I want to setup a shared data source for my blog between 2 Windows Amazon AWS EC2 instances. I don't want to share a folder off of either server, because I want to be able to stand up new instances if needed and have them point to the same shared data source.

Is there a way that I can create a shared volume of data and point IIS to it from each EC2 instance?

Would mapping a drive to S3 using tntdrive work smoothly?

Has anyone used DropBox to host static IIS websites out of from EC2 instances?


  • While researching for a similar solution and finding out AWS EFS is not "officially" supported for Windows EC2 instances (bummer), I came across another solution that leverages S3 as a backend: AWS Storage Gateway (VMs) and Windows' native iSCSI support. – Roberto Andrade Apr 13 '18 at 18:45

I would recommend using Logmein Hamachi and mapped drives as this would be the simplest way to connect the machines without a lot of overhead and configuration. Hamachi also encrypts traffic so the transfers should be secure.

| improve this answer | |
  • Hi Mike, thanks for the suggestion. However, wouldn't the blog files still have to be stored on one of the servers and then have the rest of the servers mapped to the single server? I don't want to create a single point of failure, which is why I was thinking of using S3 as the data source. – Aaron Schnieder Feb 5 '14 at 19:43
  • I'm not familiar with S3's functionality but if it allows you to map a drive then that could work as well. I've used Hamachi in the past very successfully but you are correct, it would involve mapping to a single server. To work around the this you could use Hamachi and a file/folder synch application to synchronize files between the servers. ie. Server A has a mapped drive to it's duplicate on Server B and synchronizes on a schedule. Server B points to Server A as the primary data source but if A goes down the paths on B can be updated to point to the duplicate folder. – Mike Naylor Feb 5 '14 at 19:48
  • Ah, thanks for the further explanation Mike. I will give that a try. – Aaron Schnieder Feb 5 '14 at 22:38

You can used s3 as a shared data source, although it's not a true file system.

While you can mount an s3 bucket so it looks like a disk it won't have the same performance.

Blogs, (like wordpress for instance) do have plugins where you can hook directly into an s3 bucket through their APIs.

Alternatively you can write your own hook-ins to S3 via their API.

| improve this answer | |

I would launch a windows server 2008 R2 instance and set it up as a file server.

You could change your security group, so only instances that are launched within your subnet can communicate with your file server.

| improve this answer | |

There is an S3 mapping application that does a good job and is fast. You can also use a VPC for your hosts and share in the VPC address range

| improve this answer | |
  • 4
    This answer would be more useful with a link to the mapping application. – Andrew Schulman Feb 5 '14 at 21:15

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.