Edge Latest articles from the team en 2019 Edge Network Technologies Ltd. All rights reserved. Sun, 25 Aug 2019 00:55:00 +0000 2019-08-25T00:55:00+00:00 en 2019 Edge Network Technologies Ltd. All rights reserved. Edge https://edge.network/assets/img/dadi-logo-colour.png https://edge.network Weekly update: w/c 12th August, 2019 https://edge.network/updates/announcements/weekly-update-wc-12th-august-2019 Hello everyone 👋

I’ve been on holiday this week, so this will be a fairly quick update.

Of course while I’m away I like nothing more than to work! So I Spent a bit of time updating foundthings.com, which is now running in full on the network (DNS + Content distribution).

Setup was easy and it’s pretty performant 🙂

As I mentioned last week, we’ve now added Edge-Device alongside Edge-Host in the headers returned by the network.

DEVICE_ID is the ID given to the device by Agent, the application that manages authentication and software updates. When Agent registers the device with Consul, it registers using this ID to create a Device service.

HOST_ID is the Consul ID for the host application. When Agent authenticates on the network it downloads Host and Host connects to Consul using the HOST_ID.

Why don’t we have a single ID? Because each Consul service need a unique ID. It’s important to know that if Host goes offline, the Agent application is still running.

For Telemetry it’s also important to be able to distinguish between the resources used by Device Agent, and by Host.

I see the following network headers for foundthings.com:

Edge-Device => 9e82bf98-d6f7-4559-b91c-95c578eb74a2

Edge-Gateway => 455106e7-612e-411b-9ded-38bfce8f59ee

Edge-Host => 82a3ce68-d489-4fab-afed-7046e3988752

+We also return some cache control headers:

Etag => "11ec4-wBAM3fVvut8jqaPsB2ab15I2Zng"

X-Cache => HIT

Aside from this update to the network, the network team have been working on improving the performance of the Gateway HTTP request queue, and have also been building out the transfer queue for Edge Storage. The response time for requests has been reduced by around 40% on the back of this work.

That’s from very fast, to extremely fast folks.

To achieve the performance improvement we eliminated one of the RPC (Remote Procedure Call) methods.

Up until now there were three RPC calls performed by Host:

  • Check, opens an http/2 stream and waits until there is a request
  • Read triggers when Check closes and fetches the request
  • Write returns the response

After some benchmarking and a lot of experimentation, Check was dropped and Read replaced it as the long polling stream.

This means that rather than having to make a call to Check, then Read and then Write, Host now only calls Read and Write. Which means far fewer requests in production.

Another optimization was made to the Protobuf which are the blueprints for RPC interactions. They tell the Gateway and the Host how to convert byte data into meaningful objects, rather than the traditional approach of including keys and values in the payload.

To reduce the payload size, we attributed shorthand values to some of the stricter keys, for example:

  1. Previous version of method transaction
message Request {
  string method = 1;
  bytes body = 2;
}
  1. New version using enum for keys:
message Request {
  enum Method {
    GET = 0;
    PUT = 1;
    POST = 2;
    DELETE = 3;
  }
  Method method = 1;
  bytes body = 2;
}

These small changes make a huge difference, and there are plenty more to come!

The team have also been cracking through earning/payout routines, which we expect to be releasing at the end of the month. This means we’ll be moving from monthly to much more regular payouts very soon.

And work on CLI for self onboarding has been ongoing.

We released CDN 3.6.1 with a fix for a bug raised by the Edge team. You can see this here: https://github.com/dadi/cdn/releases/tag/v3.6.1

(Note that we’ve not migrated the web services to our new Git org yet. This will happen in the coming weeks.)

And we added new functionality to the API validator module in preparation for some changes to API authentication. The new functionality will also be used in our SaaS platform.

The team also made a start and significant progress on a new authentication method for API, called access keys. This will solve the requirement from the Edge network to have bearer tokens that don’t expire, and will also allow us to persist Publish sessions for as long as we need. It will also be a cool way for people to integrate third-party apps with their Edit installations, with the ability to view, edit and revoke any issued keys.

They also made some structural changes to Publish routing, in preparation for the SaaS product of what were our web services. This adds the property name to the URL if more than one property is defined. it also stops reference selection from pushing a new URL to the history, as discussed previously, which solves a bunch of smaller issues.

We have some new business through an existing customer, which we expect to be kicking off in the next few weeks.

And we spent a bunch of time with the PPA, working up future workstreams and assisting them with their 2020 roadmaps.

Tiipr is also now ready to submit to the Apple store, a little later than planned on the back of some last minute changes, but it’s great to have it so close to the line. I’ll let you know when it moves to live.

And that’s about it for this week!

Enjoy your weekends.

