Larry Smarr: The Future of Distributed Computing Is Here

By Alex Woodie

August 25, 2020

Larry Smarr may have stepped back from full-time work in the Computer Science and Engineering Department at the University of California, San Diego, but that doesn’t mean he’s slowing down. In fact, since his schedule has opened up, he’s now free to do other things, such as evangelizing the benefits of distributed computing for local startup Kazuhm. Datanami recently caught up with Smarr to get his take on how new technologies like Kubernetes and 5G are changing the distributed computing game.

Smarr, who received his PhD. in physics from the University of Texas at Austin in 1975, has had the sort of academic career that most computer scientists can only dream of. In 1983, he famously submitted a short 10-page report to the National Science Foundation called “A Center for Scientific and Engineering Supercomputing.” Better known as the Black Proposal (for the color of the cover), it was the first unsolicited proposal accepted by the NSF, and led to the foundation of five supercomputer centers, at Cornell, Illinois, Princeton, San Diego, and Pittsburgh.

As the founding director of the National Center for Supercomputing Applications (NCSA) at the University of Illinois, Champaign-Urbana, Smarr played a role in ushering in the modern Web with the development of Mosaic, a Web browser released in 1993 that built on the HTTP and HTML protocols that Tim Berners-Lee created three years earlier. Smarr moved to California in 2000 and helped found the California Institute for Telecommunications and Information Technology (Calit2) at UCSD and then UC Irvine.

In June, Smarr stepped back from full-time duty at UCSD and serving as the director of Calit2 and took the title Distinguished Professor Emeritus. As the principal investigator on three large computing research proposals, however, Smarr has not entirely divorced himself from UCSD. In fact, the school has called him back to “active duty” to work on these projects (so long as it doesn’t consume in excess of 43% of the time, per the bureaucratic rule book).

Freed from the day-to-day requirements of his former job, Smarr signed up to be an evangelist for distributed computing with a local San Diego outfit called Kazhum. While he has no official position with the Analytic Ventures-funded startup, he clearly sees potential in the company, which develops a platform that utilizes Docker containers and Kubernetes to enable customers to distribute computing workloads to any device, including smaller devices on the IoT.

According to Smarr, the computing world is on the cusp of major changes, even if it doesn’t necessarily feel that way to many people.

“When these things come–and they come on in exponentials–people begin to hear the words, but they don’t get there’s going to be major transformation,” Smarr says. “Go back to file sharing with Napster. All the executives at the record companies were sending out lawyers to old ladies and the kids. They didn’t get that the world was going to completely change and they were going to become extinct in about three years. Or Blockbuster. Or Netflix.

“I’ve lived through so many of these transformations, and I just think the day of ubiquitous distributed computing is finally here,” he continues. “It isn’t like it hasn’t been here for decades. That’s what I, of course, have been doing for decades. But it’s getting to the point where it’s actually becoming mainstream.”

While many computing advances emerged from academia in decades past, we’re now seeing more technological breakthroughs from companies in the private sector, Smarr points out. Google, for instance, has been party to many of the advances since 2000, including the Google File System, which Doug Cutting and Mike Cafaraella used as the inspiration for the Hadoop Distributed File System; Big Table, a NoSQL database built atop the Google File System; and Spanner, a relational database built atop Big Table.

Smarr identified Kubernetes–a resource scheduler for containers created by Google to manage its internal computing resource and released as an open source project in 2014–as an especially disruptive technology, and one that will help to usher in the next generation of distributed computing.

“To me, the containers are just an extraordinary advance because what you always needed in a distributed system was a mobile way for your software to run around in distributed system, and then execute when it found a host that it could sit on,” Smarr explains. “In other words, you’re not doing RPC [remote procedure calls]. You’re not doing remote log-in on that computer to be able to use it, which is what we had to do forever. You let the software worry about that.”

For decades, companies have needed a system that could identify all the computing resources in the organization. But it’s only been in the past few years, with Docker and Kubernetes, such systems have been available and practical to use, according to Smarr.

“I think it’s the biggest change in distributed computing since Mosaic, and more importantly since Tim Berners-Lee and CERN wrote down the HTML and HTTP protocols,” Smarr says. “That unleashed an enormous capability that we didn’t know was there. Now 2 billion people per day update their Facebook page.”

Many companies that want to move their applications to the cloud have had to “containerize” their software–i.e. adapt it to run within a container (usually Docker). Those that have done that work have benefited from the increased flexibility that cloud computing brings (although it’s not necessarily cheaper as Smarr points out).

