth it. Now it's not about saving money anymore. I actually have a living, breathing server now. After these months of procrastinating, it's finally alive. Should I name it Cerberus or something else? A server needs a name. Let me know your name ideas. But over the next few days, more problems started to pop up. Like how is this even useful to us and our business? We still have to use Google Drive for active projects because we have editors that live all the way around the world. So I got to brainstorming. This is how I think it's going to work but it might get expensive and it might require me to redo everything that I just did. So currently on Google Drive, there's everything. There's old videos, active projects, future videos, everything. I want to separate that into two categories, an archive and active projects. But there's a problem. We're always still accessing old files from our archive, namely to get B-roll. So there needs to be a way to do that. So what if we move B-roll into its own library before the projects are archived? Now active projects has three data sets, B-roll, thumbnails, and active videos. But no one wants to edit off HDDs. That's loud, annoying, and slow. So to future-proof our server, it makes sense for the active projects folder to use SSDs. So when I went to look up the price of eight terabyte SSDs, the price was pretty steep. And like always, there were more problems that arose simply from me not knowing certain facts about PC building. My computer wouldn't even recognize the new SSDs and I connected the cables multiple times to make sure they worked. It wasn't until I read the motherboard manual that I finally understood the issue. If there's an NVMe slot taken, then SATA ports five and six are disabled. Now I don't know how I could have known that unless I read motherboard manuals for a living, but I don't. So I moved all my SATA cables to one, two, three, and four, and boom, it started working. Was it worth it? Spending $1,800 on two eight terabyte SSDs? Most people might say no. Heck, I even said no initially. I might still say no. But it's still cheaper than a full year of 10 terabytes on Google Drive. A lot cheaper. But the more I learned and the more I looked at my storage solution, the more I realized it wasn't enough. I already have all of our old videos on the NAS and it's taken up like 90% of the space. I just built it and I already need to upgrade it. And if I do another mirror, I only get 50% of the capacity. So you only get 16 terabytes of real data storage. But if I can change that to a RAID Z, I think a one, now I can store 24 terabytes of data. And to make that all possible, I now have a long to-do list. It just seems like the tasks are never ending. Like I have to constantly say like, oh, I gotta work on the server today. You'll see the final version of this after the list is done in the next episode where we will do something cool, which I'll tell you soon. So while I feel like I only learned what I had to learn to make this thing work, I came out way ahead financially and experience. It's definitely worth it. And I wish I had done this many, many years ago in tandem with using Google Drive. Now, if you've been wanting to try this out, I highly recommend it. It can be super affordable and fast. You throw TrueNAS onto an old PC and you just start tinkering away. It's not as scary as it looks. So my next step, next episode, setting up remote access. I'll see you there.