]]>
Mon, 19 Aug 2019 13:30:00 +0000 5d5a5466a1b958570459cd3a Joseph Denne 2019-08-19T13:30:00+00:00
Weekly update: w/c 05th August, 2019 https://edge.network/updates/announcements/weekly-update-wc-05th-august-2019 Hello everyone 👋

It’s been a jam packed week at Edge, even with the summer recess in full swing. There’s a lot to update you on this week.

First up, https://thedecentralizedweb.com/ has been updated with Edge branding.

As it’s running on the updated mainnet you can now see both the Gateway and the Host ID of the nodes that handled the delivery in the header request.

For me this looks like this:

Edge-Gateway => 370d51a7-4a48-4b6c-b52f-26d813e2f086
Edge-Host => caf4e202-350b-4d28-ad0e-ec57c5e2fe85

We will be adding the device ID to this in the near future, and further down the line the Stargate ID as well.

The Stargate ID requires the internal secure DNS consensus check to be fully operational. We’ll be writing more about this in due course, but it’s a very powerful mechanism that super charges security in the network by adding the consensus mechanism to all DNS requests.

You’re going to love it.

A single document collection feature was added to Publish, giving support for individual page control in the interface.

The team made a start on document versioning interfaces within Publish. Versioning has been part of the platform for years, but until now there hasn’t be a front end to it.

They also squashed a few bugs, fixing an issue with Reference selection in Publish and fixing an issue with document versioning + filestore in API.

User management interfaces were added to the SaaS proposition for what were our web services, and specifications were created for 1. API key management, 2. the integration of network content distribution image cropping, and 3. the integration of network content distribution with API.

Workshops with the PPA are shaping up nicely, with the first in plan for September.

We’ve established a private sector team which is gearing up for direct engagement in the UK.

Two pieces of new business have arrived through one of our advisors, with meetings to follow in early September.

An exciting partnership discussion opened up this week. Can’t talk about more than that for now though I’m afraid.

Branding and interfaces for the SaaS proposition for what were our web services are progressing nicely. We expect to be releasing this in September.

We’re continuing to work through updates to Founding Nodes, and expect to have the remaining nodes updated and back online by early September at the latest.

The BGP layer in the network had an update focused on deployment processes, which are now much improved (great instead of just very good 😂)

They have also been working on the file sharding queue, which is used to transfer file shards between Gateway and Host.

Our DNS service has been being updated to further harden the solution, adding per-domain NS server records. And the front end interfaces for Edge DNS are almost complete. I’ll be sharing designs for this with you all in the next week or so. And we expect the service to be live in September.

This is major milestone for us, and will be the first Edge service to be available self-serve. We’ve been using it for months, both internally and with existing customers, and it has proven to be very robust. Nonetheless the first release will be flagged as a beta, as scaling it out will provide us with an invaluable feedback loop for service improvement.

We picked up the domain ed.ge, which is pretty neat. We plan on using it for a series of free services all TBC in due course.

The team have been experimenting with a custom Merkel DAG as part of an update to the underlying storage solution in the network. Very early days, but it’s exciting technology that has a lot of potential for evolving the core network services in future releases. We’ll be writing about this in detail as we move forward.

And we remain on track for the release of self-onboarding at the end of August, with the first new devices expected to be bought on at the start of September.

And that’s it!

Enjoy your weekends 🙂

]]>
Mon, 12 Aug 2019 13:30:00 +0000 5d5144a8981f1e56d6522ac8 Joseph Denne 2019-08-12T13:30:00+00:00
Weekly update: w/c 29th July, 2019 https://edge.network/updates/announcements/weekly-update-wc-29th-july-2019 I’m literally just off of a nine hour flight, so this will be a relatively short update!

The network team continued their work on the network CLI, which is being written in Go, and which is used to authenticate and onboard devices into the network.

They adapted the filesharding package for object storage, introducing HTTP payload support and in-memory caching.

They also refactored the Gateway queue to store file shards as well as HTTP requests, and improved overall performance of the core interaction RPC methods used by Gateway and Host.

The team added new features to the rich editor in Publish: undo/redo icons, plus a word/character count.

Small, but powerful tools to enhance the editorial interfaces.

And they made a start on a UI library for what were our web services, which will allow the reuse of visual elements across our various properties.

They improved the API endpoints for managing collections and their fields/settings and progressed the SaaS proposition for web services so that it’s now possible to create and edit collections, as well as managing fields directly within the API service.

As part of this work-stream they released a new version of the API validator module (@dadi/api-validator), which now includes methods for validating collection schemas.

And they added more information to the document versions endpoint, namely the date and the author of the change, in preparation for the integration with Publish which should begin next week.

