Quantum computers and future of system administration
Released in 1971, the Intel 4004 was the world's first microprocessor. I was ten years old and watched on TV as David Scott and James Irwin drove a car on the moon. It was the coolest thing I had ever seen, and I could not stop dreaming about what the future would bring. Technology has come a long way since then, and I believe we are now on the threshold of the next gigantic leap, thanks to quantum computers. Sysadmins, are you ready?
This article takes an exploratory approach to what awaits us once quantum computers become commercially available. I compare our progress with history and what has been, where quantum computers are today, and onwards.
Quantum computers, where are we?
The first few quantum computers are operational, and they are as bulky and clumsy as the early generations of mainframe computers. Those mainframes were water-cooled monstrosities that were ever so sensitive and required enormous amounts of power to operate. Prone to errors, they had to be continuously supervised. Looking at the quantum computers we have today, I sense some similarities.
Most of you probably already know that Google's quantum computer completed a complex computation in 200 seconds that would take even the most powerful supercomputer we have today approximately 10,000 years. Sure, the new technology is fast, but can it do something useful? Yes, it can, although most of us have probably not even begun to understand what that is.
Each of the quantum computers in operation today has its own unique operating system. We already know that this is not an effective way to increase the usage of and development around quantum computers. To counter that, a consortium lead by a company called Riverlane has been working on a universal operating system called Deltaflow.OS for quantum computers in the hopes that it will gain followers and subsequent momentum. You can read more on that here.
How far away are we from quantum computers becoming commercially available? Well, that's up for debate, but the progress in this area is blistering, and the number of resources being poured into it is huge, so I expect us to see the future soon enough.
What's so special about a quantum computer?
Quantum computing is inherently probabilistic, which means it solves challenges based on the most probable outcome while using several dimensions simultaneously. If regular computing can be compared to flipping a coin, where the result is either heads or tails, then quantum computing is calculating the most probable outcome in a sort of "while the coin is spinning"-way. However, quantum computing is more than that: it can be both heads and tails at the same time, and any variation in-between these two states. This concept is known as superposition. This way of working allows a quantum computer to perform millions of calculations simultaneously to work out the most probable solution or even solutions, that in turn, provides us with more options to choose from.
In the world of quantum mechanics, there are phenomena like the quantum state of superposition, quantum entanglement, string theory, and more. The research in this field has good headwind and a lot of progress is being made. The mathematical models on how to use all this are evolving rapidly, and new scientific papers on the topic are being published all the time.
Quantum computers use what are called qubits to harness the power of quantum mechanics. However, the operations inside qubits are unstable and very sensitive. So far, qubits require some special, challenging requirements to function correctly. They need vacuum and temperatures extremely close to absolute zero. They also tolerate no interference, which becomes very complicated when operating on a nano scale with individual photons and electrons.
The difference between bits and qubits is that a regular computer needs eight bits to represent any number between 0 and 255. With eight qubits, a quantum computer not only represents every number between 0 and 255, but it does it simultaneously! This warps my mind, and even though I have read it so many times, I am still confused about how it actually works. But it does.
There is fascinating research on how to make quantum computers operate in near room temperature and how to decrease the sensitivity by improving the error correction models. When all that comes to fruition, the doors to the future will burst wide open.
So we have established that quantum computers behave very differently from ordinary computers. This also requires entirely new ways of thinking, new math, new models for programming, and new methods of administering the whole thing.
I have bothered myself to try and let go of the norms and mental boundaries that have taught me what is, what can, and what cannot be done. Instead, I have started to consider a quantum-powered future to exercise, on a most basic level, the emerging discipline of quantum philosophy.
Nobel prize winning American theoretical physicist Richard Feynman is sometimes quoted to have said, "If you think you understand quantum mechanics then you don't understand quantum mechanics." For me, that is a bit of a relief, since I really struggle to wrap my head around all this. His statement actually provides me with a bit of freedom to say crazy things that might be wrong or impossible to prove with today's mathematics. However, this is the cool thing about science. Quantum philosophers way smarter than I can say outrageous things, and then scientists might prove them wrong, or start working on a new trajectory thanks to a new way of thinking.
The grandfather of quantum theory, also a Nobel prize winner, Max Planck took to the field of physics even though he was advised not to because a professor told him that "In this field, almost everything is already discovered." Well, apparently not, and we are grateful he did not heed the advice.
For the good of mankind, once we get quantum computers working on a broad scale, it will be a breeze to calculate suitable vaccines to new and existing virus, or work out how to improve the distribution and usage of energy. Quantum computers could provide several alternatives for how to distribute and make use of global resources, and maybe they would even take care of it for us.
But what if I look at more abstract things like a parallel universe, time travel, warp speed, or navigation in space using black holes? Quantum computers can most likely do this and much more. However, most of us will probably use them to optimize our playlists, navigate the private economy's caveats, and improve our social interactions on all existing and emerging platforms through avatars and shadowy interpretations of our different social personalities that can be amplified through the right (or wrong) circumstances.
I do believe that quantum computers can change everything. If we can figure out the most optimal solution out of millions for almost any problem, that will change our fate. However, for this to happen, we also need sysadmins who understand and manage quantum computers, and who know what to do when something goes wrong. Sysadmins must know how to migrate old systems, how to provide data and store the results, and how to implement the next upgrade.
So what is the future for me as a sysadmin?
A quantum computer harnesses the power of quantum mechanics, so would it be like sitting next to a black hole when the thing goes wrong? Well, they are absolutely super fast in solving stuff, so that means they are equally super fast when going wrong.
