Cloud Computing

Cloud computing is now the dominant approach for deployment of scalable web services, and recent advancements in GPU virtualization technology now make it possible to use cloud computing for large-scale scientific computing, data analysis, and visualization tasks. Cloud computing involves the use of computational equipment provided by a third-party, accessed through an internet connection, and paid for on an on-demand, as-needed basis. The cloud computing model provides researchers with access to powerful computational equipment that would otherwise be too costly to procure and maintain on their own. Furthermore, structural modeling often involves the use of different software suites, and bundling them together on a cloud platform gauruntees their availablity and interoperability on a standardized system.

We have made MDFF available on the Elastic Compute Cloud (EC2) of Amazon Web Services (AWS) to highlight the platform's capability in providing easily accessible computational power sufficient for rapidly fitting structures to EM density with MDFF. EC2 provides a cost-effective and practical solution for many MDFF applications across a wide range of computational hardware.

Hardware details on the available instance types of EC2 can be found here, while pricing information can be found here. We recommend using one of the compute optimized instance types (c3 or c4) for MDFF simulations.

In order to access the EC2 implementation of MDFF, you will need to sign up for an AWS account. You should also read through Amazon's "Getting Started" Guide and the EC2 documentation.

After reading through the Amazon documentation, you can use our MDFF AMI (ami-XX) by searching for it on the public AMI list when launching an instance from the EC2 dashboard.

You can connect to your running instance through ssh, as described in the user's guide, with the user name 'ubuntu'. Alternatively, you can view the instance using a VNC client, such as NICE DCV Endstation. You can connect to the instance's public IP address on port 5901, using the password XX. In order to connect to your instance through ssh and VNC, make sure you have the appropriate ports open in your Security Group:
22 tcp 0.0.0.0/0
5900-5920 tcp 0.0.0.0/0

The MDFF AMI contains installations of the latest versions of VMD and NAMD. In addition, the MDFF tutorial files, including an example for using the beta release of the multi-copy "Resolution Exchange" MDFF (ReMDFF), are located in the home directory.