An updated image for network telemetry on Founding Nodes was finished. This update sorts telemetry connectivity in to the upgraded mainnet and will rectify the “Unable to get status” message.

The update will be pushed out to node owners over the next few weeks.

There are two methods of delivery: direct or via a new SD. A form will be provided to guide Founding Nodes owners through the process shortly.

The Ecohustler updates mentioned last week went live on Monday and has already worked to significantly increase traffic to the site.

And finally Tiipr should be moving to the App Store as a beta next week.

And that’s it!

Enjoy your weekends.

]]>
Mon, 05 Aug 2019 12:30:00 +0000 5d468da039d1dd083a574548 Joseph Denne 2019-08-05T12:30:00+00:00
Mainnet upgrade https://edge.network/updates/network/mainnet-upgrade Scale

Important in the networks growth is its ability to accept an increasing number of Hosts without compromising the speed in which vital coordination components calculate and synchronise topology.

Initial builds of the network manifest distributor operated on geographically spread Consul servers, adjacent to the Stargate services. Whilst decentralised in respect to their lack of a single point of failure, these nodes required consensus on all operations, including healthchecks, key/value updates and service metadata. Put simply, if a Host were to tell its nearest Consul server that it was up and healthy, the same data would need to be fully propagated before the write was considered successful. Operationally, this latency doesn’t scale well for our type of network, and we started to see some failing health checks (even whilst telemetry painted a far more serene landscape).

One of the largest single changes in the release was a shift away from globally propagated service data in favour of a multi datacenter approach, more akin to your typical VPC. The services continue to register with their local Consul server, operating alongside the nearest Stargate, but that’s where the syncronisation stops. Consul offers a little flexibility in the way it connects to its peers, so whilst a multi datacenter setup does not share service data globally it does permit access by proxy – something the network relies upon on rare but vital occasions, such as global load balancing during peak traffic spikes.

Performance

When debugging the latency experienced by the single data center configuration the team discovered a number of performance improvements to the way service data is stored. Whilst the refactors were complex, the result can be simplified to two key areas:

Health

The first health checks consisted of periodic writes to a key/value directory on a per-device basis. A pruning process existed to remove services that exceeded the maximum TTL.

The latest iteration of the health check process uses Consuls inbuilt healthcheck methods, including GRPC standard health endpoints for Stargates and Gateways, and TTL checks for Hosts. This meant that we were able to completely remove the key/value data which had a significant reduction in latency due to the way the Consul propagates service health.

Metadata

Until recently all devices wrote information such as their current connections and build digest to Consul key/value. Whilst it appeared to be largely an efficient method, benchmarking showed us that storing this data within the core Service metadata reduced the number of Consul requests by other services.

If, for example, a Host wanted to find the closest operational Gateway it would first need to make a call for all services of type ‘Gateway’, individually requesting the KV metadata of each Gateway before an informed preference could be made. By moving this data to the Service itself, what could be a query of 1+n became simply a query of 1.

Security

ACL tokens were initially created to offer distinguishably different permissions to the three core services, with the Host application having the lowest level of trust, and the lowest staking. The Host token had read-only access to non-sensitive information, and a small amount of write access to its own part of the network manifest making it almost impossible to use it for evil.

A single ACL token for all Hosts does come with a downside, and one we discovered when migrating to multi DC on testnet. Firstly, Consul doesn’t allow ACL migration which poses a risk when embedding a single token into the Host. It also means ACL for a single machine cannot be revoked.

The next release will include a refactored method for retrieving ACL tokens to use with Consul, which required the introduction of Vault, an accompanying service which extends Consul ACL with policy templates.

Founding Node status

As part of the network upgrade we updated Founding Nodes to the latest build of Host.

The new telemetry service in the platform also requires an update, but as a result of the permissions system in the previous build of Host, this cannot be performed automatically. This means that nodes will be reporting as “Status unavailable” within your dashboards.

Note that earnings and payouts for July are not effected, and where telemetry data is unavailable, usage approximations from Gateways will be applied.

What do you need to do?

We will need to either update your node’s telemetry system directly or send you an updated SD card to replace the one in your node.

Updating the node directly is the fastest route, but it does require a port forward to be setup on your router.

Please indicate your preference by completing the following form: https://forms.gle/Szm9jp3GSU4pVD6V8

]]>
Fri, 02 Aug 2019 17:30:00 +0000 5d447eb207b4db085ee3af72 Arthur Mingard 2019-08-02T17:30:00+00:00
Weekly update: w/c 22nd July, 2019 https://edge.network/updates/announcements/weekly-update-wc-22nd-july-2019 First up, the PPA site launched on Tuesday 💥

You can see it here: https://www.ppa.co.uk/

It’s a great showcase for our platform, demonstrating how quickly it is possible to move with the technology, as well as how easy it makes re-platforming (this time from Sitecore).

