Leonhard

From ScientificComputing
Revision as of 12:28, 10 January 2017 by Byrdeo (talk | contribs) (Compute nodes)

Jump to: navigation, search

Introduction

Leonhard is a new cluster designed for “big data” applications. This cluster is financed jointly by a small group of shareholders and IT Services, and by a special grant from the Executive Board of ETH.

As the name implies, Leonhard will be closely related to Euler. It will be operated by the same group and will offer the same software environment. One half of the cluster will be open to all users, like Euler; whereas the other half will be reserved for sensitive applications, like personalized medicine. In general, the name Leonhard refers to the whole cluster, and its two halves are called respectively Leonhard Open and Leonhard Med. For security reasons, each half will have its own dedicated storage, network and login nodes.

Life time

The cluster's components have been ordered in September 2016 and will be delivered at the end of the year. The system will be installed in January 2017. We expect that it will be open to beta users in Q1 and to all users in Q2.

Storage

Each part of Leonhard will feature high-performance parallel file system — based on GPFS — with a capacity of 2.0 PB for Leonhard Open and 1.5 PB for Leonhard Med (1 PB = 10^15 bytes).

Compute nodes

Leonhard will contain three types of compute nodes:

  • Standard nodes equipped with two 18-core Intel Xeon E5-2697v4 processors and 128 GB of memory
  • Large-memory nodes equipped with two 18-core Intel Xeon E5-2697v4 processors and 512 GB of memory
  • GPU nodes equipped with two 10-core CPUs, 256 GB of memory and eight Nvidia GTX 1080 GPUs (specifications subject to change)

Network

All compute nodes will be connected together and to the cluster's parallel file systems via a 100 Gb/s InfiniBand EDR network.