Facilities

The Parallel Distributed Systems Laboratory, located in LW-836, is comprised of a heterogeneous computing environment which includes a Multi-Core Cluster, Intel Xeon and Xeon Phi Coprocessors, Dozer (64 cores) and 25  Linux-based PCs and many other equipments. These systems are used by faculty, staff and students at Department of Computer Science & Engineering for teaching and studying related courses and for research project's lab members (refer Eligibilty for Lab Access )

!!! CSCI-5593 and CSCI-5551/7551 Class Website here

Heracles Multi-Core Cluster

Cluster

The Multi-core cluster consists of following primary components:
  • Total of 18 nodes distributed as:
  • Mellanox SwitchX-2 18-Port QSFP FDR Externally Managed Switch (1U)
    • Non-Blocking Switch Capacity of 2Tb/s
  • 256GB Total Memory @ 3200 MHz (16 x 16GB DDR4 3200 MHz ECC/Registered Memory)

 Learn more

Xeon and Xeon Phi Coprocessors

Cluster

Intel Xeon Phi coprocessors have been designed by Intel Corporation as a supplement to the Intel Xeon processor family. These computing accelerators feature the MIC (Many Integrated Core) architecture, which enables fast and energy-efficient execution of High Performance Computing (HPC) applications utilizing massive thread parallelism, vector arithmetics and streamlined memory access. The term “Many Integrated Core” serves to distinguish the Intel Xeon Phi product family from the “Multi-Core” family of Intel Xeon processors.

 Learn more

Dozer (64 Core Machine)

  • Four AMD Opteron 6274 processors, 16 cores
  • Each processor chip:
  •        - 16 Core 2.2GHz / processor chip
           - 16x16KB (Data) + 8x64KB(Instruction) L1 Cache
           - 8x2MB L2
  • 2x8MB L3 Cache
  • Memory: 128 GB RAM
  • Storage: 1,862GB SATA
  • Operating System: Cent OS 7.0
  • NVIDIA Kepler GPU (K40c) 192CUDA cores/MP, 15 Multiprocessors for a total of 2880 CUDA cores

Other facilities

Cluster The Lab room, located at LW-836, is the place where faculty, staff and students study and research in various related fields. There are 25 Linux-based PCs to access the computing resources of the cluster. Learn more