The modern technology is fascinating if one thinks about the power that every person carries in their pockets. Even a hand-sized smartphone, today, has more computing power than the military computers from 50 years back, which occupied entire rooms.
However extraordinary these leaps in the knowledge of science and research might be, there are still some problems which cannot be solved by these computers. It is believed that quantum computers are the solution to these problems
The computer industry is often found to be compelled towards making computational operations more efficient since the classical methods have reached their limits of energy efficient. The Semiconductor Industry Association claims that the world will have no energy to run any of the
machines by 2040. This is the primary reason for the computer industry to struggle in order to develop operational quantum computers at a commercial scale
Problem Statement and Opportunities
Over the course of the past sixty years, a tremendous increase has been seen in the power of classical computers. Their computational power is believed to have increased by millions of times, but there are some complicated problems which are too complex for these computers to solve. Even though research has found methods to write programs which would give the correct solution, but
the computational power is limited, and it would take billions of years to finish the calculation.
Today’s supercomputers are made up of thousands of connected resources such as processors, graphics cards, ultra-high bandwidth network connectors etc. and their speed has grown exponentially over the past few decades.
Since the inception of supercomputers in 1946, it has grown significantly and now it resolves highly complex problems such as oil prospecting, life science research, industrial design etc. IBM’s SUMMIT has taken the crown of the world’s smartest and most superpower computer by beating China’s Sunway TAIHULIGHT.
So, what’s the configuration of the modern day supercomputer? A question often asked, and below is a brief view
These modern world supercomputers can do amazing things need high-performance computing workload, from traditional simulation to advanced AI at unprecedented scale and speed on massive data sets.
While classical computing (includes supercomputers) have done amazing things, there are still problems that we cannot solve using classical computing
Quantum Computing — Two Problem Statements
The limitation of computing power has raised many problems in the areas of general optimization, protein folding, quantum simulation, deep learning, and machine learning.
Problem 1: Optimization — Best possible solution out of many possible solutions
One of the primary problems fall in this category is infamous Travelling Salesman Problem.
This problem finds the answer to the question that for a given list of cities and the respective distances between each pair of those points, what will be the shortest route that the salesman can adopt in order to visit every city and return to the starting point. In computational complexity theory, such a problem is called the NP-hard problem, and it has important results in theoretical computer science
This problem first came to know in 1930 and has been studied intensively since then, to the day. It is important in optimization, where it serves as a benchmark for several methods. Even though there is a computational hurdle for the feasibility of this problem’s solution, a large number of algorithms are known which could solve this problem to a limited extent. For instance, with the currently known algorithms, this problems can be solved for tens of thousands of cities, completely.
For millions of cities, the problem can be approximated not more than within a small fraction of 1 percent. This is not a hypothetical problem, The TSP, but it has a number of applications even when it is applied in its purest form. Areas where this problem finds applications are planning, manufacturing of microchips and logistics. In a slightly modified form, the problem represents many other areas including the DNA sequencing. In its applications, the points or cities represent the customers for logistics, DNA fragments for sequencing or soldering points for microchips. The distance represents the
traveling times for logistics, a similarity measure for the DNA and cost for the microchips.
Similar problems can also be related to astronomy when astronomers want to observe the many sources and find out the minimum time required to move the telescope between all the sources. Additional constraints can be imposed, in many applications, to limit the resources or time durations.
A whole branch of computer science, called computational complexity, is focused on solving such complex algorithms. The branch classifies them on the basis of their running time — exponential time or polynomial time.
Quantum computers employ a concept called ‘inherent parallelism’ which is a computing concept. This inherent parallelism is obtained when a quantum bit, is kept in a state of superposition, and by involving other clever algorithms to increase the processing speed to find the solution in a shorter time duration.
Problem 2: Exponentiality
A famous fable below is a good interpreter of this problem.
The emperor agreed, amazed that the man had asked for such a small reward — or so he thought. After a week, his treasurer came back and informed him that the reward would add up to an astronomical sum, far greater than all the rice that could conceivably be produced in many many centuries!
Now, there can be a simple solution to this problem — basic addition. Since there are 64 squares on the chessboard, and the number of grains is being doubled with each subsequent square, then a series can be developed and summed as: 1 + 2 + 4 + … up to 64 times. The total number of grains is then calculated to be around 18,446,744,073,709,551,615.
Contrary to popular belief, “exponential” does not mean fast. It means the growth rate is proportional to the size.
There are things which have a constant growth rate, while there are other things which grow faster since the number of these ‘things’ also tends to grow. When the growth rate increases and does not remain constant, relative to the growing total number, the growth is called to exponential
Exponential growth is complicated yet powerful. It is important to consider here that the exponential growth starts off rather slowly, but it can generate enormous quantities in a short period of time — leaving the observers shocked. In the above example, the value comes out to be 1645 times more than global wheat production in 2014. Classical computing can be ruled out as an economical method to fast process to arrive this problem.
Ability to solve such complex problems of a high magnitude that are beyond the capabilities of supercomputers, and more importantly much efficient than supercomputers of this era makes quantum computing relevant
The building blocks of quantum compute is built over the concepts of entanglement and superposition
What is Quantum Computing?
