Quantum Networking and Clustering – What Is It? Why Should You Care? Who’s Aliro?

By John Russell

October 4, 2021

Taking those questions in reverse order, Aliro Quantum is a young Harvard lab spin-out seeking to deliver the quantum networking technology many believe is critical to scaling up quantum computers. Aliro contends that clustering moderate size quantum computers, say 1000-qubit systems (still not buildable, though IBM says perhaps soon) or bigger, is the most likely way to create something resembling a monolithic 1M-qubit-or-more-sized quantum computer large enough to solve practical problems.

“That thesis is at the core of our business model and of many others,” said Prineha Narang, Aliro CTO, founder, Harvard professor. “People, mostly from the hardware side, are trying to do this with us. We’re interfacing down to the hardware – getting down to the FPGAs – we’re just not building the quantum hardware. We see the path to scale being through these networks, and those types of networks being part of the bigger picture of these entanglement-generating and entanglement-using large scale networks.”

Whether these connected quantum computers end up being called clusters or distributed computing (both terms have currency now) or something else isn’t yet clear, she said. Narang and Harvard colleague, John Philbin, have an interesting recent paper (Computational Materials Insights Into Solid-State Multiqubit Systems) in the APS journal PRX Quantum.

Prineha Narang, founder and CTO of Aliro Quantum

There are, of course, many other quantum networking applications, but in essence they all connect some sort of quantum device (e.g. a sensor) to another quantum device. It should be noted not everyone agrees that clustering is the only or best path to scale-up quantum computers. PsiQuantum, for example, says its photonics-based quantum computer that’s leveraging semiconductor fabrication techniques will scale to a million or more qubits.

That said, there is broad agreement that practical, scalable quantum networking is an important ingredient in the development of a robust quantum information sciences landscape. DOE has several related projects and there are similar efforts around the globe.

How do these devices work? Broadly, a quantum network must interface with a quantum computer (or other quantum device), capture and faithfully transmit a qubit-based information stream to another device able to use the data. Accomplishing that requires dealing with familiar quantum challenges: generating entanglement, managing coherency duration, enacting error correction, using quantum memory, and – to reach any distance (WANs) – reliable repeaters. Also, don’t forget there are currently numerous qubit technologies such as semiconductor-based superconducting, trapped ions, cold atom, photonics, etc. A ‘generalizable’ quantum network would need to be able to interface with any of them.

A positive for quantum networks is mature optical technology; high-quality optic fiber is a good transmission medium and the most viable current option. Aliro is focused on quantum network control plane and protocol development. It is leveraging quantum and traditional networking hardware. Think of Aliro as a little like Mellanox for quantum networking, said Narang.

Putting aside, for a moment, remaining technical challenges, consider Narang’s ambitious milestones for Aliro and forecast for quantum networking:

  • Two years. “Near term, the next two-year timeframe, we think that metropolitan areas is the best that we can hope for. Our focus is taking one of these proof-of-concept networks and showing a commercial POC network, which means we can deliver the network not as something that only fits in a DOE lab with all of its [support], but something that sits in a commercial setting. If we can do that over the next two years, we’ll be very happy. That’s a major Aliro milestone,” said Narang.
  • Five years. “On the five-year horizon, we expect to see repeaters that are commercially available. There are startups, particularly some of the folks in Europe, that have technology they’re trying to package and commercialize that looks like repeater technology. The expectation is that in the next few years, somebody will be able to place a purchase order for those, say order five of those from Aliro, and we can start testing those out in those commercial networks.”
  • By 2030. “By the end of the decade, we can start to see a commercial network that is city-to-city. And this is not just my set of milestones, it’s also what the field has put forward,” she said.
  • How about today. “Clustering is much more near-term. It’s something that’s happening now. People are trying to cluster quantum devices. The push to scale has hit a bit of a roadblock with these jumbo dilution refrigerators only being able to do so much. [Higher temp] trapped ion architectures are being embraced, and photonic fabric because you have lasers and laser table distancing. I think we’re going to hear a lot more in terms of clustered or distributed quantum computing over time, and also these hybrid kinds of networks,” said Narang.

Aliro was spun out of Narang’s Harvard lab in 2019 where her research focuses on a variety of quantum topics spanning quantum materials, quantum information, and quantum molecular dynamics. The Aliro headcount is roughly 20 and growing, and Narang cites the greater Boston area’s wealth of high tech resources (university, networking tech expertise, talent) as an important factor in deciding to set up shop there.

“We’ve been developing a control plane. Some of the things I’m going to say are very similar to the language you’d hear from classical networking. We’re thinking about the quantum version in the same way. It has various layers, including a physical layer, like a classical network. The temptation is always, even in classical networking, to say let’s do everything in the physical layer because the hardware is becoming better,” said Narang.

“But the reason classical networking has been so successful is that the details of the physical layer are abstracted away into some of these other layers. This is why every time there’s a hardware upgrade, how I think about the internet doesn’t change. That abstraction is what enables us to think about issues of timing, synchronization, and connecting these devices in the control plane. That’s really where the key value is.”

The idea is to hide the underlying complexity such as qubit modality. “What’s the right architecture for connecting to superconducting and trapped ion systems? What amount of architecting with a control plane is needed versus how much is overkill? So what can you do over a quantum channel versus what can you do over the classical channel in order to get the timing right or to get some of the synchronization problems solved?” said Narang citing key questions.

“We want people to come into this field and not have to figure out all of these other pieces, but to be able to easily interface with these systems. The phrase plug-and-play gets used a lot in the sales. [Quantum networking] is still not plug-and-play but it’s certainly become much more accessible to a broader set of engineers and scientists than five years ago,” she said.

Besides clustering quantum computers, there is a major push to develop a so-called quantum internet. Among potential applications identified in a 2018 Science article are: secure communication, clock synchronization, extending the baseline of telescopes, secure identification, achieving efficient agreement on distributed data, exponential savings in communication, quantum sensor networks, as well as secure access to remote quantum computers in the cloud. (Figure from Science article, 10-31-2018, shown below.)

“From a hardware standpoint, key components are missing at the moment, quantum repeaters being a significant one. This is an area where my lab does a lot of work in looking at how we think about third-generation quantum repeaters that actually can enable the kind of capacity you need to have a meaningful large-scale quantum network using solid state components. There’s a lot of talk about this now. DOE has announced a very big roadmap and blueprint to connect the various DOE labs, which, of course, are all across the country. Essentially, this is the quantum version of the ARPANET. And there are some hardware advances needed before we can talk about connecting something as big as that.

“Having said that, a metropolitan area network is definitely much more achievable. This is where you’re within the repeater-less bound. There’s actually fiber between Harvard, MIT, and Lincoln Labs and these things can be connected. Now, over a network like that, you’re not having a very high bandwidth conversation with your buddies on the other side. But these are entanglement using and generating networks, they’re a proof of concept, testbeds. We have one here. There’s one at Argonne and Fermi Lab in the Chicago area, that’s getting a lot of attention. These are attempts at figuring out what are the components, both hardware and software, that will become part of a larger scalable network,” said Narang.

Narang noted that codesign is important in quantum networking development. “As much as we are abstracting away from the hardware, there are also many hardware choices that need to be made that are informed by the protocols (software),” said Narang.

“There’s a temptation from the hardware community to say, ‘We’re going to build out the hardware, these folks are going to throw some software on there and it’s all going to be great.’ It turns out there’s some hard constraints that are imposed on the hardware, based on some of the protocols and algorithms of interest. So codesign has been a really key component here. We’re collaborating very closely with the DOE labs, in particular ESnet based out of LBNL though they (ESnet) have fiber all across the country. [We are working with them] on thinking about how we can emulate these long-haul links because that’s where we see the value. We don’t see value in simple quantum key distribution (QKD). There are people doing it, but you can look up very public documents from the DOD and NSA that say that they don’t view the path in secure communication to be at all related to QKD. So for entanglement generating and using networks, the biggest value really is from having these components and getting to getting to scale.”

