Cisco Blogs


Cisco Blog > High Performance Computing Networking

Ultra low latency Ethernet (UCS “usNIC”): questions and answers

July 17, 2013 at 5:00 am PST

I have previously written a few details about our upcoming ultra low latency solution for High Performance Computing (HPC).  Since my last blog post, a few of you sent me emails asking for more technical details about it.

So let’s just put it all out there.

Read More »

Tags: , , , , ,

Why MPI is Good for You (part 3)

June 24, 2013 at 1:05 pm PST

I’ve previously posted on “Why MPI is Good for You” (blog tag: why-mpi-is-good-for-you).  The short version is that it hides the typical application programmer from lots and lots of underlying network stuff; stuff that they really, really don’t want to be involved in.

Here’s another case study…

Cisco’s upcoming ultra-low latency MPI transport is implemented over an “unreliable” transport: raw Ethernet L2 frames. For latency reasons, it’s using the OpenFabrics verbs operating-system bypass API. These two facts mean that a) userspace is directly talking to the NIC hardware, and b) we don’t have a driver thread running down in the kernel that can service incoming frames regardless of what the MPI application is doing.

Read More »

Tags: , , , , ,

Cisco’s Philosophy on Open Source

May 27, 2013 at 4:00 am PST

Last weekend, I was fortunate enough to be able to attend the Midwest Open Source Software Conference (MOSSCon 2013).  I met some fascinating people, listened to some great talks, and learned a bunch of new things.

All in all, a win.

I also presented a talk on two things:

  1. The general open source philosophy at Cisco
  2. My specific open source work at Cisco

The slides that I presented are below (slightly edited from their original form; I used a few animations in my original slides, which don’t work on Slideshare):
Read More »

Tags: , , , , , ,

Cisco ultra low latency support for MPI

November 16, 2012 at 9:42 am PST

My team demonstrated our new ultra-low latency Ethernet solution in the Cisco booth at SC this past week (it was so busy that I didn’t get to post this until it was all over!).

The short version is that we have implemented operating system bypass and NIC hardware offload via the Linux OpenFabrics verbs API stack. We call it “userspace NIC”, or “USNIC”.

But let’s cut to the chase — what’s the performance?  Let’s break it down:

Read More »

Tags: , , , ,