When matter is discussed at an atomic or sub-atomic level, the general behavior of things change. One of the interesting underlying reasons is the co-existence of these particles in more than one states at a single time. This is exactly the same ability which the quantum computers harness to pull off extreme processing speeds
A quantum computing is the mechanism that performs calculations based on the behavior of particles at the sub-atomic level — leveraging the techniques of superposition and entanglement
The conventional computers are based on binary digits (0 or 1), however, the quantum computers use a different kind of bits — quantum bits.
An example can clarify the difference between these two kinds of bits.
Let us assume a sphere; a bit can exist at either of the two sides (or poles) of the sphere. But a quantum bit behaves differently. For this case, it is not possible to actually single out the state of a quantum bit since it can exist at any point on the sphere (O’Brien, 2007).
This also means that a computer based on qubits will be able to store huge volumes of information, and will consume lesser power than the conventional, classical computers. The traditional physics laws are challenged when the computing enters the quantum area. As an application of this, significantly faster processors will be possible to create, more than a million times faster. This sounds incredible, but the main problem is that quantum computing is not so easy to implement.
Currently, the modern technology has made transistors almost as small as atoms, which aid in switching and memory units of computers. But for quantum computing, an entirely new way of thinking and building computers, is required. Even though the conventional computers can process information at tremendous speeds, but inside, it just operates on a sequence of 1s and 0s. These bits represent two basic states — ON and OFF.
Quantum computers are not meant to substitute the modern array of conventional computers, but they are intended to be a different tool — a tool to solve complex problems which are beyond the limits of classical computers
Boolean algebra provides the basic and fundamental principles for the operation of classical computers. Usually, those computers operate with a 7-mode principle, based on logic gates; though, only three modes (COPY, NOT, and AND) can be used to implement the same operation
It is necessary to process data in an exclusive binary state, which can either a 0 for false and 1 for true. These values are called binary digits, or shortly, bits. A conventional computer is made up of millions of transistors and capacitors, all of which can only hold a single state — either 1 or 0.
Silicon chips are reaching their limits, and we’re entering into sub-atomic era..
Unlike the classical computing, the quantum computing can be operated with 2-mode logic gate — which means it can operate in 1 or 0 or a state in between — this is where the principles of entanglement and superposition come into play
For classical computing, the output is governed by a Boolean function, which is essentially a logical problem with either YES or NO for its answer. Of course, these answers can also be represented as TRUE / FALSE or 0 / 1. Generally, these gates are implemented using electrical circuits, which are integrated into the CPU and other computational components of the computer. This is true for most of the computers, starting from Babbage’s Difference Engine, the first generation computer ENIAC, the IBM’s chess-playing computer Deep Blue or the IBM Quantum Computer, announced in 2017. While those in quantum computers, the operations in on Qubits, and, therefore, quantum bits can handle data in a way that is entirely beyond possibility for classical computing — where it achieves multi-parallelism
Real World Applications
Eric Ladizinsky, the co-founder of D-wave, a popular quantum computing company, spoke at the WIRED conference, in 2014, about the changes that quantum computer will bring in the coming future.
He said that if a letter X is written on the page of a book, and the book is hidden among 50 million other books in a library, then it will not be possible for a classical computer to find the X written on the book. However, if a quantum computer is used, the scenario will be different. It will be like having 50 million parallel universes where you can look into each book, go through each page and possibly find the X.
Quantum computing is based on a similar concept, more or less.
This is to say that in the given scenario a quantum computer will split itself into 50 million different versions, to make the work astonishingly faster and easier. Surprising, but quantum computers are capable of these properties, which are beyond the possibilities for any of the classical computers
Based on the above properties of quantum computers, they can be used in a number of fields — Weather Forecasting, High Frequency Trading, Artificial Intelligence, Cyber security etc.
The quantum world is extraordinarily amazing when one thinks of the power and possibilities but it is menacing at the same time. The current challenge is to deal with the quantum security by developing such quantum-safe encryption algorithms, which are resilient against quantum attacks. Although it might sound easy, developing quantum algorithms can be a complicated thing. One specific solution to this problem is to come up with such complex encryption which would exponentially increase over time, making it difficult for a quantum computer to solve. However, creating such an algorithm will be a difficult task
At the present time, it is difficult to say for sure that how much quantum computing will revolutionize our world with its speeds and operating powers. The quantum world is an entirely new realm of physics and might yield solutions which have never been possible before.
When one considers the revolution that classical computers have brought since their invention, operating on just 0s and 1s, one can only begin to imagine the exceptional possibilities and opportunities that will be possible by tremendously fast processing power.
However, we know that it will have groundbreaking results in every field of work and industry, and will leave huge impacts on the way that we live. From inventing medicines to materials and space exploration to weather forecasting; quantum computers will affect every field. It is not foolish of the world’s largest companies such as Google, IBM and Intel to invest in this technology. They are anticipating great changes — changes that will create solutions to problems that we have not even identified yet.