Aliro envisions being able to “spec out the entire network.” This would entail having Aliro boxes at both ends and gaining access to existing high-quality fiber in between. Generating entanglement and sharing entanglement are core capabilities required. Quantum memory is also important (needed for repeaters). Routing is also a challenge and Aliro has put forward a set of protocols for routing, said Narang, who is working with the Quantum Internet Research Group (IETF Quantum IRG) seeking to set standards and is active in the NSF-affiliated Center for Quantum Networks.

She expects there to be a variety of node types, think smart and less so. Fees for entanglement-as-a-service would likely be based on both transmission fidelity and transmission rates. Big customers such as a bank might want exotic boxes and pay a premium; others might not require that. The Aliro website lists “quantum-secure communications, improved GPS precision and reliability, and accurate positioning, navigation and timing” as potential application for its entanglement-as-a-service (EaaS). That seems less connected to the idea of scaling up quantum computers via clustering. To some extent, the promotion may be aspirational as the quantum network world is still nascent.

“We’re partnering with hardware vendors that we’ve established relationships with. I’m being intentionally a little vague here. It’s a tight rope here,” said Narang. “Currently, if you deliver a network, it’s going to be slow, the entanglement rates are very slow. And it’s very small in the sense of the geography because we’re waiting for repeaters to come online to take this across the country. However, there are a lot of things that we can do to anticipate what kinds of repeaters we’ll have. We’re working with various types of architectures at testbeds like, the folks at Argonne, and the folks at Brookhaven. They have different realizations of what a repeater would look like.”

Like many, Narang thinks moving quantum processing out of icy cold dilution refrigerators (a few degrees Kelvin) will be an important step. “There’s a strong push in my research to look at quantum memories that are at temperatures above four Kelvin that can operate with reasonable fidelity at that temperature, because we think that [operating] with liquid nitrogen cooled (77 kelvin) is no problem, but as soon as you’re talking about, you know, these ultra-low temperature, pumped-helium systems, and there are.” She’s a fan of solid state memory citing diamond and silicon carbide (color centers) which could offer higher temperature operating ranges.

Choosing partners and technologies is tricky. “We’ve been in deep conversations with both Google and Amazon and they have very different roadmaps going forward and we haven’t ourselves decided if would try to achieve both or pick one. There are reasons to pick and there are reasons to not pick,” she said.

Likewise, photonics expertise will be critical. Xanadu and PsiQuantum are among the more prominent quantum computer companies trying to develop optically-based systems. PsiQuantum has been very direct about its plans to launch a million-qubit quantum computer.

“We have talked with them,” said Narang. “We don’t currently have a signed partnership with them. We’re being very cautious to not get tied to one particular photonic platform. We think that we will need a long-term photonic platform partner, whether it’s going to be PsiQuantum or Xanadu, or another, we’re not sure. They, on their roadmaps, don’t have plans for integrating quantum repeaters.”

That’s okay, said Narang, but Aliro would prefer somebody on the photonics side that’s interested both in these connected photonic processors, and how you bring in the repeater into the picture.

Stay tuned.

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!

Energy Exascale Earth System Model Version 2 Promises Twice the Speed

October 18, 2021

The Energy Exascale Earth System Model (E3SM) is an ongoing Department of Energy (DOE) earth system modeling, simulation and prediction project aiming to “assert and maintain an international scientific leadership posi 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…

Royalty-free stock illustration ID: 1938746143

MosaicML, Led by Naveen Rao, Comes Out of Stealth Aiming to Ease Model Training

October 15, 2021

With more and more enterprises turning to AI for a myriad of tasks, companies quickly find out that training AI models is expensive, difficult and time-consuming. Finding a new approach to deal with those cascading challenges is the aim of a new startup, MosaicML, that just came out of stealth... Read more…

