Introducing dHost, The Decentralized Alternative To AWS & Google Cloud
Since the inception of the World Wide Web, data centers and hosting providers have been utilized by developers to build and publish experiences that have changed how each and every one of us experience daily life. Like any technology, the way we have distributed data over the past decade has changed, for reasons ranging from performance and security, to concerns surrounding data control, along with the recent uprising of cancel culture. Lately, hosting providers like Amazon’s AWS, Google Cloud and others have been taking websites and web apps offline that don’t exactly reflect their corporate values. To add to this, the billions of dollars in servers they power on a daily basis is by all means inefficient, and by today’s standards, there are better ways to distribute and host the world’s online experiences. Even worse, the web hosting services provided by these companies typically experience various forms of downtime, are expensive and ultimately susceptible to outside attacks by malicious actors.
With the inception of dWeb’s 3.0 release (https://github.com/distributedweb/whitepaper) and dMesh’s graph-based Kademilia DHT, a truly blockchain-less web was launched that brought with it many of the services developers typically utilize via cloud computing providers, from file storage and databases to backend services and domain name systems. The main difference between the two is that dWeb-based services are hosted amongst a network of peers and their devices, rather than hosting providers and data centers. The dWeb is also a far more efficient and modern way of hosting websites, web apps and data, considering they are redundantly hosted amongst a network of personal computers in people’s homes, rather than expensive server farms. At the very same time, distributed and decentralized computing services result in much better performance grades than centralized computing services, do not suffer downtime as long as they have persistent seeding, are not susceptible to attacks due to their distributed and immutable nature and cannot be controlled or removed from the web since distributed data lacks a central point of control and/or failure.
dHost is a new distributed computing platform that brings all of dWeb’s distributed computing technologies under a single roof, giving developers an easy way to deploy dWeb-based services across a network of “hosters”, who in turn make their personal computers available to the platform as persistent peers. Hosters select the number of resources they want to make available and developers ultimately pay to deploy their services on a per-peer/per-minute basis across as many peers as they choose, depending on the amount of storage space that’s being used on a hoster’s computer(s). In exchange for persistent distributed computing services, developers pay in a new currency known as dWebCoin (DWEB) or other currencies like RISE (RIX) or Bitcoin (BTC), which is split by the dHost platform and the hosters who host their data, where payouts occur on an per-minute basis.
dHost presents a new “hosting economy”, allowing hosters to lease out their own personal computers to developers, just as AirBnb allows homeowners to lease out their homes to renters. dHost’s platform also simplifies and decentralizes the development and deployment process for developers who are looking to move their websites, apps and data to Web 3.0, without all of the complexities that typically derive from replicating distributed data sets or messing with the blockchains of the past. It also expands the dWeb to the homes of everyday people across the world, which could eventually create a bigger infrastructure footprint than the combined global footprint of all data center facilities, while also giving developers a way to move beyond expensive, centralized and tyrannical hosting providers — where their apps can exist in a more secure fashion and their online experiences can flourish.
Hosting on the regular web is so much more different than hosting on the distributed web, since hosting on the distributed web centers around peering data to the computers of others, where the execution or consumption of applications and their data eventually takes place, rather than the traditional model where the execution of application code and its data is handled by remote servers. This offloading of CPU and RAM-based requirements onto the users of an application, greatly reduces the costs associated with distributing an application and its data on the distributed web, where only the storage and bandwidth utilized by those who are peering an application’s code and data actually matters. Our work for the past year or so has centered around the development of several distributed data structures that in turn could be utilized by app developers to do pretty much anything. The launch of software like dHub (https://npmjs.com/package/dhub), which has the capability of storing, manipulating and distributing these data structures amongst a network of requesting peers, has enabled platforms like dHost to exist.
