AMA recap: January 26, 2019

The first Ask Me Anything of 2019 was covered a lot of ground. From the inner workings of the Edge network to the recent FSA consultation, this was an excellent session with a set of brilliant questions from the DADI community.

Lead article image

🔗Question 1

🔗When can we expect to see more data like performance metrics, total uptime, estimated earnings in the dashboard regarding our nodes?

Joseph Denne: We’re working on fitting more into the dashboard as the Host telemetry application moves into beta. Our expectation is to have this sometime over this quarter, with updates added over time. It’s sensitive area of the network because every call brings load, and of course we want the data to be as up to date as possible.

James Lambie: We have an explorer in beta too, the first version of which we will be sharing once we’ve put a friendlier design layer on it. Still a lot of work to do here!

Edge network explorer, betaEdge network explorer, beta

Joseph Denne: Our aim is to cover the map entirely.

🔗Question 2

🔗Many crypto companies have been laying off their employees. Is DADI looking to downsize as well? If YES, will it affect on targets in the roadmap? If NO - how many years/months will the company have enough funds?

James Lambie: It’s unfortunate to see, but a lot of crypto companies were heavily reliant on a volatile market, with no other revenue to support an aggressive roadmap. At DADI we already have products used in production and we continue to support these which goes a long way to stabilising the network funding.

Joseph Denne: It’s important to state that we’re not blind to market conditions. We’ve taken steps to protect the business and will continue to do so as the broader picture evolves. One of our core focuses for 2019 is driving adoption, which forms part of this strategy.

Follow up: How long is your runway?

Joseph Denne: Rolling. As things stand, we have enough income to maintain the business indefinitely. Pace of delivery and the shape of the business is impacted by market conditions – there’s no threat to the business itself.

🔗Question 3

🔗What should we be expecting from DADI in 2019 now that the network has been launched? Apart from the roadmap, what other milestone achievements are planned for this year?

Joseph Denne: Apart from the roadmap?! As if that wasn’t enough 😂 Well, in answer to the question the big focus for the business is our go to market: driving sales and revenue for contributing nodes. (I should note that it has been a great start so far! Chris and I have attended 30 new business meetings between us in the last fortnight.) We’re also engaging with OEMs to explore methods of distributing nodes that help us reach our “home as a data center” vision. And of course, the engineering teams have a stacked roadmap! The big deliverable is the compute layer in the network: moving this to the top of our list to enable multiple apps and languages to be run. For me this is the strongest expression of the capability of the network architecture, hence the focus we’re giving it.

🔗Question 4

🔗How long will it take to failover to another host? What is the availability (estimated 9s) you can promise to customers, when you consider host fails?

James Lambie: There’s never a single host responsible for a single application in the network, so redundancy is built in bottom up. If a Host goes offline it drops out of the network instantly. The architecture of the network means that Hosts are requesting jobs directly, and in competition with each other. This means that outages that you see with traditional setups as a result of single-instance failures are a thing of the past in Edge network.

Joseph Denne: So, I guess from an application perspective we expect 100% uptime. Where this nets out to in terms of overall network performance, we don’t know yet, but we’re targeting 99.99%

Follow up: Am I right if I say the payout system is the weak link here? If the payout contracts fails the entire network is asleep…or what happens if the payout network/system gets congested? I mean DADI does many micropayment transactions right?

Joseph Denne: The payout system is disconnected to the network itself - it runs as a side chain. So, no impact on uptime from this side of things.

Follow up So, payout system down => all traffic is free?

Joseph Denne: No, the traffic will still be being tracked by the Stargates, but payouts would effectively be on pause until the contract was available again.

🔗Question 5

🔗Can we onboard more than 1 device when self-onboarding starts?

James Lambie: Depending on your network capacity, yes. If you flood your connection with devices, your earnings will be diminished. So, there will be a balance to be struck.

🔗Question 6

🔗When does testing end and earning start?

Joseph Denne: As soon as we’re happy with our production test results, we’ll engage the POW engine. I hope that we’ll be distributing earnings from the end of February, but realistically this may happen in March. We’re still officially in production testing, so while we’re tracking POW, we’re not applying earnings. This will start e/o Feb.

🔗Question 7

🔗How many clients have been onboarded via CDN?

Joseph Denne: A handful. As we move forward and bed down capacity and stability, we will start to talk about individual success stories (including on our site). You can expect to see more of this towards the end of this quarter.

🔗Question 8

🔗Will DADI provide a framework for contributors to estimate/file taxes on earnings from providing services or would that be the responsibility of the fiat exchange provider?

Joseph Denne: Taxes are calculated based on the gain made (capital gains tax), which typically means once crystallised to fiat. In the case of POW, in the UK there is income tax due on the income from the network, which would be calculated again at the point of withdrawal to fiat. In this sense the blockchain itself provides you with all you need to calculate the gain. I like the idea of building in a guide for taxes. We’ll discuss what we can do regarding education/support with regard to taxes internally and see what we can reasonably do. (Bearing in mind that the network is already active in 96 of countries!)

🔗Question 9

🔗Can we get SSH access to the own founding node? I want to include it in my local monitoring

James Lambie: Nope, they are locked down and self-updating. You’ll be able to use the network explorer to interrogate them though!

🔗Question 10

🔗Can the team speak to the profitability of running a node? Seems to be one or two people that are extremely concerned about this.

Joseph Denne: Yes, a little. We expect a Host to be returning c25% of its POS annualised, assuming it is running at capacity; a Gateway to return c45% of its POS annualised, again assuming that it is running at capacity; and a Stargate to return c80% of its POS annualised, running at capacity. I have a whole bunch of information to share on this front (pricing, masternode system updates etc.), which we’re in the process of writing up and will be releasing soon.

Follow up: So you expect 80% PoS annualized, denominated in USD; not $DADI correct? (paid out in $DADI ofc)

Joseph Denne: In effect, yes.

🔗Question 11

🔗How can I limit the bandwidth my founding node is allowed to use in my network?

Arthur Mingard: The Founding nodes should only use a small amount of bandwidth. When the desktop application arrives you’ll be able to configure this more, but for now I wouldn’t worry. Some routers do come with bandwidth management so that’s another option to consider.

🔗Question 12

🔗Why is it that some founding nodes appear offline whereas others appear online? Can you elaborate on the ‘testing’ phase, what is this about?

Arthur Mingard: We’re improving the way we evaluate the state of a Founding Node with increased health checks. We’re rolling the updates out to some systems, and those without the latest version will temporarily not meet the required health checks to show a fully online state.

🔗Question 13

🔗Have all the Founding Nodes been onboarded or are we waiting for more to go online?

James Lambie: The majority of those who funded their wallets are online. We have a small number of nodes left over which we intend to distribute to the community in the coming weeks. Look out for an announcement about this.

🔗Question 14

🔗What happened to the market maker? Do they still own $DADI?

Joseph Denne: Yes. Still engaged and still operating.

Follow up: As was recently clarified you are required a minimum trading volume to be in Kucoin+ ranking. DADI volumes have substantially diminished in the last weeks. I would like you to elaborate further on their involvement. As they are clearly still present, but not in the same capacity. Which is why you got kicked down on that exchange. What is the value gained from them?

Joseph Denne: A fair question, and one that has been asked internally. They follow and predict market movements, and operate independently of us to support volume when it is required. I can’t say much more than that at the moment.

🔗Question 15

🔗Now that Constellation has been released, what is the process for having my website hosted by DADI?

Joseph Denne We can work with you get to you online with CDN directly - reach out at hello@dadi.cloud or hit one of us up directly. We’ll be opening up the interfaces for individuals to directly interact with the network to setup and use CDN soon. You can join the waitlist for this by setting up an account on https://my.dadi.cloud (if you don’t already have one) and hitting the “Join the waitlist” button and giving us your domain. If you want to use the technology/network outside of CDN, we can assist with that as well, via a VPC build of the Edge. Again, reach out!

🔗Question 16

🔗Will profitability be greater than the cost of maintaining a Host?

James Lambie: We believe so. The network is designed to use spare capacity – so for example running on your laptop. It also works in low powered and embedded devices. This is precisely why we chose the Pi for the Founding Node. It costs around £3.40 a year to run one 24/7 in the UK at the moment, and electricity is expensive here.

Arthur Mingard: These are calculated using a combination of dual cost (daytime £0.32/kWh, nighttime £0.26/kWh) rates on an average UK tariff (we pay a lot for our utilities) when demanding a sustained 400-500 mA, with LEDs enabled and with a wired connection. It’s likely to be lower, very unlikely to be any higher.

🔗Question 17

🔗My founding node transfers between 100 - 600 MB data daily. Since we are not getting any real traffic, what’s it doing at the moment?

Arthur Mingard: We regularly run tests on the network load to random hosts which would explain the traffic. You shouldn’t see large spikes but a slow trickle when tests are active.

🔗Question 18

🔗Will we be able to buy/order a premade plug and play node for staking?

Joseph Denne: Simple. Yes. 🕵️

🔗Question 19

🔗What do you think about the consultation paper from the FSA? Does more regulation pose a risk to DADI at all?

Joseph Denne: I read the paper from the FSA - the Financial Services Authority here in the UK - earlier this week. It’s a very measured document which is clearly seeking to support the cryptocurrency industry. As $DADI is a utility token that sits outside of the e-money framework in the UK, the document doesn’t have any implications for us as a business. But it will help bring clarity over operations in the broader space in the UK. We have a lot to feedback to them on, which we will be doing over the coming month. But of course, we welcome greater clarity regarding the regulatory environments that we work in.

🔗Question 20

🔗A suggestion has been made to change the monthly roadmap links to simply Q1, Q2, Q3 and Q4. Apparently, when hitting the month link some have interpreted those milestones as being specific for that month.

Joseph Denne: Good feedback! We’re in the process of updating the site to make the network proposition much clearer as part of our go to market strategy. This should be released next month. We’ll make some changes to the roadmap pages as part of this work.

🔗Question 21

🔗Why is it that some founding nodes appear offline whereas others appear online? Can you elaborate on the ‘testing’ phase, what is this about?

Arthur Mingard: We’re improving the way we evaluate the state of a Founding Node with increased healthchecks. We’re rolling the updates out to some systems, and those without the latest version will temporarily not meet the required healthchecks to show a fully online state. We’re testing production traffic in the network - seeing how it performs under load, what latency looks like as audience profiles shift and the like.

Follow up: Am I right in thinking you poll once an hour? (seems like that looking at the dash)

Arthur Mingard: Polling is around once every five seconds via a GRPC health check. It’s more like a persistent check than polling, but that’s about the average time for the entire network to successfully propagate health data.

Follow up: If I recall correctly the actual networking software running on hosts, gateways and stargates is being patented. Is that still the case? Do you intend to opensource it after that is done?

Joseph Denne: We’re still looking at that, yes. And yes, the intention is to move a full Open Source footing in time.

🔗Question 22

🔗Could you provide a status/update on Contract Logic per the Whitepaper, i.e. node interaction and payouts?

Joseph Denne: We’ll be publishing information on this in the next month, along with detailed pricing for the network.

🔗Question 23

🔗Will DADI be listed on masternodes.online once the nodes are up and running?

James Lambie: Yes! We plan to be listed on multiple aggregator sites, and to be available via third party hosted services soon after we move to a wide availability footing for self-build nodes.

🔗Question 24

🔗If an application requires high processing power, how is it processed in this distributed environment (collection of Pis)? In that sense, how does DADI challenge centralized cloud providers?

Joseph Denne: First up it’s important to note that the network is much more than “a collection of Pis”. A Pi is simply the base compute node size in the Edge network. And in that it is roughly comparable to a t2.micro instance in AWS (actually it’s a fair bit more powerful). The Edge network allows for different compute unit sizes to be used and targeted, meaning that in time you will be able to pick the appropriately sized resources for your applications. Of course, as the Edge is a fundamentally different architecture, it needs to be thought about differently in terms of application build: working within the limits of individual compute nodes; making use of our APIs for storage input/output; and monitoring performance etc.

🔗Question 25

🔗How do you effectively achieve east-west traffic demand on the network topology?

James Lambie: When you deploy, you deploy to the Edge network rather than to a single instance or cluster of instances (as you tend to do in a traditional cloud setting). The network itself then determines where to run the image based on emerging and predicted audience demand, effectively swarming your app to meet demand. We’ve established the backbone in the network - along with routing between nodes - to smooth out the relatively high server-to-server (east-west) traffic that this architecture requires (which it should be noted is relatively small).

🔗Question 26

🔗Any chance of you guys renting out Stargates in the backbone to the community?

Joseph Denne: Yes. To elaborate – we’re working on modelling full and partial rentals for various backbone components. We like the concept as it helps to move us towards full decentralisation, which is the ultimate goal for the Edge network.

🔗Question 27

🔗What’s the difference between the hosts that are already gaining traffic and the founding nodes, which need further testing?

Joseph Denne: We’re production testing the network at the moment. This means that live traffic is being served by it - the backbone, Founding Nodes and nodes with partners. There’s no distinction in test traffic for the mainnet.

🔗Question 28

🔗How have you progressed on providing a mechanism for clients wishing to purchase your services (which buys the underlying $DADI)?

James Lambie: We have a beta ready to be tested in a live context, which we anticipate being able to test in the coming few weeks.

🔗Question 29

🔗Can we get an update of the Agorai collaboration?

Joseph Denne: Work within our compute container is ongoing. I can’t be any more specific than that at this point I’m afraid.

🔗Question 30

🔗Is Gateway onboarding still planned for Q3?

James Lambie: Yes

Related articles

More knowledge articles
Mail icon

Like what you are reading? Want to earn tokens by becoming a DADI Node? Save money on cloud computing services? Build amazing digital product with DADI Web Services? Join our mailing list.

To hear about our news, events and products or services subscribe now. You can also indicate which services you are interested in, which we use for research and to inform the content that we send generally.

* You can unsubscribe at any time by emailing us at data@dadi.cloud or by clicking on the unsubscribe link which can be found in our emails to you. Read our Privacy Policy.