As Vice President of Software program Engineering, Doug is accountable for all facets of the Cornelis Networks’ software program stack, together with the Omni-Path Structure drivers, messaging software program, and embedded machine management techniques. Earlier than becoming a member of Cornelis Networks, Doug led software program engineering groups at Crimson Hat in cloud storage and knowledge companies. Doug’s profession in HPC and cloud computing started at Ames Nationwide Laboratory’s Scalable Computing Laboratory. Following a number of roles in college analysis computing, Doug joined the US Division of Vitality’s Oak Ridge Nationwide Laboratory in 2009, the place he developed and built-in new applied sciences on the world-class Oak Ridge Management Computing Facility.
Cornelis Networks is a expertise chief delivering purpose-built high-performance materials for Excessive Efficiency Computing (HPC), Excessive Efficiency Information Analytics (HPDA), and Synthetic Intelligence (AI) to main business, scientific, tutorial, and authorities organizations.
What initially attracted you to laptop science?
I simply appeared to get pleasure from working with expertise. I loved working with the computer systems rising up; we had a modem at our faculty that permit me check out the Web and I discovered it attention-grabbing. As a freshman in school, I met a USDOE computational scientist whereas volunteering for the Nationwide Science Bowl. He invited me to tour his HPC lab and I used to be hooked. I have been a supercomputer geek ever since.
You labored at Crimson Hat from 2015 to 2019, what have been among the initiatives you labored on and your key takeaways from this expertise?
My primary challenge at Crimson Hat was Ceph distributed storage. I might beforehand targeted totally on HPC and this gave me a possibility to work on applied sciences that have been crucial to cloud infrastructure. It rhymes. Lots of the rules of scalability, manageability, and reliability are extraordinarily related regardless that they’re aimed toward fixing barely completely different issues. By way of expertise, my most vital takeaway was that cloud and HPC have rather a lot to study from each other. We’re more and more constructing completely different initiatives with the identical Lego set. It is actually helped me perceive how the enabling applied sciences, together with materials, can come to bear on HPC, cloud, and AI purposes alike. It is also the place I actually got here to know the worth of Open Supply and how one can execute the Open Supply, upstream-first software program improvement philosophy that I introduced with me to Cornelis Networks. Personally, Crimson Hat was the place I actually grew and matured as a pacesetter.
You’re at present the Vice President of Software program Engineering at Cornelis Networks, what are a few of your duties and what does your common day appear like?
As Vice President of Software program Engineering, I’m accountable for all facets of the Cornelis Networks’ software program stack, together with the Omni-Path Structure drivers, messaging software program, material administration, and embedded machine management techniques. Cornelis Networks is an thrilling place to be, particularly on this second and this market. Due to that, I am undecided I’ve an “common” day. Some days I am working with my group to resolve the newest expertise problem. Different days I am interacting with our {hardware} architects to ensure our next-generation merchandise will ship for our clients. I am usually within the discipline assembly with our wonderful group of shoppers and collaborators ensuring we perceive and anticipate their wants.
Cornelis Networks affords subsequent era networking for Excessive Efficiency Computing and AI purposes, may you share some particulars on the {hardware} that’s supplied?
Our {hardware} consists of a high-performance switched material sort community material answer. To that finish, we offer all the mandatory units to totally combine HPC, cloud, and AI materials. The Omni-Path Host-Material Interface (HFI) is a low-profile PCIe card for endpoint units. We additionally produce a 48-port 1U “top-of-rack” change. For bigger deployments, we make two fully-integrated “director-class” switches; one which packs 288 ports in 7U and an 1152-port, 20U machine.
Are you able to focus on the software program that manages this infrastructure and the way it’s designed to lower latency?
First, our embedded administration platform gives simple set up and configuration in addition to entry to all kinds of efficiency and configuration metrics produced by our change ASICs.
Our driver software program is developed as a part of the Linux kernel. In reality, we submit all our software program patches to the Linux kernel group instantly. That ensures that each one of our clients get pleasure from most compatibility throughout Linux distributions and simple integration with different software program equivalent to Lustre. Whereas not within the latency path, having an in-tree driver dramatically reduces set up complexity.
The Omni-Path material supervisor (FM) configures and routes an Omni-Path material. By optimizing visitors routes and recovering rapidly from faults, the FM gives industry-leading efficiency and reliability on materials from tens to hundreds of nodes.