On dHost, apps could pay for 20GB of storage via 8 peers for their application code (in a dDrive [https://npmjs.com/package/ddrive]), at around $5/per-month/per-peer (8 x $5/month), which would ultimately end up costing them $40/month. Those 8 peers would provide persistence to the application, which would also be peered by its users. At the same time, the application could utilize a DHT as a multi-writer distributed datastore, which would grow organically (a truly free datastore), or could pay for 1TB of space, via 8 peers, to create persistence for a single-writer distributed datastore (like a dTree [https://npmjs.com/package/dtree]). This would cost around $40/per-month/per-peer, which would ultimately equate to $320/month if 8 peers were utilized. Those 8 peers would provide persistence to the application’s data, which would also be peered by the application’s users, just like its code is. Keep in mind, even though the peer is being paid for 1TB in space, most of those who pay for this level of storage probably will not utilize an entire terabyte. This is true in most hosting environments, where hypervisors are “oversold” and end up hosting machines that total more resources than actually physically exist on the underlying hypervisor they’re deployed upon.
When I say data or an application’s code is being peered by the application’s users, I mean that the users of the application are automatically seeding its dDrive and/or dTree, by simply accessing the data structure via dBrowser (http://dbrowser.com), since dBrowser does this seeding in the background by default. While this sort of seed growth will happen over time, some applications and most websites will need a constant level of persistence and cannot rely on “organic seeding” aka “free seeding”, due to their lack of popularity. Regardless, it’s important that an application has at least one persistent peer and I cannot stress this enough. This will be the case for most corporate websites that users simply don’t care to seed long term or websites that simply aren’t visited that often. Whatever the case may be, the costs I just discussed are much lower than what you see with web hosts on the centralized web and it’s not even close. A website could be hosted on the dWeb persistently, using a single peer, for $4/month — and never go down. The best part about dHost is, it handles peer persistence autonomously, so that if the peer you are paying for goes offline, you’re automatically shifted to another. This is why dHost’s services are paid out on a per-minute basis and just another reason why dHost is different from the hosting providers of old.
Ultimately, we’re moving away from the “hourly cloud model.” With that said, consider the scenario on the centralized web where your app is designed to automatically scale up and/or scale down as its resource requirements change via AWS or Rackspace’s public cloud API (as traffic increases). All of this has to be built in to your application code at some point and can be a pain to manage for some, or for others, confusing to implement. With today’s web, this is a major requirement for most developers, whether they are developing websites or simply working with a Wordpress-based ecommerce store. Companies demand scalability and they definitely hate downtime. With dHost, an application can be exported and minified through a bundler like Webpack, placed in a dDrive and deployed on dHost without the need to worry about scaling.
dHost scales based on peer availability and network popularity, so if a peer goes offline, that peer is immediately replaced with a peer who is online to maintain peer persistence for the underlying dWeb data structure. Likewise, if a dWeb data structure is experiencing a heavy request load across its current seeds, its peer count can be increased automatically to increase application performance. This helps us depart from the model of old which requires that applications manage their own scaling strategies, since an application’s network analytics are publicly available. This is what allows dHost to scale an application’s peer persistence levels amongst computers across the globe, just as cloud hosting platforms are capable of scaling up an application’s computing resources across multiple regions and their available hypervisors.
More importantly, dHost allows ANYONE in the world, regardless of their level of technical expertise, to earn various cryptocurrencies like DWEB, RIX and BTC by simply hosting the data structures of others. We call these people “Hosters” and dHost will feature a special “Hoster Interface”, just like Uber features a “Driver Interface”, so that hosters can bring their computer online for the purpose of peering the dWeb-based data structures of others. The Hosting Interface asks the Hoster how many resources they are willing to sell on their device, and a virtual container (a virtual machine) is deployed on their device with these resource limitations where dWeb data structures will eventually be deployed and peered. The dHost platform chooses which peers it ultimately deploys services across. At the very same time, Hosters can increase/decrease these resource limitations at any time (change the container limits), view their earnings (updated on a per-minute basis), and take a look at what their device is currently hosting. While Hosters do not choose who they host, they can stop hosting specific kinds of content if they choose to. In the case that they do choose to “dehost” a service, dHost will simply find another peer.
This essentially would allow anyone to earn hundreds or even thousands of dollars per-month (depending on how much storage space they provide to the network), if not more, for simply leasing out portions of their hard drive. In fact, one could simply purchase an external hard drive, plug it in and sell all of it for quite the profit! If you were around for the launch of Bitcoin, as I was years ago, you will remember how many people were amazed that they could make money with their personal computers by simply purchasing expensive mining equipment and running the Bitcoin core software. Obviously, it’s much more complicated these days, where you would have to pump a significant amount of capital into a mining farm to simply break even. Therefore “hosting” is such a better alternative to “mining,” considering peers on the network are not competing to host services, instead, peers are chosen evenly. At the same time, the more space a peer provides, the more money they can technically earn. In this way, peers are selling space in the digital world, just as a realtor sells space in the physical world.
