White Papers

Compute and Management Components

There are several considerations when selecting the servers for master node, login node, compute node, fat node and accelerator

node. For master and login node, 1U form factor PowerEdge R430 is recommended. Master node is responsible for managing the

compute nodes and optimizing the overall compute capacity. Login node is used for user access, compilations and job submissions.

Usually, master and login nodes are the only nodes that communicate with the outside world, and they act as a middle point between

the actual cluster and the outside network. For this reason, high availability is provided for master and login nodes in the example

solution illustrated in Figure 1.

Ideally, the compute nodes in a cluster should be as identical as possible, since the performance of parallel computation is bounded by

the slowest component in the cluster. Heterogeneous clusters do work, but it requires a careful execution to achieve the best

performance. We recommend PowerEdge C6320 as a compute node due to its density, a wide choice of CPUs and high maximum

memory capacity.

PowerEdge R930 is an optional node with up to 3TB of memory. This node is recommended for customers who need to run

applications requiring large memory.

Accelerators are used to speed up computationally intensive applications, such as molecular dynamics simulation applications.

Although there are five different configurations for four NVIDIA K80 GPUs on PowerEdge C4130, we tested configuration C for this

solution.

The compute and management infrastructure consists of the following components.

o Compute

- Dell EMC PowerEdge C6320 rack server with 4 x C6320 servers and either Intel® OPA fabric or IB EDR fabric

- Dell EMC PowerEdge FX2 chassis with 8 x FC430 chassis each and IB FDR interconnect [1]

- Dell EMC PowerEdge C4130

- Dell EMC PowerEdge R930

o Management

- Dell EMC PowerEdge R430

Dell EMC PowerEdge C6320 for compute node

High-performance computing workloads, such as scientific simulations, seismic processing and data analytics, rely on compute

performance, memory bandwidth and overall server efficiency to reduce processing time and data center costs. The next-generation

Dell EMC PowerEdge C6320 provides an optimized compute and storage platform for HPC and scale-out workloads with up to four

independent two-socket servers with flexible 24 x 2.5” or 12 x 3.5” high capacity storage in a compact 2U shared infrastructure platform

[5]. C6320 supports up to 512GB of memory per server node, for a total of 2TB of memory in a highly dense and modular 2U solution.

Dell EMC PowerEdge C4130 for accelerator node

The PowerEdge C4130 provides supercomputing agility and performance in an ultra-dense platform purpose-built for scale-out HPC

workloads [6]. Speed through the most complex research, simulation and visualization problems in medicine, finance, energy

exploration, and related fields without compromising on versatility or data center space. Get results faster with greater precision by

combining up to two Intel® Xeon® E5-2600 v3 processors and up to four 300W dual-width PCIe accelerators in each C4130 server.

Support for an array of NVIDIA® Tesla™ GPUs and Intel Xeon Phi™ coprocessors, along with up to 256GB of DDR4 memory, gives

you ultimate control in matching your server architecture to your specific performance requirements. This server is an optional

component for molecular dynamics simulation applications.

Dell EMC PowerEdge R930 for fat node

The Dell EMC PowerEdge R930 is a 4-socket, 4U platform, equipped with the Intel® Xeon® E7-8890 v4 (24 cores per socket – 96

cores in server) and is dubbed “the fat node,” because of its 6 TB of memory capacity. This processor has 3 QPI links per socket. The

particular features are needed for de novo genome assembly application like Velvet. Hosting large genomic data sets in memory with

64 cores operating on it eliminates the overhead caused by interconnects, disk look-ups, and swapping, thereby resulting in decreased

run time. This server is an optional component that can be added to the solution.