Omni-Path Specific (OPX) is our high-performance messaging software program, just lately launched in November 2022. It was particularly designed to cut back latency in comparison with our earlier messaging software program. We ran cycle-accurate simulations of our ship and obtain code paths as a way to decrease instruction rely and cache utilization. This produced dramatic outcomes: if you’re within the microsecond regime, each cycle counts!
We additionally built-in with the OpenFabrics Interfaces (OFI), an open commonplace produced by the OpenFabrics Alliance. OFI’s modular structure helps decrease latency by permitting higher-level software program, equivalent to MPI, to leverage material options with out further perform calls.
The complete community can be designed to extend scalability, may you share some particulars on the way it is ready to scale so effectively?
Scalability is on the core of Omni-Path’s design rules. On the lowest ranges, we use Cray link-layer expertise to right hyperlink errors with no latency affect. This impacts materials in any respect scales however is especially vital for large-scale materials, which naturally expertise extra hyperlink errors. Our material supervisor is concentrated each on programming optimum routing tables and on doing so in a fast method. This ensures that routing for even the biggest materials could be accomplished in a minimal period of time.
Scalability can be a crucial element of OPX. Minimizing cache utilization improves scalability on particular person nodes with giant core counts. Minimizing latency additionally improves scalability by enhancing time to completion for collective algorithms. Utilizing our host-fabric interface sources extra effectively allows every core to speak with extra distant friends. The strategic selection of libfabric permits us to leverage software program options like scalable endpoints utilizing commonplace interfaces.
May you share some particulars on how AI is integrated into among the workflow at Cornelis Networks?
We’re not fairly prepared to speak externally about our inner makes use of of and plans for AI. That mentioned, we do eat our personal pet food, so we get to make the most of the latency and scalability enhancements we have made to Omni-Path to assist AI workloads. It makes us all of the extra excited to share these advantages with our clients and companions. We’ve definitely noticed that, like in conventional HPC, scaling out infrastructure is the one path ahead, however the problem is that community efficiency is well stifled by Ethernet and different conventional networks.
What are some modifications that you simply foresee within the {industry} with the arrival of generative AI?
First off, the usage of generative AI will make folks extra productive – no expertise in historical past has made human beings out of date. Each expertise evolution and revolution we’ve had from the cotton gin to the automated loom to the phone, web and past have made sure jobs extra environment friendly, however we haven’t labored humanity out of existence.
By way of the applying of generative AI, I consider corporations will technologically advance at a sooner price as a result of these working the corporate could have extra free time to deal with these developments. For example, if generative AI gives extra correct forecasting, reporting, planning, and many others. – corporations can deal with innovation of their discipline of experience
I particularly really feel that AI will make every of us a multidisciplinary skilled. For instance, as a scalable software program skilled, I perceive the connections between HPC, massive knowledge, cloud, and AI purposes that drive them towards options like Omni-Path. Outfitted with a generative AI assistant, I can delve deeper into the which means of the purposes utilized by our clients. I’ve little question that this may assist us design much more efficient {hardware} and software program for the markets and clients we serve.
I additionally foresee an total enchancment in software program high quality. AI can successfully perform as “one other set of eyes” to statically analyze code and develop insights into bugs and efficiency issues. This might be notably attention-grabbing at giant scales the place efficiency points could be notably tough to identify and costly to breed.
Lastly, I hope and consider that generative AI will assist our {industry} to coach and onboard extra software program professionals with out earlier expertise in AI and HPC. Our discipline can appear formidable to many and it may possibly take time to study to “assume in parallel.” Basically, similar to machines made it simpler to fabricate issues, generative AI will make it simpler to contemplate and purpose about ideas.
Is there the rest that you simply wish to share about your work or Cornelis Networks usually?
I might wish to encourage anybody with the curiosity to pursue a profession in computing, particularly in HPC and AI. On this discipline, we’re geared up with essentially the most highly effective computing sources ever constructed and we convey them to bear in opposition to humanity’s biggest challenges. It is an thrilling place to be, and I’ve loved it each step of the best way. Generative AI brings our discipline to even newer heights because the demand for growing functionality will increase drastically. I am unable to wait to see the place we go subsequent.
Thanks for the nice interview, readers who want to study extra ought to go to Cornelis Networks.