White Papers

Dell HPC NFS Storage Solution - High Availability Configurations
Page 4
1. Introduction
Clusters have become one of the most popular architectures for High Performance Computing (HPC)
today.
(1)
The disparity between the time taken by the processor and network to read or write data and
the slower speed of storage devices makes the storage subsystem of a cluster, in most cases, a
bottleneck that negatively impacts the overall system performance. Furthermore, cost-effective
clusters are usually built with commodity servers and storage devices and component failures or faults
cannot be completely avoided in such systems. Maintaining high levels of reliability and availability in
such an environment is an important issue. A well-designed cluster must balance computational and
I/O needs and take reliability and availability factors into account as well.
Parallel file systems have gained popularity on high performance clusters, especially for clusters that
require high end storage (>300 TB) and have intensive I/O demands. This requirement is met by the
Dell | Terascala HPC Storage Solution (DT-HSS) that offers a high throughput scale-out storage
appliance based on the Lustre file system and Dell PowerVault storage arrays.
(2)
For HPC users who do
not have intensive I/O workloads with MPI-I/O oriented applications and the need for high scalability,
the Dell NFS Storage Solution (NSS) is an attractive option. The NSS solution uses the NFS file system on
top of the Red Hat Scalable File System (XFS) with Dell PowerVault storage to provide an easy to
manage, reliable, and cost-effective solution for unstructured data.
(3)
With this solution guide, Dell has updated the current NSS offerings to include highly available options
(referred to as NSS-HA configurations in the rest of this document). The goal is to improve on the
service availability and data integrity of NSS configurations in the presence of possible failures or
faults, and to minimize the performance loss in the failure free case. For example, when a cluster
component fails, the impact to cluster clients is minimized and both system availability and user data
integrity are maintained.
This solution guide provides details on such NSS-HA solution offerings. The following sections describe
the NSS-HA architecture in detail with special focus on the high availability capability. Subsequent
sections discuss the performance of the NSS-HA solution offerings. An extensive appendix with detailed
steps on configuring an NSS-HA environment is included towards the end of this document.
The Dell NFS Storage Solution is delivered as an all-inclusive storage solution and is available with
deployment services and full hardware and software support from Dell. The solution is designed using
standards-based HPC components and focus on ease of deployment and ease of use.
Technical details of the solution are provided in subsequent sections of this solution guide while a
summary of the NSS options available from Dell is provided below.
1.1. NSS Solution Offerings From Dell
Dell HPC NFS Storage Solution is available in the standard configurations listed below. Contact a
Dell Sales Representative to discuss which offering works best in your environment, and the
ordering process. Customers can order any of the pre-configured solutions or customization can be
performed to fit specific needs. Based on the customization selected, some of the best practices
discussed in this document may not apply.
Details of the offerings without HA are covered in the solution guide “Dell NFS Storage Solution for
HPC (NSS)”. Details of NSS offerings with HA are provided in this solution guide.