Cisco Blogs

Cisco Blog > High Performance Computing Networking

Speaking at Lawrence Berkeley National Lab next week

Are you in the Northern California Bay Area and want to hear about Open MPI and/or Cisco’s usNIC technology next week?

If so, you’re in luck!

I’ll be speaking at Lawrence Berkeley Lab (LBL) next Thursday, November 7, 2013, at 2:30pm.  Click through to see the location and directions and whatnot (LBL requests that you RSVP if you plan to attend).

Read More »

Tags: , , ,

EuroMPI’13 Cisco slides: UCS, Nexus, usNIC

A few people asked me to post the slides that I just presented in the Cisco vendor session at EuroMPI’13.  In short, I gave a brief overview of our servers and switches, and then some technical details of how we use SR-IOV in our usNIC, etc.

Here’s the slides: Read More »

Tags: , ,

usNIC released!

I’m excited to announce that Cisco has just released usNIC as a feature of the UCS C-Series Rack Servers product line.

usNIC is available since the release 1.5(2) of the Cisco UCS C-Series Integrated Management Controller.

Read More »

Tags: , ,

Why is SR-IOV relevant in the HPC world ?

One feature of the usNIC ultra-low latency Ethernet solution for the UCS Cisco VIC that we think is interesting is the fact that it is based on SR-IOV.

What is SR-IOV, and why is it relevant in the HPC world?

SR-IOV (Single Root I/O Virtualization) is commonly used in the server virtualization world. The most commonly described purpose of SR-IOV in the hypervisor world is to allow a device partition, called VF (Virtual Function), to be mapped in the guest operating system address space. This allows the guest operating system to enjoy higher I/O performance and lower CPU utilization as compared to the alternative: software-emulated devices that are traditionally implemented in hypervisors.

Compared to the old world before hypervisors came along, that use of SR-IOV seems to allow to regain back some performance lost due to the hypervisor software intervention in the I/O data path. But why should I care about SR-IOV in the world of my network-latency-bound HPC applications running on common operating systems on bare metal servers?

Read More »

Tags: , , , , ,

Short message latency and NUMA effects

I’ve previously written a bunch about the effects of location, Location, LOCATION! on MPI applications.

Here’s another subtle NUMA effect that a well-tuned MPI implementation can hide from you: intelligently distributing traffic between multiple network interfaces.

Yeah, yeah, most MPI implementations have had so-called “multi-rail” support for a long time (i.e., using multiple network interfaces for MPI traffic).  But there’s more to it than that.

Read More »

Tags: , , , , ,