HPCatDell home button

HPC cluster design requires a deep understanding of the trade-offs between price, performance, and energy efficiency. This blog shares the results of studies conducted by Dell's HPC engineering team that can help inform your cluster design decisions.


Start Blog Entry

Performance of the Dell PE C6100 and C410x based GPU Solution - NAMD


Link to Blog

Saeed Iqbal

Start Blog Entry

Performance of the Dell PE C6100 and C410x based GPU Solution


Link to Blog

Saeed Iqbal and Shawn Gao

Start Blog Entry

Performance of the Dell PE C6145 and C410x based GPU Solution


Link to Blog

Saeed Iqbal and Shawn Gao

Start Blog Entry

Dell NFS Storage Solution with High Availability - an overview

The Dell NFS Storage Solution (NSS) now comes in two flavors – with and without High Availability. This blog discusses the High Availability Solution introduced in April 2011.

Link to Blog

Xin Chen and Garima Kochhar

Start Blog Entry

Advantages of GPUDirect for HPC


Dell's Saeed Iqbal shows performance improvements due to GPUDirect on HSG Models.

Link to Blog

Saeed Iqbal

Start Blog Entry

Method for automated BIOS and BMC configuration of a Linux HPC for performance and manageability


Dell's Calvin Jacob describes a set of scripts that can be used to set HPCC specific BIOS options and configure the IPMI controller on each node.

Link to Blog

Calvin Jacob

Start Blog Entry

Dell™ PowerVault™ MD1200 Performance as a Network File System (NFS) Backend Storage Solution whitepaper

6/3/2010

This whitepaper builds a standards based HPC storage architecture over NFS and describes the performance characteristics of the PV MD1200 arrays when used with NFS in an HPC environment.

Link to Blog

Aziz Gulbeden

Start Blog Entry

Comparing Different PV MD1200 Connection Topologies

4/7/2010

The Dell PowerVault MD1200/1220 arrays provide a convenient way to expand the PowerEdge servers to gain access to more drives. The MD1200 and MD1220 models use the 6Gb SAS technology, and they can be daisy chained up to 4 units. This allows a maximum of 8 arrays to be connected to a single adapter since PERC H800 adapter has two ports. When one has four MD1200 arrays, there are a number of different ways that these arrays can be connected to the PERC H800 adapter. The goal of this post is to study the performance of three possible topologies under a workload where a number of threads perform sequential write and read operations. This post is an intermediate study and the results will be used to configure an NFS server to share MD1200 storage to an HPC cluster through NFS.

Link to Blog

Aziz


Start Blog Entry

A Comparison of Ext2, Ext3 and Ext4 Performance

3/24/2010

The goal of this post is to study the sequential read and write performance that one can get from a RAID array using various local file systems available in Linux, specifically the ext2, ext3 and ext4 file systems. Ext3 file system adds journal feature to ext2 for better reliability. Ext4 is an incremental update to ext3, it has new features such as being able to create up to 1 exabyte volumes and up to 16 terabyte files. It introduces further performance related enhancements such as using checksums in the journal, using extents (rather than block maps) and delayed block allocation that improve large file performance.

Link to Blog

Aziz


Start Blog Entry

CASC -- HPC 20 years from now

10/26/2009 -

The Coalition for Academic Supercomputing (CASC) is an educational nonprofit dedicated to advocating the use of advanced computing technology to accelerate scientific discovery. Last September, CASC had its 20th anniversary meeting. Academic and industry leaders gathered in Washington DC to discuss the impact of HPC on science and engineering. Dell Global Solutions Engineering director Reza Rooholamini participated in a panel discussion entitled "HPC: The Next 20 years."

Link to Blog

Jacob


Start Blog Entry

Impact of C1E on PowerEdge 11G Servers

10/9/2009 -

C1E (Enhanced halt state) is a new BIOS option which is added to 11G Servers with the latest BIOS version 1.2.6. Since all other BIOS options (Turbo mode, C-states and various power management profiles) are already covered in the whitepaper “Optimal BIOS Settings for High Performance Computing (HPC) with PowerEdge 11G Servers” published by the HPCC Engineering team, I just wanted to use this space to talk about some of the details on C1E. I will also be covering the impact of C1E on performance/power consumption based on a number of HPC applications.

Link to Blog

Toby

Start Blog Entry

iSCSI for HPC
9/8/2009 -


Last week the Dell HPC engineering team presented a poster at the IEEE Cluster 2009 conference in New Orleans describing iSCSI I/O subsystems for HPC. Commonly used in Enterprise computing, the HPC community has been slow to adopt iSCSI storage for clusters. This paper describes two use cases for iSCSI based I/O subsystems for HPC clusters. The first approach -- called an indirect SAN -- has servers re-exporting iSCSI storage via a traditional parallel file system. With the second approach -- called a direct SAN -- the compute nodes attach directly to a clustered file system on the iSCSI storage via software iSCSI initiators.

Link to Blog

Jacob


Start Blog Entry

HPCC in ParadiseJacob Liberman at Le Perouse Bay, Maui
8/26/2009 -


In Polynesian culture, mana means life force or power. To say that someone has mana means that they have influence or authority. The Maui High Performance Computing Center increased its mana with a new 1,152 node cluster of PowerEdge 11G servers. The compute nodes are M610 blades connected together via an Infiniband DDR fabric. To my knowledge, this is the largest PowerEdge 11G cluster deployed by Dell to date.

I was lucky enough to help deploy and benchmark Mana. I describe some impressions in this post.

Link to Blog

Jacob

Start Blog Entry

BIOS Optimization for HPC on 11G
8/25/2009 -


Dell's PowerEdge 11G servers -- based on Intel's Nehalem micro-architecture -- include new features intended to improve performance and energy efficiency across a wide range of server workloads. Dell's HPCC engineering team recently published a whitepaper that describes optimal BIOS settings for High Performance Computing.

Link to Blog

Jacob


Start Blog Entry

Memory Wall
7/08/2009 -


In 1994, Wulf and McKee predicted that computer performance improvement would soon cease. They reasoned that CPUs were becoming too fast for their own good. CPUs would soon become so much faster than memory that program execution time would depend entirely on how fast the memory could feed CPUs data. They dubbed this problem hitting the memory wall, a natural consequence of Moore’s Law.

John McCalpin (the inventor of the STREAM memory bandwidth benchmark) plotted the projected divergence on his web page...

Link to Blog

Jacob

End Blog Entry

11G -- As Good As Advertised?

7/02/2009 -

Last March, Dell released the 11th Generation PowerEdge servers based on Intel’s Nehalem micro-architecture. In my 10 years at Dell I can’t remember a product launch accompanied with such high expectations. Some of the performance projections I saw were frankly hard to believe. As the performance lead for Dell’s High Performance Computing (HPC) engineering team, I couldn’t wait to get these servers in my lab and try to bring them to their knees.

Link to Blog

Jacob