IBM, Google & Nasa Quantum Research Labs – Quantum computers promise new levels of performance


 

The race to build a quantum computer is heating up.

The U.S. government announced recently that it would provide a multi-year funding grant to IBM in order to allow it to continue its quantum computing research.

In its latest move to build a practical quantum computer, IBM Research for the first time ever is making quantum computing available in the cloud to anyone interested in hands-on access to the company’s advanced experimental quantum system.

To learn more about IBM’s quantum computing research and get access to the IBM Quantum Experience, visit: http://ibm.com/quantumcomputing

The UK recently announced a £270 million investment to bring quantum technology including a quantum computer to market by 2020, and Google has hired researchers to build its own quantum processors.

 

Tianhe-2 (Milkway-2)

Currently, the fastest computer cluster in the world is Tianhe-2 in China, which sits on 8,000 square feet of space, and is able to process 34 trillion operations per second or 34 petaflops based on the Linpack benchmark. The Tianhe-2 supercomputer was developed by China’s National University of Defense Technology (NUDT) in collaboration with the Chinese IT firm Inspur.

As recently as 2000, Inspur was a local company based in Shandong, with its business activities spanning North China. The company later moved its marketing center to Beijing, the IT industry hub of China, and expanded its operations to the national level.

In 2005 it was reported that Microsoft had invested 20 million US$ in the company. On 18 April 2006, the company switched its name from “Langchao” to “Inspur” in hopes it would increase its sales from overseas markets by as much as thirty percent by 2010.

Inspur announced several agreements with virtualization software developer VMware on research and development of cloud computing technologies and related products.

In August 2009, Inspur acquired the Xi’an-based research and development facilities of Qimonda AG for 30 million Chinese yuan (around US$4 million).

The centre had been responsible for design and development of Qimonda’s DRAM products.

In November 2011, Shandong Inspur Software Co., Ltd., Inspur Electronic Information Co., Ltd. and Inspur (Shandong) Electronic Information Company, established a cloud computing joint venture, with each holding a 33.3% stake.

Inspur participated in the production of printed circuit boards and the installation and testing of the Chinese Tianhe-2 Supercomputer, revealed in June 2013 as the most powerful supercomputer in the world.

In Sep 2015, Cisco Systems announced a $100 million joint venture which includes reselling Cisco networking gear and jointly developing a broad spectrum of software and hardware as part of a $10 billion China investment initiative.

According to NUDT, Tianhe-2 is used for simulation, analysis, and government security applications. Tianhe-2 is a cluster of 16,000 computer nodes, each comprising two Intel Ivy Bridge Xeon processors and three Xeon Phi chips, it represents the world’s largest installation of Ivy Bridge and Xeon Phi chips, counting a total of 3,120,000 cores. Each of the 16,000 nodes possess 88 gigabytes of memory (64 used by the Ivy Bridge processors, and 8 gigabytes for each of the Xeon Phi processors). The total CPU plus coprocessor memory is 1,375 TiB (approximately 1.34 PiB).  Yahoo, Google and others are rumored to have even larger clusters in operation. This information has not been made public.

The Tianhe-2 cluster is powerful by today’s standards. However, it is limited by the present silicon based design using millions of switches per CPU that represent on or off, a 0 or 1.

What is quantum computing?

Quantum computers would offer atom-sized bits commonly referred to as a qubits within a quantum CPU (qCPU) that is able to represent 0, 1, or a superposition of 0 and 1 at the same time. These additional states make quantum cpus faster than even the most powerful supercomputers.

Quantum bits (qubits) used within a qcpu are fragile requiring them to be shielded from heat and electromagnetic interference in near absolute zero temperatures. IBM best reported attempt at creating a qCPU has eight qubits, severely limiting its ability. Another company, D-Wave claims to have created a quantum CPU with a 1,000 qubits but on closer testing, the qubits do not “appear” to behave correctly.

Superposition and entanglement

It’s only when we review quantum particles (atoms, electrons, photons) that we find the behavior in nature known as superposition and entanglement that a quantum CPU and its qubits demonstrate.

