2018-07-30
量子计算机更快，更好，具有潜在的危险性，能够超越传统计算机的边界，但是能到什么程度呢？本文将探索一些可能性。
数字计算有一些严重的局限性。虽然过去几十年来的技术进步令人印象深刻，如体积更小、处理器更快、用户界面（UI）更人性化、内存和存储空间更大，但是量子计算机可以更好地解决一些问题。
一方面，量子计算机比传统的计算机更快。它们也能够解决传统计算机在合理的时间内处理不好或不能处理的问题。
英特尔实验室量子硬件总监吉姆�克拉克（Jim Clarke）说：“量子计算利用物理学的基本原理，以新的方式解决复杂的计算问题，如发现疾病如何发展并创造更有效的药物来对抗它们等。一旦量子系统上市，人们就可以用它们模拟自然，以促进化学、材料科学和分子建模方面的研究。例如，帮助创建新的催化剂（以螯合二氧化碳）或室温超导体。“
量子计算还将推动业务优化，有利于机器学习和人工智能的发展，并改变密码学的格局。
Deloitte董事总经理大卫�沙茨基（David Schatsky）表示，常见的是有多个可能答案的优化问题，任务是找到正确的答案。例如投资管理、投资组合管理、风险缓解以及通信系统和运输系统的设计。物流公司正在探索路线优化，而国防工业正在考虑通信应用。
沙茨基说：“一年前，量子计算被更多地认为是一种物理实验，但是这种观点迅速改变。在过去的三个月中出现了突破性的进展，包括基础工程突破和商业产品公告。”
可以确定的是，我们无法很快地拥有量子计算机，但是任何使用浏览器的人都可以通过云访问IBM的5和16量子位（qubit）计算机。今年早些时候，IBM宣布将推出世界首个商用量子计算系统IBM Q。IBM还宣布，它已经构建并测试了两个量子计算处理器，包括16量子位开放处理器，供公众使用，并为客户提供了17量子位商用处理器。
根据《自然》杂志上的IBM文章，科学家们成功地使用了一个7量子位处理器来解决氢化铍（BeH2）的分子结构问题，这是迄今为止在量子计算机上模拟的最大分子。
IBM Systems的量子计算技术战略与转型副总裁兼首席技术官斯科特�克劳德（Scott Crowder）说：“量子计算目前还处于早期阶段，但它将会迅速扩展。当您开始谈论数百或数千个量子位时，您就可以解决传统计算机处理不好的问题了，如量子化学和某些类型的优化问题（指数问题）。”
指数问题是指随着元素数量呈指数级变化的问题。例如，根据目标，可以以多种方式优化涉及50个地点的路线，例如识别最快的路线。这个看似简单的问题实际上涉及到一千万亿种不同的可能性，传统计算机很难处理，克劳德说。
2015年，英特尔与荷兰学术伙伴QuTech合作。此后，英特尔实现了多个里程碑，例如演示集成低温CMOS控制系统的关键电路块、开发了自旋量子位制造流程和超导量子位的独特封装解决方案（在2017年10月10日推出的17位量子超导测试芯片中进行了展示）。一周之后，在加利福尼亚拉古纳的《华尔街日报D.Live大会上，英特尔首席执行官布莱恩�科兹安尼克（Brian Krzanich）说计划在2017年年底之前推出49量子位量子芯片。
英特尔公司的克拉克表示：“最终目标是开发一种商业化的量子计算机，一种通用的、能够影响英特尔底线的量子计算机。”
为此，英特尔与QuTech的合作涵盖了从量子位设备到整体硬件架构、软件架构、应用和互补电子产品的领域。
克拉克说：“量子计算本质上是并行计算，能够解决传统计算机无法处理的问题。但是，实现量子计算需要将优秀的科学、先进的工程技术和持续发展的传统计算技术结合起来，英特尔正在通过各种合作伙伴关系和研发计划朝着这一方向努力。
关于量子计算机是否会使当前的加密方法过时，是有争议的。以暴力破解攻击为例，黑客不断猜测密码，并使用计算机加速这一过程。量子计算将进一步加速这种攻击。
“几乎今天使用和部署的所有安全协议都容易受到量子计算机的攻击，”IEEE量子标准工作组主席威廉�赫利（William Hurley）说，“量子信息可以让我们以完全无法攻破的方式来保护信息，甚至是对抗量子攻击。”
按照这些原则，我们正在努力开发一种不利用量子力学的新型安全协议。赫利表示，他们正在使用非常困难的数学问题，即使是量子计算机也无法解决，这被称为“后量子加密”。
IEEE量子标准工作组正在研究其它量子技术，包括量子传感器和量子材料。该研究所汇集了物理学家、化学家、工程师、数学家和计算机科学家，以确保研究所能够迅速适应变化。
Deloitte的沙茨基表示，合成生物学和基因剪辑也有潜在的危险性，主要是因为这些能力的发展速度比人们理解如何明智地应用这些技术的能力更快。许多新兴技术也是如此。
量子计算正在迅速发展，所以明智的做法是思考这种能力对企业能够带来什么好处。现实情况是，没有人知道量子计算的所有使用方式，但它最终会影响到许多不同行业的企业。
量子计算机是否会取代传统计算机？还是说两者会共存？在可预见的未来，共存是这个问题的答案，因为二进制和量子计算机各自擅长不同的问题。
您认为量子计算的“杀手锏”是什么？做过实验吗？您想解决什么问题？我们很乐意在评论部分与您讨论可能性。
[仍然在纠结量子计算到底是什么？]
以下是来自《麻省理工科技评论》的最简洁的定义之一：
“量子计算的核心是量子位（qubit），这是一种基本的信息单元，类似于您的计算机中由晶体管表示的0和1。由于两个独特的属性，量子位比传统的位数多得多：它们可以同时表示1和0，它们可以通过称为量子纠缠的现象影响其它量子位，这使得量子计算机在某些类型的计算中能够更快地获得正确答案。”
10/26/2017
03:00 PM
Faster, better, potentially dangerous. Quantum computers will push way past the boundaries of traditional computers, but to what ends? We explore some of the possibilities.
