-2

I don't know if this is a good place to beg for help, but here it goes: basically, I need to run a secure recommender system simulation (C++ console application) in order to meet tonight's deadline, and the faculty's server grid decided to go offline. I could really use something like 10+ (actually, about 16 would be required to meet the deadline) virtual instances of some Linux that has GMP installed... Ideally, they should all have the same specs, because a part of the simulation will represent performance benchmarks.

If my question is inappropriate in any way, I kindly ask the administrators to remove it.

Mihai Todor
  • 222
  • 1
  • 3
  • 13

1 Answers1

4

Amazon Elastic Computing Cloud is what you're looking for. Those are relatively cheap servers rent on demand.

32 Extra Large servers for 8h will cost you a bit under $164. Those are 15 GB, 4 core, 8 thread machines with 1690GB of storage.

If you don't need something as big, there are machines from 1 core, 1 thread, 1.7GB or RAM up. There are also memory and CPU optimised instances.

If you want to have an already configured cluster machine, Quadruple Extra Large machines (23 GB memory, 33.5 cores) cost $1.3 per hour...

Hubert Kario
  • 6,361
  • 6
  • 36
  • 65