Superposition is the ability of a quantum CPU to be in multiple states at the same time (0, 1, or a superposition of 0 and 1 at the same time).

Entanglement is an extremely strong correlation that exists between quantum particles. Two or more quantum particles can be inextricably linked. The particles remain perfectly correlated even if separated by great distances. The particles are intrinsically connected in instantaneous, perfect unison, regardless of distance.

What is required to build a quantum computer or qcpu?

Qubits that can be controlled.  They are notoriously difficult to manipulate since any disturbance causes them to fall out of their quantum state (or “decohere”). IBM is working on quantum error correction which examines how to stave off decoherence and combat other errors. One design encodes some imperfect physical qubits into a logical qubit that can perform quantum computations without errors, which could serve as the foundation for quantum computing.

“What we’ve done thus far is to demonstrate some of the concepts of error correction and detection,” says Jerry Chow at IBM’s Thomas J. Watson Research Center in Yorktown Heights, New York. “What we’re doing with this program is aiming for larger system sizes which permit the ability to encode a logical qubit.”

Chow says they need around 20 physical qubits to create one logical qubit, but packing the qubits close together is difficult. “When you put many qubits together, it becomes more difficult to get the correct behavior. “How to properly engineer this larger chip is a big challenge.”

“It’s becoming more of a competitive field now,” says Chow. “It’s a healthy type of competition that will push the entire field forward.”

Why are powerful computers needed?

This level of computer power is useful in a number of fields such as AI, complex modeling simulations involving thousands of independent variables such as air traffic control, molecular modeling (creating faster Quantum computers) and dealing with large numbers.

Molecular modeling

Quantum computers will be able to simulate quantum systems. Simulation of quantum systems will allow the study of the interactions between atoms and molecules. This could help design new materials, such as superconductors that work at room temperature. This could also be used for the development of new drugs.

Handling large numbers

Calculating the factors of a large number (500-digits) is considered near impossible for classical computers. In 1994, mathematician Peter Shor (MIT), calculated a working quantum computer could factor large numbers. Factoring large numbers is the basis for much of our present day cryptography. RSA encryption, the method used to encrypt credit card numbers for shopping online, relies on factoring. Since factoring is difficult, unless someone used a quantum computer and Peter Shor’s algorithm, current RSA encryption remains secure.

Decode encrypted data

Although certain aspects of classical cryptography would be jeopardized by quantum computing,  quantum mechanics would offer new highly secure cryptography.

Say party A and party B share a long string of random zeros and ones, a secret key.

As long as they only use this key once to send a secret message, no eavesdropper will be able to decipher the message. The main difficulty with the one-time secret key is the distribution of the secret key. Quantum Key Distribution (QKD) allows for the distribution of random keys at a distance.

Secure encryption

Quantum key distribution relies on another interesting property of quantum mechanics, any attempt to observe or measure a quantum system disturbs it.

The Institute for Quantum Computing (IQC) is home of one of the few QKD prototypes in the world. “Alice,” a device located at IQC headquarters, receives half of the entangled (highly correlated) one of the photons generated by a laser on the roof of a building at the University of Waterloo. “Bob” is housed at the nearby Perimeter Institute, and receives the other half of the entangled photons. Photons have a unique measurable property called polarization.

Since the polarization of each individual photon is random, there’s no way of knowing the unique properties of each photon in advance. But here is where entanglement becomes interesting: if Alice and Bob measure the polarization of the entangled photons they receive, their results will be the same (remember, “entangled” means the particles are highly correlated with each other, even at great distances). Depending on the polarization of each photon, Alice and Bob ascribe either a “one” or a “zero” to each photon they receive. Therefore, if Alice gets a string like 010110, Bob also gets a 010110. Unless, that is, an eavesdropper attempting to spy on the signal. This would disturb the system, and Alice and Bob will instantly notice that their keys don’t match.

Alice and Bob keep receiving photons until their identical keys are long and identical providing ultra-secure keys for encrypting communications.

Google and NASA’s Quantum Artificial Intelligence Lab

Leave a comment