The New Stern Cloud Computing Environment

July 29, 2008

The Stern Center for Research Computing is proud to announce the availability of a new set of capabilities based on "Cloud Computing". This new environment will also us to rapidly provision new machines for special puposes, while doing so at a much lower cost than previously. The Stern Research Computing Cloud (SRCC) will initially consist of 5 physical servers connected to approximately 30,000 GB (30 TB) of shared disk. Each server will be able to support many virtual machines, running linux, Windows XP, Windows Server 2008, Windows Vista, Windows 7, ...

The virtual machines will be able to be accessed from anywhere on the internet. Windows users will connect to the machines using Windows Remote Desktop. In most cases, the machine will feel faster than their local machine. CPU, memory, disk upgrades can be done almost immediately. The machines will automatically be migrated to faster facilities as they become available, with NO work necessary by the faculty member. Machines can be moved from one server to another, WHILE they are running. This allows upgrades to the environment to be done with no disruption to users.

We already have a number of machines being piloted in this new environment, including VLAB, the Volatility Institute web server, and several special purpose faculty servers. We will soon be moving the matlab, splus, and sas license serves to this new environment. Eventually, we hope to have an off-site server, so we can keep vital services running even if there is a problem in the facility at 44 West 4th Street. Stern IT recently did this during the recent power outage.

Faculty who need access to these facilities should contact Norman White (nwhite@stern.nyu.edu). There will likely eventually be some type of charge, either to Stars or to grant money for machines that use significant resources, but in any scenario, the charge will be significantly less than purchasing the equivalent capability for a standalone machine, and there will be much less hassle in ordering, configuring, maintaining, and disposing of it.