Meeting the ChallengesBy Alex Fuss | Posted 2009-09-24 Email Print
How Real-World Numbers Make the Case for SSDs in the Data Center
Nature is full of quantum computers, but we are just beginning to figure out how to control them. When we do, we’ll be able to solve once-intractable problems.
Meeting the Challenges
The challenge for companies and organizations working today to build these machines—such as D-Wave Systems, Princeton Center for Theoretical Physics and The Mitre Corp.—is to capture, observe and control qubits in large enough quantities to be useful.
D-Wave has a quantum computer it expects to commercialize in one or two years, according to the late Chris Hipp, who had been director of marketing. The company is working on a 128-qubit chip made of niobium, but it needs to be able to control at least 1,000 connected qubits to handle complex, multivariable, combinatorial optimization problems.
Princeton and Mitre are taking a more controlled approach to harnessing quantum electrons. They expect to have a working version in their labs in a year or two.
Some pressing real-world problem areas include:
Code Breaking: How can the Defense Intelligence Agency, CIA and FBI break the codes our enemies and criminals create with public key encryption methods? Using classical computers to derive the private key for a 50-digit number would take 3 million years, according to Julian Brown in his book The Quest for the Quantum Computer. Quantum computers could do it in minutes.
Package Delivery: How can FedEx or UPS maximize the number of packages they deliver, while minimizing the number of hangers, terminals and planes; the amount of fuel; the number of employees; and the time required to satisfy customers and beat the competition?
Air Traffic Control: How can the Federal Aviation Administration maximize the number of planes it can keep in the air and land safely, with the fewest number of people and the least amount of radar and communications equipment—given the constraints of specific planes, airports and runways, as well as weather and emergency conditions—while minimizing delays and fuel consumption?
Project Scheduling: How can a large consulting firm use its staff resources efficiently, given specific project requirements, various consultants’ skills, vacation schedules and geographical constraints, while maximizing revenue by getting all projects in on time and under budget?
Pattern Matching: How can Google, Yahoo, Microsoft or the Air Traffic Security Administration compare images taken at different times from different angles—from cameras with different resolutions—and determine if the images are of the same person or thing? This requires examining thousands of pixels per pair; mapping those pixels into features; and abstracting the features to compare them to a database of known objects. Today, this can only be done quickly and reliably with human help.
The human brain is designed for rapid pattern matching: People recognize individuals even if they have gained weight, grown a beard or changed their hair color. Classical computers struggle to do this and cannot help with the challenges listed above because they require more computing power than a serial-oriented digital processor can muster. Quantum computers promise to change all that.
Quantum computation is found all around us in nature, which is made up of subatomic particles that we are just beginning to figure out how to control. When we eventually do, we’ll be able to solve once-intractable problems.
Alex Fuss is a managing partner at DigitalThis, a consulting firm specializing in leading-edge technologies and strategic implementations.