Scheduling Proof of Scalability

[abit]

1. I'm contributing with the following amount of BTS: 2500
2. I'm contributing with the following resources (for testers only):  N/A
3. I'm aiming to contribute following amount of transactions per second (for testers only): N/A

[betax]

1. I'm contributing with the following amount of BTS: 2500
2. I'm contributing with the following resources (for testers only): I will be testing using Azure instance up to ~ £80. If we are going to test transaction volume, we could get AWS and Azure together. This way we could test WAN and LAN.
3. I'm aiming to contribute following amount of transactions per second (for testers only): Depends on the instance.. up until now I have been running the test on a Standard A2
https://azure.microsoft.com/en-gb/documentation/articles/virtual-machines-size-specs/.

[clayop]

1. I'm contributing with the following amount of BTS: 2500
2. I'm contributing with the following resources (for testers only): I am not that a tech person, so I just can make spam transactions via VPS
3. I'm aiming to contribute following amount of transactions per second (for testers only): n/a

[puppies]

Using cubes description above, in regards to phase 1 testing (lan test) I am not 100% sure what has been proposed, so here is my pending pledge.  More information would be appreciated.
1. I'm contributing with the following amount of BTS: 2500
2. I'm contributing with the following resources (for testers only):I haven't played with AWS much.  If I can spin up an instance and get a gigabit connection to other instances I am more than willing to spin up an instance or two.  It will depend upon price and the time chosen.  I don't have a lot of time off during the next few weeks. 
3. I'm aiming to contribute following amount of transactions per second: TPS transmission seems to be limited by processor speed running the cli_wallet.  My i5 desktop at home seems capable of sustaining around 10-20 tps, or bursting up to about 50 tps with the flood_network command.  I would guess that each instance will be able to do about the same unless someone can find a way to optimize it.

As far as phase 2 goes.  (wan test)
1. I'm contributing with the following amount of BTS: n/a
2. I'm contributing with the following resources (for testers only): I could commit to running at least 5 servers.
3. I'm aiming to contribute following amount of transactions per second (for testers only): Should be able to sustain around 50-100 tps if I have enough CORE.

I would suggest that we complete phase 2 testing first.  If we can get 100 nodes flooding we should be able to break 1000tps.  With the windows binary available this may be possible.  We could get a set of simple directions to install and sync the client, (block production is not needed)  set up a wallet and account, and run xerocs script.  We could use mumble or irc to coordinate and hand out core. 

This should help give us the information we will need to successfully complete a proof of concept 100k tps test on aws instances.  This seems like it is going to need to be run on much more expensive hardware.  I think we have a better chance of it being successful if we have done some tests on cheaper hardware first.

[xeroc]

I could assist with scripting (when I have time) ... but I need a clear list of things you need implemented

[nethyb]

  Great work community...

1. I'm contributing with the following amount of BTS: 10,000
2. I'm contributing with the following resources (for testers only): I'm also prepared to also contibute $300 USD in AWS server resources
3. I'm aiming to contribute following amount of transactions per second (for testers only): n/a

I don't have a lot of time to contribute, but could spin up all the AWS servers / create in image etc if someone was to the initial dev/scripting/image work etc.


i.e. The $300 in AWS could be used for 1000 x 5hrs of m3.xlarge (0.06)  spot instances, or 150 x 5hrs of m4.10xlarge , happy to support whatever combination will get the best result.

m3.xlarge
4 vCPU's
15 GB Mem
2 x 40 (SSD)

m4.10xlarge
40vCPU's
160 Gb Mem
EBS Storage
10 Gigabit

[kenCode]

1. I'm contributing with the following amount of BTS: 2500
2. I'm contributing with the following resources (for testers only): n/a
3. I'm aiming to contribute following amount of transactions per second (for testers only): n/a

[kenCode]

I for one, would be happy to contribute some resources (donate some BTS) or spin up a number of AWS instances to help with load testing, and I'm sure many others in the community would also help if someone could develop the instructions or scripts to perform a private testnet load/tps test.
Using AWS, you could effectively have a 10Gb private LAN with the ability to spin up a significant number of servers for a short period of time at reasonable cost using spot pricing.
Here is an atricle on how someone has configured AWS instances to achieve 1 Million TPS For Just $1.68/Hour - http://highscalability.com/blog/2014/8/18/1-aerospike-server-x-1-amazon-ec2-instance-1-million-tps-for.html
If we chime and say we (as the community) would be willing to fund/support/provide resources for this test - it may encourage one us to take it on...

 
@nethyb @kenCode @cube @sudo @phillyguy @emailtooaj @onceuponatime @godzirra @rnglab @bobmaloney @ccedk @liondani @tonyk @Akado @mike623317 @CLains @DataSecurityNode @puppies @clayop @betax @abit @chryspano @Slappy @Xeldal @merockstar @tbone @Thom @Fox
 
Removing the WAN limitations could prove our scalability beyond 100K tps.. I'm looking forward to this AWS test.
It would be great if we could all record it on video with our phones. Split screen CLI and xeroc's tps gauge. Other ideas welcome.
 
This is 28 people so far willing to put in some BTS for this 100K tps LAN test.
Anybody else?

[emailtooaj]

1. I'm contributing with the following amount of BTS: 5000
2. I'm contributing with the following resources (for testers only): n/a
3. I'm aiming to contribute following amount of transactions per second (for testers only): n/a

If I had the technical know how I would be all in doing the test phase.
So please, anyone willing to step up and become a tester to spread some nods around the globe and make this push happen!!   

[BunkerChainLabs-DataSecurityNode]

1. I'm contributing with the following amount of BTS: 2500
2. I'm contributing with the following resources (for testers only): N/A
3. I'm aiming to contribute following amount of transactions per second (for testers only): N/A

Sorry I couldn't be more involved with the testing end of things at this time.

I am already maxed out between dposhub beta release and other bitshares related projects right now.

Really looking forward to the results.

  to Akado for wrangling this together.

[godzirra]

1. I'm contributing with the following amount of BTS: 2500
2. I'm contributing with the following resources (for testers only): n/a
3. I'm aiming to contribute following amount of transactions per second (for testers only): n/a

[rnglab]

Why do people have to donate bts? We can demonstrate scalability in a test net.

 A testnet made of nodes with typical Internet connections and processing power will let us know the starting point performance. Bypassing bandwidth and latency bottlenecks on a LAN testnet will bring scalability results, as Internet speed constantly grows (it likely follows Moore's law I guess) and that scalability should make it profitable for nodes to pay for better infrastructure and processing power when more TPS becomes necessary (as it means success for the DAC)

Then a testnet running on a cloud computing virtual LAN allows everyone to monitor that scalability, making results publicly verifiable.

[clayop]

Why do people have to donate bts? We can demonstrate scalability in a test net.

Perhaps it needs high performance VPSs, which cost at least $1 per hour.

[clout]

Why do people have to donate bts? We can demonstrate scalability in a test net.

[clayop]

2) Local Area Network would need to be in a gigabit environment so that we can test the possibility of 100K tps.  nethyb mentioned about aws having such a network. That is cool. Depending on the load test script created, we may end up either a MS Windows or a Linux environment.  The cost may vary slightly. 

We need to determine how much load (of the 100K tps) to offload in order to find out the optimal number of computers/aws instances to pump these transactions.  We probably need to do a test to find out the max load a computer/instance can take.

I am not sure if there are performance test scripts already present in the graphene test suite (I have not taken a look yet).  If not, we will need to develop one - possibly modifying xeroc's python rpc suite.  We will still need to write scripts that generate dummy accounts and dummy transactions.  A big part of the work is right here.

Edit: I just checked the graphene repository.  I am afraid there is no such transaction test script/program available yet.  We need to look for load test input and capture tool and use it to write our input scripts.

Fully agreed with this argument.