Digital computing has some serious limitations. While the technology advances made over the past few decades are impressive such as smaller footprints, faster processors, better UIs and more memory and storage, some problems could be solved better by quantum computers.
For one thing, quantum computers are faster than classical (traditional) computers. They are also able to solve problems that classical computers can't do well or can't do within a reasonable amount of time.
"Quantum computing exploits fundamental laws of physics to solve complex computing problems in new ways, problems like discovering how diseases develop and creating more effective drugs to battle them," said Jim Clarke, director of quantum hardware at Intel Labs."Once quantum systems are available commercially, they can be used to simulate nature to advance research in chemistry, materials science and molecular modeling. For instance, they can be used to help create a new catalyst to sequester carbon dioxide or a room temperature superconductor."
Quantum computing will also drive new levels of business optimization, benefit machine learning and artifical intelligence, and change the cryptography landscape.
David Schatsky, managing director at Deloitte, said the common thread is optimization problems where there are multiple probable answers and the task is to find the right one. Examples include investment management, portfolio management, risk mitigation and the design of communication systems and transportation systems. Logistics companies are already exploring route optimization while the defense industry is considering communications applications.
"A year ago [quantum computing] was thought of more of as a physics experiment [but] the perception has changed quickly," said Schatsky. "In the last 3 months there have been a flurry of breakthroughs including fundamental engineering breakthroughs and commercial product announcements."
Test drive a quantum computer today
It's probably safe to say that none of us will have a quantum computer sitting on our desks anytime soon, but just about anyone with a browser can get access to IBM's 5 and 16 quantum bit (qubit) computers via the cloud. Earlier this year, the company announced IBM Q, an initiative intended to result in commercially available quantum computing systems. IBM also announced that it had built and tested two quantum computing processors including the 16 qubit open processor for use by the public and the 17-qubit commercial processor for customers.
According to an IBM paper in Nature, scientists successfully used a seven-qubit quantum processor to address a molecular structure problem for beryllium hydride (BeH2), the largest molecule simulated on a quantum computer to date.