The cloud is a homogenous computing environment, with Linux on X86 being the norm, even for Microsoft Azure. But with tools like Kazuhm, which works with heterogenous devices spanning Linux, Windows, Mac, VMware, and other operating environments for servers, desktops, laptops, smart phones, tablets and IoT devices, companies now have a way to manage their entire IT infrastructure using containers and Kubernetes.

“It’s not obvious to me that Kubernetes is an appropriate way to orchestrate containers in a normal enterprise. It’s a wonderful thing if you go, say, between the clouds or anything else,” Smarr says. “But I’m also sort of struck by the number of folks who haven’t yet understood that containers change everything and they’re still trying to figure out, ‘You mean I’ve got to containerize? What the hell does that mean?’

“We’re on that part of the learning curve where there’s still an awful lot of very smart, savvy IT people who haven’t made the connection,” he continues. “One of the things that really helps people is you have something like Kazuhm where it’s a tool that you can go into your existing enterprise– you don’t have to change any of your stuff–and it will actually map out that enterprise that you own but you maybe don’t know.”

Smarr says we’re on the cusp of another breakthrough, around connected devices and the IoT. The rapid increase in the capabilities of system-on-a-chip designs are enabling companies to place more powerful devices in more locations in the real world. As the 5G rollout brings fast network access with low latencies, the potential to leverage all that previously isolated computing power will open up new business opportunities.

“One of the things that we think are not on people’s radar yet is just how rapidly IoT processors are gaining in capacity in speed and complexity,” Smarr says. “What I see coming is this partnership between the cloud and an exponentially larger set of machine intelligent sensors on the edge of the net.”

With Calit2, Smarr was involved with WIFIRE, a project that used remote sensors to monitor for wildfires in San Diego County. The project includes a high-speed wireless grid that connects 60 backcountry stations, where optical sensors use machine learning algorithms trained on large neural nets to identify smoke plumes. This is an example of how distributed computing applications will be deployed in the future, he says.

“You’re doing a bunch of training on a neural net for what does a smoke plume look like. You don’t want to do that on the camera. You want to do that on the cloud with a zillion instances of images, and then you want to deploy that trained, weighted net into a low-energy, fast device on the edge,” Smarr says.

“Well, if you actually look at those devices, how they’re executing, many of them are becoming quite powerful,” he continues. “Right now we think of the server room as the heavy weight IT. And then we have all these hundreds or thousands of laptops. That’s lightweight. Now imagine you’ve got tens to hundreds of thousands of IoT devices. That’s the next layer coming.”

With a wide deployment of 5G antennas and many connected devices with increasingly fast processors and useful memory footprints, companies will essentially have their own private cloud at their disposal. With Kubernetes and containers, and software like Kazuhm’s handling the low-level technical details of distribution of workloads and interrupts, companies will be equipped to use that private cloud to do useful work on the massive amount of data they’re collecting.

We’re in the midst of “an invisible rising tide of new computing parallelism,” Smarr says. “It’s like somehow we did a phase change and dissolved all the barriers between all of our computing devices with containers.”

Smarr has been on the forefront of these kind of phase changes before. He points out that, when he helped start Calit2 in 2000, the group’s byline was “The Internet moving to the physical world.” That was before Wi-Fi was widely available, he points out, and three years before users started accessing the Internet on their cell phones.

“It’s kind of scary to live long enough to see your prophecies coming true,” Smarr says. “I do thank the Lord that that I’ve lived this long so that I can see some of this happening. But it was inevitable. And as it happens, as you get these hugely disruptive things happening, whether it’s the Internet or the Web or social media, people are still stuck in the older world. As William Gibson said, the future is already here. It’s just unevenly distributed.”

This story originally appeared on sister publication Datanami.

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industy updates delivered to you every week!

