Virtual Computing Lab (VCL)

Introduction

The Virtual Computing Lab (VCL) is a system that enables faculty, students, and staff to remotely access specialized software on any computer. If you are a Mac user and need to run Windows software, no problem, the VCL makes this possible.

 

Instructions

Instructions: 

Please refer to the related instructions on the Academic Technology Services page.

Service Categories

Service Categories: 
385494

Audience

Audience: 
Students
Faculty
Staff
Five College Students

Tags

Tags: 

Stata/MP in the Virtual Computing Lab

Introduction

Stata/MP is a special multi-processor-aware version of Stata, that can be used for processing large datasets. This software is available to a limited group of faculty and students. If you would like to be added to this group, please contact the Help Desk.

In addition, this software is made available in a private cloud-computing environment, hosted at Amherst College. This means that the compute environment is dynamically provisioned based on need, rather than running at all times on a single shared server. This also means that when your "reservation" (or "session") has ended, you will need to make a new reservation in order to continue your work. It is possible to extend reservations for several hours beyond the initial request

Please note: any data stored directly on the Stata/MP server during this reservation will be erased after the session ends. Given that, a persistent datastore is provided where data can be stored and accessed from within the Stata/MP application. It is recommended that users place large datasets directly on this data server rather than loading data directly onto the Stata/MP server.

 

Instructions

Instructions: 

Initial Setup

Depending on how you plan to use Stata/MP, you can interact with the application either entirely through the command line (perhaps best for running batch scripts) or you can use a full graphical interface (perhaps the more familiar option). Some initial setup may be required, which is detailed below.

Mac Users

In order to use Stata's graphical interface, you will need to install an X-Windows environment. This can be done by going to the XQuartz website and downloading the software available there.

Windows Users

All Windows users will need to install PuTTY. There are several download options; the MSI installer is recommended. That download link will look similar to this: putty-0.xx-installer.msi

Any Windows users who plan to use Stata's graphical interface will also need to install an X-Windows environment. XMing is the recommended option for this:

  1. Go to https://sourceforge.net/projects/xming
  2. Download Xming by clicking on the large green download link. Make sure you are downloading a file whose name looks like Xming-x-x-x-x-setup.exe
  3. Install Xming by running the setup executable. The default options will suffice.

Reserving a Stata/MP environment

There are several ways to access and use the Stata/MP software. All of them require first logging into the Virtual Computing Lab and following these steps:

  1. Navigate to Reservations -> New Reservation, and select "Stata/MP" from the drop-drown list.
  2. Select the duration of the reservation (default = 1 hour), and click "Create Reservation"
  3. When the reservation is ready, click the "Connect!" button

Look for where the page reads "Connect to reservation using ssh on port 22". You will use that information below.

Note, also, in the box below that command, under "Connection information". If the page says "Use your campus password", you will use that to connect; otherwise, a one-time password will be generated for you and printed on the screen. 

Connecting to a Stata/MP environment

Mac Users

All connections to Stata/MP will begin by launching the Terminal application (Applications -> Utilities -> Terminal) and using ssh. The ssh command can be simply copied from the displayed web page.

If you plan to use the simple command-line version of Stata/MP, you can launch it by typing the command: stata-mp.

If you plan to use the full graphical interface, you can launch that by typing the command: xstata-mp.

Windows Users

All connections to Stata/MP will begin by launching the PuTTY application. You will need to provide the host address to PuTTY, which will be available on the displayed webpage. You will be propmted automatically for your username and password. If you plan to use Stata's graphical user interface, you will need to also enable X11 forwarding. You do this by looking on the left side of the PuTTY window, under Connection; open the SSH option and select X11. There, click on "Enable X11 forwarding".

If you plan to use the simple command-line version of Stata/MP, you can launch that by typing the command: stata-mp.

If you plan to use the full graphical interface, you can launch that by typing the command: xstata-mp.

Accessing Persistent Data

For large datasets, it is important to store these in a persistent location so that they can be accessed and/or modified across multiple Stata/MP sessions.

This data store is made available automatically when a user accesses a Stata/MP server. The data can be accessed from a user's home directory in one of two directories: ./data or ./shared. The ./data directory is readable and writable only by the user with the reservation. The ./shared data directory is readable and writable by everyone.

This data directory can also be accessed from outside a VCL reservation. For instance, to upload data from your own computer before creating a session or downloading data after the session has ended.

To access this network share, you will need to use an SCP or SFTP client to copy data to/from your home directory on vcl-data.ats.amherst.edu. Mac users can use the SCP or SFTP clients that come pre-installed; they can be accessed with the Terminal application. Windows users can use PSCP or PSFTP; if you installed the PuTTY MSI as described above, then these programs are already available.

If you need assistance with any of these steps, please contact the Help Desk.

Service Categories

Service Categories: 
385494

Audience

Audience: 
Students
Faculty

Tags