Democratizing the Corporatized Cloud
The world is on the cloud now. Your backups, your music, your photos, your social media…basically everything that you use on a day-to-day basis resides on servers around the world, and very little actually runs on your personal machine anymore. This has had some very positive side effects — we can do things with computing that we never could before due to near-infinite computing resources being available in a more on-demand fashion, and new services have been able to get started because they no longer have to foot the up-front infrastructure construction costs.
However, there are some negative consequences of this cloud revolution, both from a consumer and developer perspective. These pain points with the existing cloud infrastructure led me to embark on a new experimental project — one that will hopefully change the way that cloud works in the future in order to be better for both the quality of service and the safety of users, whether it be a developer or a service consumer, and all while giving the general person the ability to make money for doing nothing other than having a computer.
I’m going to go through (at a very high level) a little bit about the uncertainty that most people have about how the cloud works, how this solution works, and what the implications are for every player. Stay posted over the next few weeks though, because I will break down all of this in greater detail in a number of subsequent articles.
Wait, so how exactly does the cloud work?
There is a pervasive question that most people outside of the technical world (and even some in it) have, which is how the heck does the cloud even work. It seems etherial — I have the cloud, magically stuff syncs to it, and it goes into this cyber world. However, it’s much simpler than a lot of people might have you believe.
At its core, “the cloud” is just someone else’ computing and storage resources. When you sync to the cloud, instead of files being stored on your own personal, local storage, your data is stored on a hard drive in a data centre, or split into pieces and stored on any number of hard drives in data centres, somewhere else in the world. When you are using a cloud-based software solution, the computations and execution of that software is happening on a computer or a number of computers in a data centre somewhere else in the world. These computing and storage resources are, in their largest capacity, controlled by four major players — Amazon (Amazon Web Services), Google (Google Cloud Platform), Microsoft (Azure), and IBM (BlueMix).
So, at the end of the day, if you are using the cloud for anything, more than likely you data is landing onto the servers and hard drives or either Amazon, Google, Microsoft, or IBM, rather than your data residing on your personal computer.
Okay, so nothing is really running on my computer anymore?
That’s a smart observation there, dear reader. In large part, most of the software that you are running today is likely being computed and stored elsewhere in the world, with limited local computation and temporary localized storage. Anything persistent is limited mostly to what can be crammed into a cookie on your browser.
What does this mean for you? On top of your data being somewhere else, your personal computer has a ton of its own resources that are sitting there, not being utilized, while you’re paying to offload these resources to some massive company who is running a power-intensive, massively-scaled, hard-to-maintain infrastructure core.
The corporatized cloud problem?
There are a few glaring issues with this model of centralized, corporatized cloud, which are all problems that the decentralized model addresses.
The first is your data security, privacy, and anonymity. While privacy is kept paramount for most of these companies, with end-to-end encryption, having all of the data and computation centralized to one location makes that location a major target for nefarious parties. If compromised, the amount of data that can be drawn from one of those services is immense, and can be personally compromising to any individual that has used the service which, based on the prevalence of the services, is likely almost anyone in the world. Meanwhile, with all of the data being in one location, all this data can be associated to other data in the system which can lead to even more compromising correlations being made, and a complete lack of anonymity. Also, for government agencies, it gives a single, easy place to draw from which to draw all of the data, sometimes without the knowledge of the organization controlling the resources. All it takes is one compromised machine in the network to compromise the system.
A decentralized model addresses all of these issues. By breaking everything down into fragments and distributing tasks and data across any number of machines that are not centrally managed in a data centre, the system is not sharing data but rather only results. This makes it so there isn’t a single location to tap or hack, making the network far less susceptible to being dangerously compromised, and localizing the types of compromises that could happen, rather than a global compromise of the system.
The second is paying for resources while not utilizing your own. The cloud model is necessary for many applications, but wouldn’t it be nice if you could use your own resources to their maximum potential and make money in the process. After all, you’ve already paid for those resources. A decentralized model allows you to run a client on your computer which will make it part of the cloud network. You get to help run the world’s software using the computer resources that you have that would otherwise be sitting idle, and get paid for the resources that you are providing to the network. For developers, this decreases costs since there aren’t duplicate resources being purchased and maintained by a company. The system simply compensates people for their pre-existing resources.
The third is resiliency. As was seen not too long ago with the massive Amazon Web Services outage which took down many of the world’s internet-based services, having everything centralized means that a failure is catastrophic. With the ad-hoc nature of a decentralized system, resiliency is core to the model of the system, wherein resources appear and disappear constantly and the network is designed to be able to operate with this in mind. It is not possible to have the entire network go down, because that would mean that every computer around the world would have to be taken down. This resiliency can give everyone more peace of mind in knowing that their web applications will be available when they need them. This ad-hoc model also has the side effect of being scalable based on exactly the resources that are needed, which further minimizes the prices of services by only paying for exactly the resources being utilized.
What to expect
Keep posted in coming weeks and months as I talk about the intricacies of this new, disruptive democratized cloud model that can revolutionize how cloud computing works. This is an experiment too, so I would love to hear your feedback as this moves forward. I’m super excited to have you on this journey.