If you have been following along with the Let Freedom Stream (LFS) series and you’re a developer, you probably saw this coming. As distributed services on the dWeb expand (we just introduced dTrees and other structures in the 3.0 release), dHost will provide peering for these services within dHost, just as cloud platforms like AWS provide new services as they come into existence on the centralized web. An example of this would be the upcoming launch of dServe, which will allow anyone and everyone to host a dDrive on the centralized web, via a service that simply serves up a dDrive over HTTP/S, via a dDrive’s peers. This would allow a decentralized website or application to be accessed from a regular web browser over https://, rather than a decentralized web browser like dBrowser. This is just a single example of how dHost will expand over time and how dHost is destined to change web hosting forever.
I have spent many years of my professional career in the web hosting business and I know how cut throat it is. From the days when Rackspace elected to steal our “EasyCloud” concept that we presented to them at “The Castle” in San Antonio (which eventually became known as “Cloud Sites”), to Amazon’s patent attorneys who refused to inform us of their obvious conflicts of interest when they chose to steal our Virtual Desktop Infrastructure (VDI) project (which eventually became known as “Workspaces”), I’ve been waiting a long time for payback. The creators of the dHost project came to me and asked if I would be interested in being the lead developer on the project, considering my knowledge set surrounding dWeb’s protocols — and I was floored when they asked if I would accept their offer — especially since they chose to build on the dWeb.
I’m excited to be a part of an alternative to these tyrannical companies and nobody knows more about these tyrants than I do. Since my days of working with NASA’s OpenStack project and others, to my controversial blockchain days, my life and professional career have consisted of distributed networks and cloud computing platforms — and now I get to combine the best from both worlds. For the past few months, we have been freely distributing RiseCoin (RIX) to people around the world so that they themselves can own a piece of the next web. A web that’s truly owned and operated by the people, rather than a few tech giants who are only interested in the growth of their existence — rather than humanity itself. We gave the coin away for free because it was the right thing to do. Those who know me, know that I have spent my career giving to others. In fact, even from behind this fence, I have continued to help countless individuals when and where I was able to do so. I do it because it is the right thing to do. Some may be annoyed by it — but I refuse to be like those that I fight to defeat each day. Greed has never been an issue for me, nor will it ever. I have dedicated my life up until this point to fighting for people like you and I will continue to do so. That is what these projects are all about — pushing the human race forward and fighting for the freedom of others. I do not believe there is any fight that’s greater than that.
From the launch of dSocial, to dHost, there will be plenty of ways that people like you can earn a few dollars for simply using their computer. Imagine that! Enriching yourself, rather than Google, doing what you love to do each and every day! That was the dream at least, three years ago and today, well, you see the Legos on the ground. We are getting ready to deploy some monumental software. With companies like Peeps and dHost entering the dWeb fray, those days are upon us. I’m happy to say that I’m now developing for two companies who are both fighting, day in and day out for your online freedom. Being able to roll all this stuff out amidst the challenges I currently face, without a computer or Internet access, has allowed for me to find my purpose. Most would say that what I’m claiming is simply impossible and probably wouldn’t believe it, but there are people in the @peepsology Telegram room who have sat in those jail cells with me, and they’ve witnessed the thousands of pages of code that have been written from this heinous place.
With that said, for all of those who were worried about blockchain-based dApps that may have beat us to the punch, like other decentralized social networks or ride sharing applications that have recently launched, they simply did not see the bigger picture. By focusing on a foundation that can scale the world, I can say that our apps are able to operate at global scale and that those “other” apps will ultimately face the bottlenecks associated with the blockchains they are permanently built upon. While many were racing to build apps, we were racing to build a stage for globally scalable applications. That stage will be there as Peeps and dHost roll out the ALPHA launch of dSocial over the next few weeks, along with the dHost whitepaper and the new DWEB cryptocurrency, where we will finally showcase the power of the dWeb. That is, if we have not shown it already.
Let the rollouts begin!
Until next time,
Jared Rice Sr.