New York University Faculty of Arts and Science College of Arts and Science Graduate School of Arts and Science

Compute Servers

Courant has a number of systems that can be used for CPU intensive jobs and programming assignments for courses. These machines are restricted to logins from within the Courant network, so if you are coming from outside of CIMS, you will have to first login to one of the remote access servers, then use ssh to get to them.

Please do not use access.cims.nyu.edu for cpu-intensive jobs or course work.

General Compute Nodes (See current status)

 These machines have been allocated for generic number smashing and medium memory consumption:

Hostname CPU Memory SSH Fingerprint
mauler.cims.nyu.edu 4 Core 2.6GHz AMD Opteron 8GB (7GB/user limit) Verify Host Identity
brawler.cims.nyu.edu 4 Core 2.6GHz AMD Opteron 8GB (7GB/user limit) Verify Host Identity

 

 

 These energy efficient cpu powerplants are for general compution with small memory footprints:

Hostname CPU Memory SSH Fingerprint
energon1.cims.nyu.edu 8 Core 1.8GHz Intel Xeon 8GB (4GB/user limit) Verify Host Identity
energon2.cims.nyu.edu 8 Core 1.8GHz Intel Xeon 8GB (4GB/user limit) Verify Host Identity
energon3.cims.nyu.edu 8 Core 1.8GHz Intel Xeon 8GB (4GB/user limit) Verify Host Identity
energon4.cims.nyu.edu 8 Core 1.8GHz Intel Xeon 8GB (4GB/user limit) Verify Host Identity

 

 

These Sun Solaris systems should be used for class programming assignments.  For mysql database related usage, see the creating a MySQL database instructions:

Hostname CPU Memory SSH Fingerprint
courses1.cs.nyu.edu Dual UltraSparc IIIi 2GB Verify Host Identity

 

The following Linux systems should be used only for CPU and memory intensive processes:

Hostname CPU Memory SSH Fingerprint
crunchy1.cims.nyu.edu 4 16-Core 2.1GHz AMD Opteron 6272 256GB Verify Host Identity
crunchy2.cims.nyu.edu 4 16-Core 2.1GHz AMD Opteron 6272 256GB Verify Host Identity
crunchy3.cims.nyu.edu 4 8-Core 2.4GHz AMD Opteron 6136 128GB Verify Host Identity
crunchy4.cims.nyu.edu 4 8-Core 2.4GHz AMD Opteron 6136 128GB Verify Host Identity
crunchy5.cims.nyu.edu 4 16-Core 2.1GHz AMD Opteron 6272 256GB Verify Host Identity
crunchy6.cims.nyu.edu 4 16-Core 2.1GHz AMD Opteron 6272 256GB Verify Host Identity

 

 

The following Solaris systems should be used only for CPU and memory intensive processes:

Hostname CPU Memory SSH Fingerprint
crunchy12.cims.nyu.edu 4x900MHz UltraSparcIII 16GB Verify Host Identity

Programs that use the Message Passing Interface (MPI) library can be compiled and run on the Crunchy HPC Cluster.

 

Specialty Nodes

Cluster Computing (See current status)

The EULER cluster is a small Linux cluster for submitting batch jobs and gridMathematica processes.

Web Service Computing

 The following machines are good for both internal and external web based development and deployment: 

Hostname CPU Memory SSH Fingerprint
linserv1.cims.nyu.edu 2 Core 1GHz AMD Opteron 2GB Verify Host Identity

 

We have some general instructions for setting up an Apache Webserver, Apache Tomcat, MySQL, and other services.

 

GPU Computing - NVIDIA/CUDA (See current status)

A pair of Tesla S1070 systems that can run CUDA programs are available for parallel processing. Each Tesla unit has 4 Tesla S10 GPUs. Each GPU has 240 streaming cores and 4GB of memory for a total of 960 processor cores and 16GB of memory. One Tesla unit is connected to each of energon3 and energon4. To use these systems please see the instructions.  Host nodes are:

Hostname CPU Memory SSH Fingerprint
cuda1.cims.nyu.edu 16 CPU cores, 2 Tesla S10 GPUs (480 streaming cores total) 16GB System Memory, 4GB Video Memory Verify Host Identity
cuda2.cims.nyu.edu 16 CPU cores, 2 Tesla S10 GPUs (480 streaming cores total) 16GB System Memory, 4GB Video Memory Verify Host Identity
cuda3.cims.nyu.edu 16 CPU cores, 2 Tesla S10 GPUs (480 streaming cores total) 16GB System Memory, 4GB Video Memory  Verify Host Identity
cuda4.cims.nyu.edu 16 CPU cores, 2 Tesla S10 GPUs (480 streaming cores total) 16GB System Memory, 4GB Video Memory  Verify Host Identity

 

GPU Computing - AMD/OpenCL (See current status)

A pair of OpenCL systems that supports AMD's Accellerated Parallel Processing (AMD APP). Each OpenCL machines has 2 FirePro V7800 GPUs.  Each GPU has 1440 streaming cores and 2GB of video memory. To use these systems please see the instructions.  Host nodes are:

 

Hostname CPU Memory SSH Fingerprint
opencl1.cims.nyu.edu 4 CPU cores, 2 AMD FirePro V7800s (2880 streaming cores) 6GB System Memory, 4GB Video Memory Verify Host Identity
opencl2.cims.nyu.edu 4 CPU cores, 2 AMD FirePro V7800s (2880 streaming cores) 6GB System Memory, 4GB Video Memory Verify Host Identity

 

 

Virtual Architecture

Courant als provides a number of dedicated virtual nodes for some specialty computing needs.  The virtual architecture is currently comprised of:

  • Self-Managed Compute Nodes (see virtual nodes for more information)
  • Public Cloud Computing (see virtual nodes for more information)
  • Private Cloud Computing (see virtual nodes for more information)

You can request your own private computing resources by submitting an online request.

 

Legacy Computing

 We currently provide some legacy computing for use with software that will only run on legcay systems:

Hostname Legacy Support CPU Memory SSH Fingerprint
stratus-1-11.cims.nyu.edu RHEL5 6 Core 12GB Verify Host Identity
stratus-2-11.cims.nyu.edu RHEL5 6 Core 12GB Verify Host Identity
stratus-3-11.cims.nyu.edu RHEL5 6 Core 12GB Verify Host Identity
stratus-4-11.cims.nyu.edu RHEL5 6 Core 12GB Verify Host Identity

 

Other Machines

There are also systems associated with research groups. They are restricted to users who have been approved by a faculty member associated with the relevant research group.

One such group of systems belongs to the Center for Atmosphere Ocean Science (CAOS).