SPEC Introduces SPEChpc 2021Suite for Heterogeneous Systems

October 28, 2021

SPEC – the Standard Performance Evaluation Company – introduced its newest benchmark suite today, SPEChpc 2021, intended to measure “intense compute parallel performance across one or more nodes.” Founded in 1988 Read more…

Rockport Networks Launches 300 Gbps Switchless Fabric, Reveals 396-Node Deployment at TACC

October 27, 2021

Rockport Networks emerged from stealth this week with the launch of its 300 Gbps switchless networking architecture focused on the needs of the high-performance computing and the advanced-scale AI market. Early customers Read more…

AWS Adds Gaudi-Powered, ML-Optimized EC2 DL1 Instances, Now in GA

October 27, 2021

As machine learning becomes a dominating use case for local and cloud computing, companies are racing to provide solutions specifically optimized and accelerated for AI applications. Now, Amazon Web Services (AWS) is int Read more…

Fireside Chat with LBNL’s Advanced Quantum Testbed Director

October 26, 2021

Last week, Irfan Siddiqi led a “fireside chat” with a few media and analysts to introduce the Department of Energy’s relatively new Advanced Quantum Testbed (AQT), which is based at Lawrence Berkeley National Labor Read more…

Graphcore Introduces Larger-Than-Ever IPU-Based Pods

October 22, 2021

After launching its second-generation intelligence processing units (IPUs) in 2020, four years after emerging from stealth, Graphcore is now boosting its product line with its largest commercially-available IPU-based sys Read more…

AWS Solution Channel

Royalty-free stock illustration ID: 577238446

Putting bitrates into perspective

Recently, we talked about the advances NICE DCV has made to push pixels from cloud-hosted desktops or applications over the internet even more efficiently than before. Read more…

Quantum Chemistry Project to Be Among the First on EuroHPC’s LUMI System

October 22, 2021

Finland’s CSC has just installed the first module of LUMI, a 550-peak petaflops system supported by the European Union’s EuroHPC Joint Undertaking. While LUMI -- pictured in the header -- isn’t slated to complete i Read more…

SPEC Introduces SPEChpc 2021Suite for Heterogeneous Systems

October 28, 2021

SPEC – the Standard Performance Evaluation Company – introduced its newest benchmark suite today, SPEChpc 2021, intended to measure “intense compute paral Read more…

Rockport Networks Launches 300 Gbps Switchless Fabric, Reveals 396-Node Deployment at TACC

October 27, 2021

Rockport Networks emerged from stealth this week with the launch of its 300 Gbps switchless networking architecture focused on the needs of the high-performance Read more…

AWS Adds Gaudi-Powered, ML-Optimized EC2 DL1 Instances, Now in GA

October 27, 2021

As machine learning becomes a dominating use case for local and cloud computing, companies are racing to provide solutions specifically optimized and accelerate Read more…

Fireside Chat with LBNL’s Advanced Quantum Testbed Director

October 26, 2021

Last week, Irfan Siddiqi led a “fireside chat” with a few media and analysts to introduce the Department of Energy’s relatively new Advanced Quantum Testb Read more…

Killer Instinct: AMD’s Multi-Chip MI200 GPU Readies for a Major Global Debut

October 21, 2021

AMD’s next-generation supercomputer GPU is on its way – and by all appearances, it’s about to make a name for itself. The AMD Radeon Instinct MI200 GPU (a successor to the MI100) will, over the next year, begin to power three massive systems on three continents: the United States’ exascale Frontier system; the European Union’s pre-exascale LUMI system; and Australia’s petascale Setonix system. Read more…

D-Wave Embraces Gate-Based Quantum Computing; Charts Path Forward

October 21, 2021

Earlier this month D-Wave Systems, the quantum computing pioneer that has long championed quantum annealing-based quantum computing (and sometimes taken heat fo Read more…

LLNL Prepares the Water and Power Infrastructure for El Capitan

October 21, 2021

When it’s (ostensibly) ready in early 2023, El Capitan is expected to deliver in excess of two exaflops of peak computing power – around four times the powe Read more…

Intel Reorgs HPC Group, Creates Two ‘Super Compute’ Groups

October 15, 2021

Following on changes made in June that moved Intel’s HPC unit out of the Data Platform Group and into the newly created Accelerated Computing Systems and Graphics (AXG) business unit, led by Raja Koduri, Intel is making further updates to the HPC group and announcing... Read more…

Enter Dojo: Tesla Reveals Design for Modular Supercomputer & D1 Chip

August 20, 2021

Two months ago, Tesla revealed a massive GPU cluster that it said was “roughly the number five supercomputer in the world,” and which was just a precursor to Tesla’s real supercomputing moonshot: the long-rumored, little-detailed Dojo system. Read more…

Esperanto, Silicon in Hand, Champions the Efficiency of Its 1,092-Core RISC-V Chip