"It is early days, but it's going to scale rapidly," said Scott Crowder, vice president and CTO, Quantum Computing, Technical Strategy & Transformation at IBM Systems. "When you start talking about hundreds or low thousands of qubits, you can start exploring business value problems that [can't be addressed well using] classical computers such as quantum chemistry [and] certain types of optimization problems that are also exponential problems."
An exponential problem is one that scales exponentially with the number of elements in it. For example, planning a route involving 50 locations could be optimized in a number of ways depending on the objective, such as identifying the fastest route. That seemingly simple problem actually involves one quadrillion different possibilities, which is too many possibilities for a classical computer to handle, Crowder said.
Intel is making progress too
Intel teamed up with QuTech, an academic partner in the Netherlands in 2015. Since then, Intel has achieved milestones such as demonstrating key circuit blocks for an integrated cryogenic-CMOS control system, developing a spin qubit fabrication flow on Intel’s 300mm process technology and developing a unique packaging solution for superconducting qubits that it demonstrated in the 17-qubit superconducting test chip introduced on October 10, 2017. A week later, at the Wall Street Journal D.Live conference in Laguna, Calif., Intel CEO Brian Krzanich said he expects Intel to deliver a 49-qubit quantum chip by the end of 2017.
"Ultimately the goal is to develop a commercially relevant quantum computer, one that is relevant for many applications and one that impacts Intel’s bottom line," said Intel's Clarke.
Toward that end, Intel’s work with QuTech spans the entire quantum stack from the qubit devices to the overall hardware architecture, software architecture, applications and complementary electronics that workable quantum systems will require.
"Quantum computing, in essence, is the ultimate in parallel computing, with the potential to tackle problems conventional computers can’t handle," said Clarke. "But, realizing the promise of quantum computing will require a combination of excellent science, advanced engineering and the continued development of classical computing technologies, which Intel is working towards through our various partnerships and R&D programs."
Decryption and other threats
There is a debate about whether quantum computers will render current encryption methods obsolete or not. Take a brute force attack, for example. In a brute force attack, hackers continually guess passwords and use computers to accelerate that work. Quantum computing would accelerate such an attack even further.
"Virtually all security protocols that are used and deployed today are vulnerable to an attack by a quantum computer," said William "Whurley" Hurley) Chair of the Quantum Standards Working Group at the IEEE. "Quantum information allows us to secure information in ways that are completely unbreakable even against a quantum attack."
Along those lines, there are efforts to develop a new type of security protocol that doesn't necessarily leverage quantum mechanics. Hurley said they're using extremely difficult mathematical problems that even quantum computers won't be able to solve, which is referred to as "Quantum-ibmSafe Cryptography" or "Post-Quantum Cryptography).
The IEEE Quantum Standards Working Group is working on other quantum technologies including, quantum sensors and quantum materials. The research institute has brought together physicists, chemists, engineers, mathematicians and computer scientists to ensure that the institute can adapt rapidly to change.
Deloitte's Schatsky said synthetic biology and gene editing are also potentially dangerous, mainly because capabilities can be developed faster than one's ability to understand how to apply such technologies wisely. The same could be said for many emerging technologies.
Quantum computing should be on your radar
Quantum computing is advancing rapidly now so it's wise to ponder how the capabilities might benefit your business.聽 The reality is that no one knows all the ways quantum computing can be used, but it will eventually impact businesses in many different industries.
Will quantum computers overtake classical computers, following the same evolutionary path we've seen over the past several decades or will the two co-exist? For the foreseeable future, co-existence is the answer because binary and quantum computers each solve different kinds of problems better than the other.
What's your take?
What do you envision quantum computing's "killer app" to be? Have you experimented with the technology? If so, what problem(s) are you trying to solve? We'd love to discuss the possibilities with you in the comments section.
[Still wrestling with what quantum computing really is?]
Here's one of the most succinct
definitions, drawn from the聽MIT Technology
Review:
"At the heart of quantum computing is the quantum bit, or qubit, a basic
unit of information analogous to the 0s and 1s represented by transistors in
your computer. Qubits have much more power than classical bits because of two
unique properties: they can represent both 1 and 0 at the same time, and they
can affect other qubits via a phenomenon known as quantum entanglement. That
lets quantum computers take shortcuts to the right answers in certain types of
calculations."
附件：
微信公众号