DOGE Mining Guide

The Amazon EC2 Edition, for absolute newbies

This guide entails setting up the custom OS Template "DOGE GPU" on Amazon EC2 - automatically configured to mine 200kh/s from an Amazon EC2 Instance.

We will go through setting up your AWS account, all the way to logging in to your server remotely, and entering your DOGE Pool details, for instant mining.

Overview

This is an overview of the general challenges in setting this up - those who don't care about the technical details behind how this is set up, can skip freely. This is simply for shibes who are interested in the workings of the system. As nothing in the system itself is locked down or encrypted, you can also edit, or add to it as you see fit. It's entirely open. Much happy!

GPU Mining is the most common way of earning doges, as it is the fastest, most cost-effective way of building up your doge riches. It is also the most common form of mining with other crypto-currencies; such as litecoins. However, those without either mid-range+ Radeon cards, or high end nVidia cards, will find their hashrates to be much low - many uncompetitive. Not wow :(

This is where EC2 steps in. Amazon have a great system where you can reserve a spot instance of a server - i.e: You'll only pay for it, when it drops within your price range. GPU Servers on Amazon are not often used - but not due to expense! Mostly because they are difficult to set up. GPU Instances are used by some individuals to mine litecoins - and some guides are available online, but all require a good working knowledge of Linux (and SUSE to an extent) in order to utilise, and set up. To set up a DOGE or LTC Miner on Amazon EC2 in the past, you would have to:

  • Rent a GPU instance at full price (around 80c/hour), so it doesn't shut down while you're configuring
  • Download and install deprecated nVidia drivers, and the CUDA5.0 toolkit
  • Download and install deprecated nVidia drivers, and the CUDA5.0 toolkit
  • Force the nVidia drivers and CUDA kit to actually install. Fuck SUSE.
  • Set up required dependancies.
  • Compile cudaminer from source, to run on Linux. Much dos2unix, wow.
  • Set up an after.local or init.d service to launch cudaminer on boot.
  • Ensure output is logged for shibes concerned with pool downtime.

This is where the DOGE Miner mark 1 comes in - I have taken the liberty of setting up an AMI which is now publicly available on Amazon, which does ALL of the above for you.

Getting Started

We're going to go through a guide on setting up cloud instances on Amazon EC2 - which basically means that you set a maximum price you're willing to pay for a DOGE Miner to come online, and do 200kh/s. Instead of worrying about the above setup, you can select the now publicly available, entirely free DOGE-Powered Operating system, and just enter your pool details. Whenever the hourly price drops on gpu instances to your desired value, your server will automatically boot up, connect to your pool, and start mining. Much wow.

STEP 1: Setting up an Amazon AWS Account


If you already have experience with Amazon AWS, You can entirely skip this section.

Setting up: You can set up a free Amazon AWS Account at http://aws.amazon.com/ . If you already have an Amazon account, simply log in. Note that you wilL NEED to verify this account with a Credit Card. It's instant, and does not apply any charge. Setting up is a bog-standard "give us some of your shibe details" system, ala first name/last name, etc etc.

You'll get your standard confirmation email, and you're ready to Doge. Now, firstly - a little information on what AWS is, for those who have NEVER used it before. AWS lets you rent servers on an hourly basis - and even provide a free 750 hours of server rental for their tiny tiers (which do about 4kh/s on cpu mining, for the very poor shibe). The more resources you want on your server, the more it'll cost - for example, a server with 16 CPU's and 32 Threads will cost about $2 per hour, and earn you about 90kh/s. Much expense! Avoid at all costs.

Now, you can rent servers based on RAM, HDD Space, CPU Power, etc etc - but you can also rent a server that has quite little amounts of RAM, HDD, CPU etc.. but has access to a dedicated GPU: The nVidia GRID. It's not Radeon, but it is one of highest performance nVidia cards out there.

There are two ways to rent these servers: either "on-demand", or "spot-request". On-Demand means you need it right now, it must stay online until you say stop, and will pay top doge for it. "spot-request" means you only want the server to come online when it reaches a price you have specified. So for example: If you can only afford 10c per hour of server time, your server will only come online when demand has dropped low enough, that Amazon can afford to rent it to you for this price.

What we want is the second one: only the rich, investment-banker shibes want to pay full-whack. Full price for a GPU instance is around 80c/hour. However, on spot-request, it reaches as low as 15c/hour. At the current market rate, this means you only need to mine more than 105 Doges to be in profit for that hour. Much easy with 200kh/s.

So, let's look at AWS again, and see how easy it is for Shibe to set this up. Up at the top-right of your AWS homepage, after logging in, you'll see a drop-down bar that says "My Account/Console". Mouse-over this, and click on "AWS Management Console". You'll then see all the services you're entitled to: and up in the top left, you'll see EC2. Click on this.

This will bring you to your dashboard - much information! However, all you need do is click "Instances" on the left-side. From here, you will be able to see all your running instances, and if they're online and mining your doges. If this is your first time here, you'll see nothing. Click the button that says "Launch Instance" at the top.

You'll see a bunch of operating system choices appear - from Amazon Linux to Ubuntu, and even Windows servers. We don't want any of these though. We want DOGE Power. Click "Community AMIs", on the left-side bar. Then, in the search bar on top of this page, type in "DOGE". You will see the following appear:

DOGE Coin - GPU Miner (such profit)


Now, what's very very important is the ami number! As this is an open operating system, anyone can edit it, and publish their own. Some may optimise it and improve it! However, some bad shibes may put not so nice things on there. The AMI Number depends on which location you're setting up your EC2 instance in - you can see this in the upper-right of your screen. Typically, the cheapest locations are Oregon, and N. California. To ensure you've got the base package, the very first DOGE Miner, as described on here, please refer to the below list - and be SURE your ami number matches up!

  • N. Virginia : ami-019db468
  • Oregon : ami-5085e160
  • N. California : ami-62a59527
  • Ireland : ami-62917915
  • Singapore : ami-eaebbfb8
  • Tokyo : ami-fbbedcfa
  • Sydney : ami-8f72edb5
  • South America : ami-db50f1c6

Hit "Select" on the right, to the DOGE GPU Miner instance. Now, you'll be asked to choose an Instance type. We want to GPU Mine - so on the left-hand side, select "GPU Instances". You should an option on this page, called "g2.2xlarge". Click it, and hit "Next: Configure Instance Details" at the bottom of the page.


From here, you can select how many instances of the DOGE Miner you want to run. I recommend you start with just one. Now, most important: There is a "purchasing option" section, with a checkbox that says "Request Spot Instance". If you do not tick this, you will pay the on-demand price, which is around $0.80 per hour. If you want to only pay money when the price drops below a certain threshold, tick the checkbox - and you can enter your "Maximum Price" in here. The next most important part is "Persistent Request" - if this tickbox is checked, it means that your server will stop running when the pricerange goes too high - and start running again when the price drops back down. If you're doing a spot instance, you'll want to have this checked.


As you can see, the cheapest instance available for GPU mining right now (3pm EST), is 0.21c/hour. This generally drops closer to 15c/hour in off-peak hours. Now, we don't need to add extra storage or any of the likes - so hit "Review and Launch" at the bottom of the page.

On the next page, you'll see lots of information: the most important part here is that this instance is not eligible for your free 750 hours. Unfortunately the GPU instances are too powerful to give away for free. Depending on the choices you made above, you'll be paying either 0.80c/hour for an on-demand instance, or a lower price on a spot-instance. Remember that an on-demand instance will have 100% uptime, wheras your spot-instance may not come online for some time.

When you're ready, hit the "Launch" button at the bottom of the page. Now, here's where things get a little complicated, so do try to follow. Best Shibe Concentration face.

To connect to your server, you need an authentication key - this is similar to a password, except you send the server your key instead of typing in a password. Select "Create a new key pair" from the drop-down bar, and give it a name. Any name is fine!


When you're done, hit "Download Key Pair" - this will download a .pem file. Keep this safe! You'll need it to access your server. Now, hit "Request Spot Instance" at the bottom of this window - or in the case where you're doing an on-Demand instance, you should have a button that just says "Launch".

If you're going for a spot-instance, you will need to play the waiting game. If you check under your "spot instances", you will see your pending request: it might fulfill immediately, or you might have to wait until demand drops enough for your price to be accepted.



Your server will instead show "active", under the state column, when it's ready to come online.

Step 2: Logging in to your new Server, and updating it with your pool details.

Now that you've got your server up and running, it's time to connect to it with your key file, and enter your details. You'll need two pieces of software to do this: PuTTY, and PuTTYGen.
PuTTY is used for connecting to your server - you can log in, and get what's basically a command prompt to run your server from.
PuTTYGen is what we'll use to make your key-file into a format that PuTTY can send to your server, to authenticate you.

Download PuTTY , and PuTTYGen from: http://www.chiark.greenend.org.uk/~sgtatham/putty/download.html
(Direct Links):
PuTTYGen: http://the.earth.li/~sgtatham/putty/latest/x86/puttygen.exe
PuTTY: http://the.earth.li/~sgtatham/putty/latest/x86/putty.exe

Once downloaded, run PuTTYGen. Click "File" from the top-bar, and "Load Private Key". Change the file extension (in the bottom-right) from "PuTTY Private Keys" to "All Files". Find the .pem file you downloaded from Amazon, and open it. You'll get a success notification - which you can hit OK on, and then select "Save Private Key" from the PuTTYGen window. Save this file to your desktop.
Now, we can use this to access your server.

Open PuTTY. For server name/Host IP, we need to get the address from your AWS Console. Under "Instances" on your AWS Console, select any running/active instance. In the Description bar at the bottom, you should see a "Public DNS", as follows:


For your host in PuTTY, you need to enter (without quotations): "root@publicDNSAddress". So for the above example, we would enter: "root@ec2-50-112-137-162.us-west-2.compute.amazonaws.com". Now we need to load our Key file - on the left-side menu on PuTTY, Expand the "Connection" section - and then expand the "SSH" section. Click on "Auth". You'll then see a browse button, which allows you to select the file you created with PuTTYGen. Select this, and hit the "Open" button at the bottom of the window.


Now, if you've followed everything correctly - you're now logged into your server, and will see something like this:


wow, such matrix.


Now, we're going to enter some commands on here - namely, we're going to input your pool details, and then reboot the machine.
type in the following command to your console, and hit enter:

nano /cudait
Now, you'll be given a very short file which will have configuration details. In here is a default account set up on netcodepool - you NEED to change this. I cannot stress this enough - if you do not change your details to match your details, you will not mine anything. You will infact, mine for an account owned by me. Perhaps Shibe is generous? Perhaps not. Perhaps update your details and dig some doges for your own account.
Your screen will look like this inside the config:




Ignore the first line - that's just so the launch protocol can find the right folder to start from.
The second line however, we need to change to update with our details. Use your arrow keys to move around - and replace the following:

stratum+tcp://doge.netcodepool.org:4094
If you do not mine on netcodepool, change this to match your own pool. If you don't have a pool yet, check out netcodepool! Much reliable, good community.

visali.Teslaminer:wowsupercompute
Here's the important part - enter your own details, in the following fashion:
accountName.workerName:workerPassword
Once you're done, hold down CTRL and hit X. (Command+X for Mac Shibes). You'll be asked if you want to save - hit Y for Yes. You'll now be brought back out to your console. Simply type "reboot", and hit enter. Your machine will shut down, and boot back up, and instantly start mining doges for your account. You can now close PuTTY.
And that's that - you can terminate or close your instance from Amazon's AWS at any time, and log back in via console at any time.

At current difficulty rates, you need to earn around 105 doges per hour for this to be profitable - 200kh/s will currently net you around 500-700 doges per hour, making for an excellent turnover.

If you would like to thank me for this guide, you can tip me some doges: D6NQHGMi4RgqubvkaBiLMBoDT1VCMxsULh

Or alternatively, please visit The DogeCoin Subreddit , join our community, and tip doges to users who helped you out, made you smile, or just because. Let's make this the cryptocurrency top doge.