A second chance to review this week's AMA with CEO Joseph Denne and the senior engineering team for the DADI network.
Updates
Last Updated:
September 2019
DADI CDN pricing is c.60% less than like-for-like in AWS (a benchmark for DADI CDN prepared in hand with an existing customer). How is pretty simple: we have next to no capex and the costs involved for individuals running nodes are comparatively very low.
Faster machines will dequeue at the highest rate possible based on their network capacity. The network will also provision limits based on the perceived machine performance, to restrict low resource and low latency machines from taking on more requests then they can handle.
To begin with we’ll be using some comprehensive machine learning metrics – we expect to introduce A.I. as the network matures.
We are planning to run the network on more than one Raspberry Pi! 😬
A single Pi 3 can handle around 40 concurrent requests in our decentralized stack. That's a lot given its footprint and energy consumption. And of course, we've only used the Pi to illustrate platform efficiencies: most of the Node-level power in the network will be far greater than this.
Obviously, a single Pi isn't more powerful than a data centre (unless that DC only has a single Pi in it of course!), but a network of tens of thousands of connected devices in our architecture will provide far greater performance than traditional operations can achieve. It's better for the environment too.
Companies that host in their own DCs today do so because until now there hasn't really been an alternative.
We're not in a position to be able to answer this today as it's commercially sensitive information. Sorry!
Our strategy remains exactly as defined in our roadmap – and pleasingly we remain on track. In terms of PR, we continue to work with our contracted partners on a mix of articles and interviews for mainstream and crypto media – links will be shared as they go out.
Existing marketing has been maintained and will continue to be stepped up in coming months (we can't go into too much detail as it is commercially sensitive, but you will keep seeing the results). We also 'newsjacked' Microsoft's press last week regarding their submarine datacentre experiments.
DADI's dApps can be thought of as SaaS from the point of view of deployment. We're working to make the process as simple as possible, and this will continue as we get to the development of the marketplace, which will enable one-click deployment for a blog, an e-commence site, etc.
The deployed containers run in secure hardware enclaves which effectively ensures security when running trusted code on an untrusted machine.
They are working with our marketplace: an open container that is designed to run third party software. Specifically, we are working together to explore the potential for running their masternode setup within the DADI network. They are currently running in AWS. And in terms of benefits, competitive price points for them, platform use and revenue for us.
Staking requirements will be reviewed regularly and adjusted to meet capacity requirements. In time this will become an automatic review, handled at Stargate level. Capacity is driven by customer demand, which means more revenue, which means greater capacity can be added to the network.
Exposure. When people (businesses, agencies, individual devs) use our tech, they love it. The greatest channel then is getting it in front of enough people. The strategy here is pretty complex, involving marketing, PR, direct sales, partner programs, events and more besides those.
This is all about the sales approach. To a traditional business we are selling on a service by service basis - our go to market in the first instance is with DADI CDN. This is a traditional sell with three main sharp edges:
Talking about decentralization in this context is actually just a distraction. It's part of the narrative, but decentralization in itself is not a benefit, rather it enables the benefits. And we sell the benefits. So, any trust issues that may exist are mitigated by this approach. However, it's worth noting that none of our existing customers have voiced concern. On the contrary, they are excited by the potential of the technology and are talking to us about how they can leverage it.
The return on POS requirements in the Masternode system document were modelled with the token price at a specific point in time. Bearing in mind that pricing in the platform for services is benchmarked in USDT, when the token value drops (and we haven't adjusted the POS requirement), the return on investment increases accordingly. Right now for example, that figure for a DADI Host looks like 15-20%+ ... we'll provide a calculator for this soon.
Our Crowdsale was completed in January under the jurisdiction of the UK and the entity into which all funds were raised is a not-for-profit asset locked vehicle whose objects are to use funds to deliver on the DADI network plans. At the time of the Crowdsale the results of the Howey Test were made available on our Crowdsale site. At the time of the DADI Crowdsale the DADI testnet was operational with web services already in use by customers. There is currently no enterprise selling the DADI tokens: this was concluded as part of the Crowdsale.
The purpose of the DADI token is to buy and sell services through a decentralised network governed by a DAO. It is important to note that whilst the tokens were widely sold, to c8,000 purchasers, this represents a small proportion of the DADI community that we are actively looking to contribute to the DADI network (a key objective of the Crowdsale and the hence limit on tokens that could be purchased by an individual buyer). Furthermore, as stated within Crowdsale and other communications expected return for DADI token holders is from participation in the DADI network, whilst the listing of DADI on exchanges is an integral part of providing liquidity and ensuring the availability of tokens for buying and selling of services through the DADI network.
As detailed in our Tokenomics paper we want (and expect) holders of DADI to be active contributors to the network providing computational power. Our decentralised network is created by token holders providing a proof of stake and initially we will be rewarding contributors based on availability of capacity whist demand for the services develops. Our ambition is for all token holders to contribute to a decentralised network of tens of thousands of nodes. This decentralised network is a reality today and new products are being launched right now to generate a demand for the network with DADI.
To the very specific question raised for question 13 (paraphrased as “given the latest communication from the SEC (William Hinman), will DADI seek to be a regulated security or withdraw from the US market”):
Given DADI's purpose, structure and the current status of the DADI network, and specifically in light of current guidance, we do not see DADI as requiring regulatory approval nor do we anticipate it becoming a security in the future. As we roll out the DADI network, DADI and its advisers, will continue to review and cooperate with regulatory authorities as appropriate and necessary to ensure the acceptance and use of DADI tokens. We believe there is an important role for businesses to play in protecting the crypto community (incl. token purchasers) and addressing fraud - this is key for innovative technologies to be adopted and deliver on the enormous value potential they hold. We see the recent announcement by the SEC as positive and a progressive step in supporting the adoption of Blockchain and cryptocurrencies.
You can build as many Gateways as you like, as long as each is provisioned the minimum proof of stake. The network handles load-balancing and resilience, but for the individual Gateway, the higher stake represents a greater requirement for stability in the hardware.
We anticipate some downtime, and factor that in throughout the payout mechanics, so rather than aiming for 100% uptime, it’s best to aim for scheduled maintenance windows. We’ll be looking into introducing a maintenance channel to reduce penalties for scheduled downtime. We’ll be getting into more detail on how to handle downtime and maintenance soon.
You'll need to stake your tokens through running a node in order to generate revenue.
We're already pretty active in this space, but yes, efforts will be stepped up as we move forward, particularly as we take DADI CDN to market.
The price of the Token will not affect the returns as services are pegged to $USDT and benchmarked with competitors. Price compression in terms of the Services such as CDN or other Web Services will impact revenues, but the return will be managed through the proof of stake. Our model provides for a wide range of open source web services to be sold through the network, so there is the opportunity for margin protection based on a mix of services. Furthermore, a strong channel strategy will target customer partnerships where there is price resilience due to a long-term commitment to DADI, based on fundamentals such as lower total cost of ownership, community participation and greener internet initiatives.
2.5GHZ+ is the advised base frequency for a single core. This is down to basic throughput requirements, where regardless of the number of cores, the threaded processes of the Gateway application need a consistently fast clock speed. If the over-clocked boost satisfies the peer review benchmarking requirements, then there won’t be an issue. We’ll be providing some build examples and details on the profiler in due course.
We’re looking into running the Pi’s with full disk encryption ahead of the release of ARM/Broadcoms TrustZone. We’re closely following the development of secure enclaves on all of the major chipset manufacturers.
The big point of difference between DADI and AWS in relation to redundancy is that an AWS sysadmin must provision redundancy on a per-application basis. That’s a huge amount of wasted resource.
Take the example of a website that targets a UK audience. The hosting stack is likely to be almost entirely redundant, with always-on hardware sitting idle. The DADI Network appropriates idle hardware, across every connected device, thus creating a single pool of devices and more accessible scalable redundancy at a much lower financial footprint.
The idle payout mechanism encourages cloud contributors to commit resource even when it isn’t required for active services, another of many considered measures to maximise on spare computational capacity.
As an aside, during the crowdsale we were heavily targeted, and because of the restrictions placed on us by AWS, we were unable to scale to a comfortable number of instances.
We’re in discussion with them about this and should have an update in the next week or so.
DADI is developing a marketplace proposition — a containerised service for approved partners that enables the running of third party software within the DADI network.
Agorai are working with us to deploy their technology in to the marketplace. They are also likely to use DADI API and Store for key components in their setup.
We are heading into an era of massive data centre growth as demand is huge and more capacity is needed. There are real concerns over the green impact of data centres and renewable energy alone not the answer. DADI makes use of the massive unused capacity that exists in homes and businesses — a far greener proposition that anything that has come before it.
DADI’s decentralised Cloud offers the opportunity for increased performance due to dispersion and locality of nodes. DADI’s dapps enable the development of products and services far faster than what’s achievable in traditional platforms:
“Products built with DADI are delivered in days and weeks, not weeks and months.” — Matt Hobbs, CTO Bauer
Finally, there’s an opportunity to re-distribute value to communities increasingly reliant on the Cloud for services and experiences. Yes, there is a significant value proposition based on reductions in price, but overall, it’s less about a price war and more about wealth re-distribution.
More information on how we are working together, ahead of our joint appearance at AI Summit this week.