Currently, the challenge is to keep the qubits' integrity, but with that sorted and the system being reasonably stable, we will still have work to do. The days of "reading the log" will be over, because how do you log what's going on in a superposition? Perhaps entanglement with string connections is easier to fix if you can identify that the link is broken – but on the other hand, string theory would also allow for errors to be identical in a remote system without any form of classic connection like cable or wifi. So we definitely need something new and MUCH faster to tame the beast. We need to think outside the box if we are going to succeed in resurrecting a quantum computer gone wrong.
We probably won't be using tape backups or "snapshot" technology to restore operations, because how can you repair something that happens "mid-air" and has several states at the same time? We will still back up the result, but given the fact that a quantum computer can have many results based on probability, the storage volumes will probably explode.
We need tools to know when a quantum computer goes wrong and probably another quantum computer to intervene and, if possible, fix the problem or simply take over operations. But how do we know that the quantum rescue computer is not the one that has gone wrong?
Our role will be to understand why things go wrong and work on improvements. Or will the quantum computer do that all by itself, and we are just the hands and feet that add disks, unpack the new hardware, connect cables, and feed the cat?
In Stanley Kubrick's movie "2001: A space odyssey", co-written by Arthur C. Clarke, the sinister red eye of the on-board computer HAL 9000 began to malfunction and do bad things to the humans on board the spaceship. HAL managed everything on the ship, and that was, of course, a comfortable blessing for the crew, until he went wrong. Perhaps this is so not far off. A quantum computer needs, besides a lot of power, two more things: a vacuum and extreme cold. Now, as far as I know, space offers these conditions, in abundance. Hopefully, we won't have to sit alone in space and sysadmin evil computers, but once we start traversing space, some of us will surely be out there. The malfunction of a "living computer" was not that hard to imagine even back in 1968. The resolution to HAL's malfunction was surprisingly similar to what we do today… a hard reboot.
The outcome from a quantum computer will surely be more precious than gold and need to be kept safe. Every day new data and new analysis made by quantum computers coupled with human decisions that allow us to create new estimates based on quantum computer-calculated suggestions will keep business and life in general moving at a pace we have never experienced.
The exciting thing in this context is that quantum computers can work on any pile of data. It could be encrypted or even bad data that is inconsistent and complicated. A quantum computer would just breeze through it since it can try multiple solutions at the same time. We would break new ground we did not even know existed—provided we can write good enough code for the quantum computers to work on.
Complexity is easy and comes all by itself, and that is probably even more true with quantum computers. Simple is hard, so to run a clean and well-maintained environment will become even more challenging, especially if the quantum computer can bridge the gap between older and newer systems. It's time to start thinking about how this will be done. We already have computers managing computers, and it will be the same in the quantum future. You need one to know one. The main difference will be" speed and power," to quote a famous TV-personality from a British car show.
Mixed landscape and migration
Just because VAX computers have not been manufactured since 2005 does not mean that old systems are no longer used. Moving into the glorious future of sysadmin tasks, we are sure to have a mix of the very old and very new. Migration is a tedious task and having done quite a few myself, I am well aware of the challenges of preparing older and undocumented systems for the move to newer platforms. Unforeseen problems come like a string of pearls. The solutions must be worked out quickly and are often more unconventional and complex than what the project plan had taken into account. Perhaps the new quantum computers will take the edge off migration and resolve incompatible data stacks and programs, so the job is done even before we have taken our finger off the [Enter] button.
To add a comic twist: In a quantum world, the migration is the probability of multiple successful solutions completed before started while they are still ongoing and ready to begin.
Scared of what's to come
Back in the day, we were both intrigued and scared when the first monitoring systems that came with remedial actions became available. Stacks that could hold the backup tapes and ingenious mechanics that could change them were cool to look at, but it was like we were not needed anymore. There was software that read the log files and issued the commands that we usually would do. Staff reductions were on the agenda, and we went from hero to zero, or so it felt. However, it was not long before we were knee-deep in work because of all the issues the monitoring tools came up with and what sort of tosspot actions they tried to invoke to solve "problems." Many times these were issues that the servers had been living with quite happily for a long time but were now shown in red on some dashboard.
Will it be the same when the new generation of quantum computers starts to perform sysadmin tasks? Or perhaps the quantum computers will be so awesome that human intervention is no longer necessary? Will we end up with the short end of the stick, or will we finally have time to explore and investigate all those new things where we never had the time? Can we focus on innovation with the support of quantum computers and get a multitude of options laid out before us to choose from?
When we start to mix artificial intelligence with quantum computers, it may be time for us to look for a different career. Who knows?
How can we keep things safe?
Red Hat Principal Product Security Engineer Huzaifa Sidhpurwala wrote a blog post in 2017 on Post Quantum Cryptography where he talked about how quantum computers will be able to decrypt most of the cryptographic systems in use today. There are initiatives in progress to combat this expected weakness, but I believe it takes one to know one. So again, you most likely need a quantum computer to outsmart another quantum computer.
We see how the new technology of quantum computers is emerging, and it will change everything because it operates in a completely new way. To sysadmins, a quantum computer will be very different as it can work in pairs with other computers without physical connections. Quantum computers approach a problem by checking all possible solutions at once and then opting for the most probable outcome. Backups will change, patching, and more. Figuring out when a quantum computer goes wrong can be done through error-correcting code but might also require another quantum computer to sort it out or take over operations. The future is just around the corner, so if you are a sysadmin, I suggest you get yourself busy understanding this new technology. When it becomes commercially available, those of you who are prepared will have a tremendous advantage.
If you want to explore more about quantum computing, there is a fascinating Ted Talk by Shohini Ghose, where she, in just 10 minutes, explains quantum computers. There is, of course, a wiki page you can use as a starting point to navigate the subject.
[ Want to test your sysadmin skills? Take a skills assessment today. ]