NSF Awards $11M to SDSC, MIT and Univ. of Oregon to Secure the Internet

October 14, 2021

From a security standpoint, the internet is a problem. The infrastructure developed decades ago has cracked, leaked and been patched up innumerable times, leaving vulnerabilities that are difficult to address due to cost Read more…

SC21 Announces Science and Beyond Plenary: the Intersection of Ethics and HPC

October 13, 2021

The Intersection of Ethics and HPC will be the guiding topic of SC21's Science & Beyond plenary, inspired by the event tagline of the same name. The evening event will be moderated by Daniel Reed with panelists Crist Read more…

AWS Solution Channel

Cost optimizing Ansys LS-Dyna on AWS

Organizations migrate their high performance computing (HPC) workloads from on-premises infrastructure to Amazon Web Services (AWS) for advantages such as high availability, elastic capacity, latest processors, storage, and networking technologies; Read more…

Quantum Workforce – NSTC Report Highlights Need for International Talent

October 13, 2021

Attracting and training the needed quantum workforce to fuel the ongoing quantum information sciences (QIS) revolution is a hot topic these days. Last week, the U.S. National Science and Technology Council issued a report – The Role of International Talent in Quantum Information Science... 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…

Royalty-free stock illustration ID: 1938746143

MosaicML, Led by Naveen Rao, Comes Out of Stealth Aiming to Ease Model Training

October 15, 2021

With more and more enterprises turning to AI for a myriad of tasks, companies quickly find out that training AI models is expensive, difficult and time-consuming. Finding a new approach to deal with those cascading challenges is the aim of a new startup, MosaicML, that just came out of stealth... Read more…

Quantum Workforce – NSTC Report Highlights Need for International Talent

October 13, 2021

Attracting and training the needed quantum workforce to fuel the ongoing quantum information sciences (QIS) revolution is a hot topic these days. Last week, the U.S. National Science and Technology Council issued a report – The Role of International Talent in Quantum Information Science... Read more…

Eni Returns to HPE for ‘HPC4’ Refresh via GreenLake

October 13, 2021

Italian energy company Eni is upgrading its HPC4 system with new gear from HPE that will be installed in Eni’s Green Data Center in Ferrera Erbognone (a provi Read more…

The Blueprint for the National Strategic Computing Reserve

October 12, 2021

Over the last year, the HPC community has been buzzing with the possibility of a National Strategic Computing Reserve (NSCR). An in-utero brainchild of the COVID-19 High-Performance Computing Consortium, an NSCR would serve as a Merchant Marine for urgent computing... Read more…

UCLA Researchers Report Largest Chiplet Design and Early Prototyping

October 12, 2021

What’s the best path forward for large-scale chip/system integration? Good question. Cerebras has set a high bar with its wafer scale engine 2 (WSE-2); it has 2.6 trillion transistors, including 850,000 cores, and was fabricated using TSMC’s 7nm process on a roughly 8” x 8” silicon footprint. Read more…

What’s Next for EuroHPC: an Interview with EuroHPC Exec. Dir. Anders Dam Jensen

October 7, 2021

One year after taking the post as executive director of the EuroHPC JU, Anders Dam Jensen reviews the project's accomplishments and details what's ahead as EuroHPC's operating period has now been extended out to the year 2027. Read more…

University of Bath Unveils Janus, an Azure-Based Cloud HPC Environment

October 6, 2021

The University of Bath is upgrading its HPC infrastructure, which it says “supports a growing and wide range of research activities across the University.” 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…

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…

CentOS Replacement Rocky Linux Is Now in GA and Under Independent Control

June 21, 2021

The Rocky Enterprise Software Foundation (RESF) is announcing the general availability of Rocky Linux, release 8.4, designed as a drop-in replacement for the soon-to-be discontinued CentOS. The GA release is launching six-and-a-half months... 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 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…

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…

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…

Leading Solution Providers

Contributors

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…

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…

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…

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…

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…

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…

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…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire