Skip to main content

Hardware Overview

Our members enjoy access to several resources, including:

Testbed Systems

Overview

The Gnosis Research Center (GRC) Lab manages several cluster computers to support the group's research. Cluster resources within the lab are controlled via a batch queuing system that coordinates all jobs running on the clusters. The nodes should not be accessed directly, as the scheduler will allocate resources such as CPU, Memory and Storage exclusively to each job.

Once you have access to use the cluster, you can submit, monitor, and cancel jobs from the head nodes, ares.cs.iit.edu and hec.cs.iit.edu. These two nodes should not be used for any compute-intensive work, however you can get a shell on a compute node simply by starting an interactive job. You can use the cluster by starting batch jobs or interactive jobs. Interactive jobs give you access to a shell on one of the nodes, from which you can execute commands by hand, whereas batch jobs run from a given shell script in the background and automatically terminate when finished.

If you encounter any problems using the cluster, please send us a request via grc@iit.edu and be as specific as you can when describing your issue.

Usage Policy

Regular members of the GRC enjoy access to the resources. If you wish to gain access to the cluster and you do not belong to the core team, please submit and request via grc@iit.edu and state the following: (i) your CS login ID, (ii) name of professor you're working with (and put him under cc on the form) (iii) reasons for requesting access (i.e., research project description) (iv) projected time period for which you would need access (v) resources that you may interfere other uses significantly (e.g., global file system, network) (vi) commands that you need to run as root privilege.

If we have any trouble with your job, we will try to get in touch with you but we reserve the right to kill your jobs at any time. If you have questions about the cluster, send us a request at grc@iit.edu.

Local Computing Resources

The GRC manages two cluster computers, Ares and HEC, each for different research scope. Our flagship cluster is Ares with 1576 cores and a 30TFLOPs peak performance. HEC is a smaller 128 core machine that specializes to network research. All HEC nodes are connected with InfiniBand network powered by Mellanox InfiniHost III Ex adapters. You can find the detailed hardware configurations below.

Ares Cluster

The Ares cluster is composed of one rack of compute nodes. All the nodes share a 48TB RAID-5 storage pool comprised of eight 8TB 7200K SAS hard drives. Nodes in each rack are connected with 40Gbps Ethernet with RoCE support. One 200Gbps uplink connects two racks of nodes. The compute rack consists of one ThinkSystem SR650 master node, 32 ThinkSystem SR630 compute nodes. In total, the compute rack has 66 2.2GHz Xeon Scalable Silver 4114 processors with boosted frequency up to 3.0GHz, which leads to 660 cores and 1320 threads. The master node and the compute nodes are equipped with 96GB and 48GB DDR4-2400 memory, 128GB and 32GB M.2 SSD for OS, respectively. 24 compute nodes are equipped with one 250GB M.2 Samsung 960 Evo SSD. The other eight are equipped with one 256 GB M.2 Toshiba RD400 SSD.

Read more in the Ares User Guide

HEC Cluster

The HEC cluster is comprised by 16 Sun Fire X2200 nodes and one Sun Fire X4240 head node. This mini cluster is ideal for InfiniBand-related tests. All nodes are connected with InfiniBand network powered by Mellanox InfiniHost III Ex adapters. The 5TB RAID-5 storage pool provides the global storage for all the nodes. The X4240 node, serving as the master node, is equipped with two quad-core 2.7GHz Opteron 2384 processors, 32GB DDR2-667 memory and two 250GB Samsung 860 Evo SATA SSDs in RAID-1 configuration. The X2200 nodes, serving as the compute nodes, are equipped with two quad-core Opteron 2376 processors running at 2.3GHz, 16GB DDR2-667 memory, one 100GB OCZ RevoDrive X2 PCIe SSD, and one 1TB Seagate 7200K SATA hard drive.

External Computing Resources

We also have access to the Chameleon Cloud platform. It is consisted with two clusters located in Texas Advanced Computing Center (TACC) at Texas and University of Chicago. It has 338 compute nodes connected with 10Gbps Ethernet network. Among all compute nodes, 41 of them are connected via InfiniBand as well. Each compute node has four 6-core (12 threads) Intel Xeon E5-2670 v3 "Haswell" processors and 128GiB RAM. There are also 24 storage nodes with 16 2TB hard drives and 20 GPU nodes. In total, the Chameleon Cloud platform has 13,056 cores, 66 TiB of RAM, and 1.5PB of configurable storage.

Conferences

Journals