• Rayhan Memon
  • Posts
  • #1 - Getting Quantum Computers Ready for Prime Time

#1 - Getting Quantum Computers Ready for Prime Time

Quantum Computers will soon be integrated into data centers, creating a slew of new opportunities for founders and investors.

August 23, 2024 • Estimated Reading Time: 4 minutes

I like to cover important updates in the world of computing.

This week, Hybrid Quantum Computing…

All the computers in our homes, offices, and data centers are Classical Computers, meaning they represent and process data as strings of bits. A bit can only be in two states: 1 or 0.

Quantum Computers perform their calculations using qubits, which can be in both states at once: 1 and 0.

If you want a deeper explanation, these animated ducks do a pretty good job of it.

But for our purposes, all you need to know is that Quantum Computers can tackle certain math problems that even the most powerful Classical Computers could never solve in a reasonable amount of time. Like simulating the interactions between molecules to formulate new drugs and discover new materials.

Here’s the rub: Classical Computers are better at almost everything else…

That’s why supercomputer facilities around the world like Germany’s Jupiter, Japan’s Fugaku, and Poland’s PSNC are developing hybrid quantum-classical computers that combine the best of both approaches.

IBM — a leader in the space with their IBM Quantum System Two — calls this vision Quantum-Centric Supercomputing. These systems orchestrate work across quantum computers and advanced classical compute clusters in the same data center.

Hybrid systems are nothing new. Most computers nowadays have a combination of CPUs, GPUs, and other co-processors to perform specialized tasks like signal processing, networking, or encryption.

However, all these processors are built on the same CMOS technology and operate in fundamentally similar ways, whereas quantum processors are completely different.

So as early users of quantum computers look to integrate quantum computers into their data centers for improved security and latency, there are going to be a lot of problems (i.e. opportunities) that need addressing.

Hardware Problems:

  • Data centers are full of cooling fans and rogue electromagnetic radiation coming from high-powered electronics. This is a terrible environment for Quantum Computers, whose qubits are hypersensitive to noise.

  • Quantum Computers look different than the uniform server cabinets in today’s data centers. They need to operate at cryogenic temperatures, typically submerged inside a specialty liquid-helium fridge called a ‘dilution refrigerator’

Software Problems:

  • Workload managers (the software that orchestrates work across various compute resources) are designed to deal with long-running algorithms that need lots of compute and memory. Whereas quantum runtimes often take just a second to execute.

Workflow Problems: 

  • Quantum Computers are analog devices, and need to be regularly re-tuned. This is not something that data center engineers are accustomed to doing.

  • Quantum engineers are research scientists. They now need to get serious about manufacturing, customer support, and treating these computers as production systems rather than science experiments.

No one has yet conclusively demonstrated “Quantum Advantage”—that is, a quantum computer that outperforms the best classical one on a real-world relevant task. But these recent developments make me optimistic that that moment may come in the next decade.

Sources: