Author Topic: Guide to Cloud Mining Memory Coin  (Read 13806 times)

0 Members and 1 Guest are viewing this topic.

Offline ManeBjorn

  • Full Member
  • ***
  • Posts: 174
    • View Profile
I hate to necro bump but I received a $100 AWS credit.
I should be able to run an instance for quite a while for this it sounds like.
Any rough ideas on cost so I can stop it before the credit runs out?

I ran 100 instances at $0.27/hour - so $27/hour. $648/day. $20K/mo.
Tips Appreciated MMC | MQotia5NmKTsaXXPaN8hw2ze3Z1DNwi39s
https://www.multibet.io/index.php?ref=ManeBjorn

Offline ManeBjorn

  • Full Member
  • ***
  • Posts: 174
    • View Profile
Roughly how much per month does it cost you to run?


I have a couple questions.
I can use AWS (Amazon) is there an easy setup for that service if I choose to go that direction?

That's what I use. When setting up your instance (I used Linux AMI (HVM) instance), under user data, paste this script:

#! /bin/bash
wget https://www.dropbox.com/s/yam_miner
tar zxvf yam-yvg1900-M7k-linux64-sandy-bridge.tgz
cd yam-yvg1900-M7k-linux64-sandy-bridge
cd linux64-sandy-bridge
wget https://www.dropbox.com/s/your_yam_config
./yam --config yam-mmc.cfg

User data is automatically run when the instance starts up. Thus, it will automatically connect to a pool and start mining. Obviously, setup your yam-config file beforehand, then add both the yam miner and yamconfig to your dropbox and edit the script with those URLs. With new yam releases, you may need to edit the folders the script opens (cd yam-yvg1900-M7k-linux64-sandy-bridge -> cd yam-yvg1900-M7m-linux64-sandy-bridge for yam 'm' version).

cc2.8xlarge uses sandy bridge architecture. All of the c3.#xlarge use haswell. Change your yam miner and edit the script as needed, depending on which miner you use.

I get ~34hpm with cc2.8xlarge. Feel free to ask any other questions.
Tips Appreciated MMC | MQotia5NmKTsaXXPaN8hw2ze3Z1DNwi39s
https://www.multibet.io/index.php?ref=ManeBjorn

Offline Linolis

  • Jr. Member
  • **
  • Posts: 22
    • View Profile
I have a couple questions.
I can use AWS (Amazon) is there an easy setup for that service if I choose to go that direction?

That's what I use. When setting up your instance (I used Linux AMI (HVM) instance), under user data, paste this script:

#! /bin/bash
wget https://www.dropbox.com/s/yam_miner
tar zxvf yam-yvg1900-M7k-linux64-sandy-bridge.tgz
cd yam-yvg1900-M7k-linux64-sandy-bridge
cd linux64-sandy-bridge
wget https://www.dropbox.com/s/your_yam_config
./yam --config yam-mmc.cfg

User data is automatically run when the instance starts up. Thus, it will automatically connect to a pool and start mining. Obviously, setup your yam-config file beforehand, then add both the yam miner and yamconfig to your dropbox and edit the script with those URLs. With new yam releases, you may need to edit the folders the script opens (cd yam-yvg1900-M7k-linux64-sandy-bridge -> cd yam-yvg1900-M7m-linux64-sandy-bridge for yam 'm' version).

cc2.8xlarge uses sandy bridge architecture. All of the c3.#xlarge use haswell. Change your yam miner and edit the script as needed, depending on which miner you use.

I get ~34hpm with cc2.8xlarge. Feel free to ask any other questions.
« Last Edit: February 13, 2014, 06:19:47 pm by Linolis »

Offline ManeBjorn

  • Full Member
  • ***
  • Posts: 174
    • View Profile
Thank you.  :)

I have a couple questions.
I can use AWS (Amazon) is there an easy setup for that service if I choose to go that direction?
I'll create an AMI for that and explain how to use it.
Tips Appreciated MMC | MQotia5NmKTsaXXPaN8hw2ze3Z1DNwi39s
https://www.multibet.io/index.php?ref=ManeBjorn

Offline itsik78

  • Sr. Member
  • ****
  • Posts: 297
    • View Profile
I have a couple questions.
I can use AWS (Amazon) is there an easy setup for that service if I choose to go that direction?
I'll create an AMI for that and explain how to use it.

Offline ManeBjorn

  • Full Member
  • ***
  • Posts: 174
    • View Profile
I have a couple questions.
I can use AWS (Amazon) is there an easy setup for that service if I choose to go that direction?
Tips Appreciated MMC | MQotia5NmKTsaXXPaN8hw2ze3Z1DNwi39s
https://www.multibet.io/index.php?ref=ManeBjorn

Offline captain1821

  • Newbie
  • *
  • Posts: 8
    • View Profile
No, there's no optimized version for that architecture.

I do not know about which of the 2 architectures you talk about, but it seems that for the Azure one, there is not even a working version...

Offline seraphim

No, there's no optimized version for that architecture.
Meet you on STEEM

Offline captain1821

  • Newbie
  • *
  • Posts: 8
    • View Profile
What's the output of
cat /proc/cpuinfo | grep "model name"

The os you choose won't make a (big) difference

At DigitalOcean the output is: QEMU Virtual CPU version 1.0

So far the best combination is Ubutu with nehalem

++++++++++++++++

At windows Azure the output is AMD Opteron (tm) Processor 4171 HE

Any ideas about the best combination?
« Last Edit: February 12, 2014, 06:38:23 pm by captain1821 »

Offline seraphim

What's the output of
cat /proc/cpuinfo | grep "model name"

The os you choose won't make a (big) difference
Meet you on STEEM

Offline captain1821

  • Newbie
  • *
  • Posts: 8
    • View Profile
Yeah, it depends on the CPU your cloud service uses. I use Amazon's previous-generation CPUs, which are sandy bridge. If yours are the newest, Haswell will probably be the best yam version to use. You can post here what CPU you are using (example Xeon ####) and I can tell you what architecture it is.

So far I'm testing at DigitalOcean. I tried to find what kind of CPUs they use but failed.

I test on "Ubuntu 13.10 x64" and 4 CPUs. Sandy-bridge failed, Ivy failed, Generic WORKS!

Will test Haswell, barcelona and nehalem now.

Please let me know if I should test any of the rest.

Also please let me know if it makes sence to test on delian.

Will report with my results!

Thank you!


Offline Linolis

  • Jr. Member
  • **
  • Posts: 22
    • View Profile
Yeah, it depends on the CPU your cloud service uses. I use Amazon's previous-generation CPUs, which are sandy bridge. If yours are the newest, Haswell will probably be the best yam version to use. You can post here what CPU you are using (example Xeon ####) and I can tell you what architecture it is.

Offline seraphim

1.: threads=0 uses all available cores, no need to change
2.: no idea what cpu they use. i'd advise you to test several versions on one box before
Meet you on STEEM

Offline captain1821

  • Newbie
  • *
  • Posts: 8
    • View Profile
the letter is like the version number, m > k

You need to download the file for your architecture and extract it. There's a yam-mmc.cfg inside the tar. Modify that file to use your addresses and upload it to your own dropbox account. Upload the tar too.
Then, modify the script to use your dropbox urls and the file/foldernames for your chosen architecture. You can add a cd ~ after #! /bin/bash to be sure it's happening in your home folder.

It seems that I start to wake up now... LOL!

So, I have to edit this adding my own addresses and I’m done:

threads = 0
mining-params = mmc:av=1&aesni=on&m=1024
mine = getwork://MTb1JHjoJCSJBjTLUgqeyJ1nXxYxBRJEUX@work.mmcpool.com:80:8880:8881:8882:8883/mmc
mine = getwork://MSkLPYJQuskzp8wNqvZCpiTNYw7pJ83zGe@mmcpool.1gh.com:8080:8081:8082:8083/mmc
mine = getwork://yvg1900.mmc_1:x@mmc.gpools.com:8080/mmc
#proxy = socks4a://127.0.0.1:9150
compact-stats = 1
print-timestamps = 0


Final questions

1. If I create a droplet with 4 CPUs do I have to edit the “threads = 0” or something else?

2. Since I’m creating clouds to services like DigitalOcean, HP Cloud and Windows Azure, will yam-yvg1900-M7m-linux64-sandy-bridge.tgz do for “my” architecture?

Big thank you to all of you that help!

Offline seraphim

the letter is like the version number, m > k

You need to download the file for your architecture and extract it. There's a yam-mmc.cfg inside the tar. Modify that file to use your addresses and upload it to your own dropbox account. Upload the tar too.
Then, modify the script to use your dropbox urls and the file/foldernames for your chosen architecture. You can add a cd ~ after #! /bin/bash to be sure it's happening in your home folder.
Meet you on STEEM