Skip to main content

Linux Foundation Open Networking Summit

The Open Networking Summit took place on April 3-6 – where Enterprise, Cloud and Service Providers gathered in Santa Clara, California to share insights, highlight innovation and discuss the future of open source networking. I was invited to give a talk about Web Virtual Reality and aframe at it.
So, Open Networking Summit (ONS) actually consists of two events – there might be more, but I was involved with two. ONS is the big event itself. There is also the Symposium on SDN Research (SOSR). This is an academic conference that accepts papers. 
There were some pretty fantastic papers at the conference. My favorite one – there was a system called “NEAt: Network Error Auto-Correct”. The idea here is that the system keeps track of what’s going on with your network and notices problems and automatically corrects them. It was designed for an SDN setup where you have a controller that is responding to changes in the network and telling systems what to do.

The event was held at San Jose Convention center and was pretty packed up. 
Keynotes were sprawled across the first floor with a very big auditorium that encompassed the whole of the first floor. The individual talks were assigned different rooms on the two floors.
Poster sessions were held on the second floor near another hall where the accompanying talks with the poster were going on.
The talks were not recorded. I had roughly 35 people in my talk and that was a pretty perfect number of audience to have without being too overwhelmed.
A previous version of my talk is available here. It would be great to have some feedback on it, though the content has changed quite a bit after that

I frankly received quite a lot of interest in the talk and questions regarding it. The questions mostly were involving authoring tool for WebVR and about how we can create scene's that can interact with industrial hardware.
Something that urged me to work on some pet projects I will write on later about.

What do you think about how networking and industry can merge with WebVR and VR in general? Let me know by comments or tweeter. I will be posting soon my take on it with a few live example and demos.

Labels:

Comments

Post a Comment

Popular posts from this blog

Racecraft (Project Koru) · Prologue — The Origin Story

Racecraft · Prologue , The Origin Story It Started With a Wine List and a Question About Racing How a happy-hour conversation in the Bay Area turned into a trustable AI race coach , and then into a second version that runs entirely on a phone, on the NPU. This is the prologue to a five-part series. Two years ago(1st November, 2024) I was in the Bay Area for a GDE Summit. If you've never been: it's a couple of days of talks among Google Developer Experts, the kind of people who get unreasonably excited about a new on-device runtime, and then , mercifully , a happy hour where everyone stops performing and just eats. We ended up at a restaurant(Puesto Santa Clara), a long table of GDEs, and I was doing the most important engineering of the evening: trying to decide which wine to order. Across the table was Ajeet Mirwani . I don't even remember how the wine talk turned into racing talk , these things drift , but the moment the word "racing" ...
Post a Comment
Read more

A Split‑Brain Neuro‑Symbolic Training Method for High‑Velocity Autonomous Coaching from Telemetry

 Author: Rabimba Karanjai Scope: Problem statement + data methodology + model training (no deployment discussion) Abstract Real‑time coaching in motorsport is a safety‑critical learning problem : a system must map noisy, high‑frequency telemetry to short, actionable guidance that remains physically consistent and avoids hazardous recommendations . This paper proposes a “Split‑Brain” training formulation that separates (i) a semantic coaching target (what action/critique should be expressed) from (ii) a reflexive interface (how actions are represented as compact, verifiable tokens). The approach trains a Small Language Model (SLM) in the Gemma family [1] using QLoRA fine‑tuning [2] , and introduces a telemetry tokenizer plus teacher‑student synthesis pipeline to generate instruction‑action pairs at scale. Core contribution: a reproducible method to convert “ golden lap ” differential tel...
Post a Comment
Read more

The Throughput Trap: Benchmarking vLLM on OpenXLA and the Reality of Production LLM Serving

vLLM Systems · DevLab 2026, Deep Dive I was recently invited by the Google TPU team to speak at the OpenXLA Summer DevLab 2026 . This post breaks down our deep-dive evaluation of the matured vLLM + OpenXLA stack, the fundamental engineering mismatches between CUDA and XLA serving paths, and why traditional capacity metrics are lying to you. If you are operating large language models at enterprise scale right now, your platform architecture team is likely staring at a massive infrastructure crossroads: Should we migrate our core serving workloads from GPUs to TPUs? Historically, NVIDIA's CUDA ecosystem was the only serious option for user-facing, low-latency LLM generation. But here in 2026, the economics and infrastructure options have transformed. Google TPUs are highly available, cheaper per chip, and the open-source serving stack built around vLLM and OpenXLA has officially achieved absolute production readiness. Yet, when our infrast...
Post a Comment
Read more