August 27, 2021

Esperanto Technologies made waves last December when it announced ET-SoC-1, a new RISC-V-based chip aimed at machine learning that packed nearly 1,100 cores onto a package small enough to fit six times over on a single PCIe card. Now, Esperanto is back, silicon in-hand and taking aim... Read more…

US Closes in on Exascale: Frontier Installation Is Underway

September 29, 2021

At the Advanced Scientific Computing Advisory Committee (ASCAC) meeting, held by Zoom this week (Sept. 29-30), it was revealed that the Frontier supercomputer is currently being installed at Oak Ridge National Laboratory in Oak Ridge, Tenn. The staff at the Oak Ridge Leadership... Read more…

Intel Reorgs HPC Group, Creates Two ‘Super Compute’ Groups

October 15, 2021

Following on changes made in June that moved Intel’s HPC unit out of the Data Platform Group and into the newly created Accelerated Computing Systems and Graphics (AXG) business unit, led by Raja Koduri, Intel is making further updates to the HPC group and announcing... Read more…

Ahead of ‘Dojo,’ Tesla Reveals Its Massive Precursor Supercomputer

June 22, 2021

In spring 2019, Tesla made cryptic reference to a project called Dojo, a “super-powerful training computer” for video data processing. Then, in summer 2020, Tesla CEO Elon Musk tweeted: “Tesla is developing a [neural network] training computer... Read more…

Intel Completes LLVM Adoption; Will End Updates to Classic C/C++ Compilers in Future

August 10, 2021

Intel reported in a blog this week that its adoption of the open source LLVM architecture for Intel’s C/C++ compiler is complete. The transition is part of In Read more…

Hot Chips: Here Come the DPUs and IPUs from Arm, Nvidia and Intel

August 25, 2021

The emergence of data processing units (DPU) and infrastructure processing units (IPU) as potentially important pieces in cloud and datacenter architectures was Read more…

AMD-Xilinx Deal Gains UK, EU Approvals — China’s Decision Still Pending

July 1, 2021

AMD’s planned acquisition of FPGA maker Xilinx is now in the hands of Chinese regulators after needed antitrust approvals for the $35 billion deal were receiv Read more…

Leading Solution Providers

Contributors

Intel Unveils New Node Names; Sapphire Rapids Is Now an ‘Intel 7’ CPU

July 27, 2021

What's a preeminent chip company to do when its process node technology lags the competition by (roughly) one generation, but outmoded naming conventions make i Read more…

HPE Wins $2B GreenLake HPC-as-a-Service Deal with NSA

September 1, 2021

In the heated, oft-contentious, government IT space, HPE has won a massive $2 billion contract to provide HPC and AI services to the United States’ National Security Agency (NSA). Following on the heels of the now-canceled $10 billion JEDI contract (reissued as JWCC) and a $10 billion... Read more…

Quantum Roundup: IBM, Rigetti, Phasecraft, Oxford QC, China, and More

July 13, 2021

IBM yesterday announced a proof for a quantum ML algorithm. A week ago, it unveiled a new topology for its quantum processors. Last Friday, the Technical Univer Read more…

The Latest MLPerf Inference Results: Nvidia GPUs Hold Sway but Here Come CPUs and Intel

September 22, 2021

The latest round of MLPerf inference benchmark (v 1.1) results was released today and Nvidia again dominated, sweeping the top spots in the closed (apples-to-ap Read more…

10nm, 7nm, 5nm…. Should the Chip Nanometer Metric Be Replaced?

June 1, 2020

The biggest cool factor in server chips is the nanometer. AMD beating Intel to a CPU built on a 7nm process node* – with 5nm and 3nm on the way – has been i Read more…

Julia Update: Adoption Keeps Climbing; Is It a Python Challenger?

January 13, 2021

The rapid adoption of Julia, the open source, high level programing language with roots at MIT, shows no sign of slowing according to data from Julialang.org. I Read more…

Frontier to Meet 20MW Exascale Power Target Set by DARPA in 2008

July 14, 2021

After more than a decade of planning, the United States’ first exascale computer, Frontier, is set to arrive at Oak Ridge National Laboratory (ORNL) later this year. Crossing this “1,000x” horizon required overcoming four major challenges: power demand, reliability, extreme parallelism and data movement. Read more…

D-Wave Embraces Gate-Based Quantum Computing; Charts Path Forward

October 21, 2021

Earlier this month D-Wave Systems, the quantum computing pioneer that has long championed quantum annealing-based quantum computing (and sometimes taken heat fo Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire