Super User is a question and answer site for computer enthusiasts and power users. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

So, i'm looking at amazons high-cpu on demand instances, they have an 8 virtual core extra large high-cpu instance, but it only has 8GB of RAM.

I'm trying to offload a load of circuit simulations that I've written, however they're all only single threaded, so I can run 8 different simulations at once, on 8 cores, but they each consume 2GB of ram each for the duration.

So is there anyway to purchase more memory for that instance? or would it be better to use the 'Cluster Compute Quadruple Extra Large Instance', which although expensive already has the necessary amount of ram.


share|improve this question

closed as off topic by squircle, random Jul 4 '11 at 22:51

Questions on Super User are expected to relate to computer software or computer hardware within the scope defined by the community. Consider editing the question or leaving comments for improvement if you believe the question can be reworded to fit within the scope. Read more about reopening questions here.If this question can be reworded to fit the rules in the help center, please edit the question.

Do you know that each process uses 2gb of dedicated memory? I'd imagine that a lot of memory can be shared between these processes, since they're likely using the same shared libraries, etc.

If you have control over the application, and can refactor it slightly to fork 8 instances, rather than starting 8 instances individually, you can probably share even more memory, since Linux will only duplicate dirty pages in a forked child, meaning that even more memory can be shared in many cases.

Although maybe you've already considered all of this, and know that you truly need 2gb of dedicated memory per process.

share|improve this answer

I've looked for ways to expand memory capacity on EC2 and have yet to see a means of doing it, other than changing instance types. So, I think your only option without getting some custom help from AWS, is to use the next instance type up the list.

Alternatively, you may want to look at other VPS services. Linode, for example will let you purchase extra specific resources a la carte. Though you should talk to them first to see what their limits are, as your needs sound fairly industrial strength.

share|improve this answer
I already have a linode for a web server, what I'm looking at is a high cpu host for a large amount of calculations. Something which linode is not very suitable for, I'm looking more towards using the amazon cluster compute instance, but that leaves me with 13GB of free memory. oh well – Bob Jul 4 '11 at 17:50
Ahh, well, maybe someone else will have a helpful tip... Though I think you may have to work with what you have for now - maybe multiple instances would hack it for you? Good luck at any rate. – Doc Jul 4 '11 at 17:53

Not the answer you're looking for? Browse other questions tagged .