The launch has been widely publicised in the UK to the publishing industry here.

There are a bunch of changes to the Ecohustler site, which will be going live next week. These bring the site up to date with the latest releases of our tech and inline with the recent changes to mainnet content distribution.

I mentioned in a previous update that we would be moving one of our existing customers with global offices to the network. This work is now underway.

The engineer team have built out a web app and accompanying API capable of creating deployments of what were our web services and talking to their corresponding API instances.

They’ve also fixed some bugs in Publish, mostly relating to the rich editor and the date picker. These will be released along with the React version of our interfaces next week 🎉

Work on improving the performance of the request queue in Gateways is ongoing, specifically in relation to object storage and how it supports the process of file sharding.

And they are continue to work on the registration app for nodes, as well as building out the end points required for network CLI in support of self onboarding.

We have been interviewing engineers to join the networking team. There are a lot of great candidates! If you’ve applied and haven’t heard from us, just hang tight. We’re working through the applications.

There have been a handful of new business meetings, with particularly interesting opportunities in education and marketing.

And finally we have made a whole world of updates to our new site on edge.network, largely on the back of community feedback. For which thank you 🙏

The feedback loop with you all is invaluable to us, and really helps to drive forward what were building.
We have a few remaining updates to make to our social channels and are waiting on exchanges to update our ticker (which we expect to see played through over the next one-two weeks).

Oh, and we have a new intro film in production. Expect to see this launched and added to the site soon.

Enjoy your weekends!

]]>
Mon, 29 Jul 2019 21:30:00 +0000 5d40848d82e98f0808995b22 Joseph Denne 2019-07-29T21:30:00+00:00
Weekly update: w/c 15th July, 2019 https://edge.network/updates/announcements/weekly-update-wc-15th-july-2019 This week has been pretty bumper, and totally full on!

First up, https://edge.network is now live 💥

(Thanks for the cached social card Telegram!)

The new site focuses on network level services, and provides us with a solid base for the future. Over the coming weeks there will a whole bunch of updates and additions to the site.

The roadmap has been updated on site to reflect our most recent iteration. It will added to as we bring more of our R&D efforts to the fore in the coming months.

https://edge.network/roadmap

my.edge.network and the explorer are being updated as I type…

Ticker updates ($DADI → $EDGE) will be happening over the coming days, as will changes to our social channels.

This is a process rather than a button push, because there many parties involved.

We’ve migrated our codebase to the Edge github organisation: https://github.com/edge

The network engineering team have been cracking through updates on the back of the update to mainnet last week.

They have been working on a re-registration script to handle migrating Founding Nodes onto the latest software inline with the update to mainnet.

They’ve also improved DNS Exchange performance, and have fixed a bug with SSL certificate generation.

Build services have been migrated and deployed the replicate services for Consul.

Plans have been finalised for the self onboarding CLI, which will be released to a closed group at the end of August.

Our second live service, Edge DNS, is now live in the mainnet and in production. It’s running edge.network, amongst other domains, with a whole bunch in plan for migration later this month.

You can read more about the service here: https://edge.network/en/dns

The team also released a beta version of API, which includes the schemas-in-database piece, some fixes around search and, most importantly, a new URL format for collections.

And they released beta versions of @dadi/api-wrapper and @dadi/api-testbed to reflect the new URL format.

Publish has been updated to reflect the new URL format as well.

Updates were made to our in-house continuous integration service (used by Publish), so that it’s possible to specify the API structure as well as the API version that accompanies the Publish branch being built.

The build of the supporting site for the new stand alone web services proposition was kicked off in preparation for branding and content, and the process of migrating documentation pages was started.

The roadmap for the new proposition was finished.

And the team made a start on the control API that will orchastrate the SaaS platform, and handling deployments in to the Edge network.

Our support team continued efforts to migrate one of our key customer’s API services in to the network.

They have also been making some last minute changes for the PPA, which is now locked in to launch next week.

Final testing have been progressing well for Tiipr. Their MVP launch is on track for later this month.

We’ve had a bunch of articles about the rebrand in the press this week, along with an opinion piece I penned:

Finally our charitable foundation was nominated for a Noonie! If you haven’t already, please take a momnt to vote them here: https://noonies.hackernoon.com/award/cjxvrv4p26gd40b40cdlfmwwy

Oh! And check out what our team do for fun when not building out the Edge: https://twitter.com/imdsm/status/1152245649452228608

Awesome stuff.

And that’s it from me! I think you’ll agree that it’s been a great week 😀

Enjoy your weekends.

]]>
Mon, 22 Jul 2019 21:30:00 +0000 5d40e49533bf6f0840dc344a Joseph Denne 2019-07-22T21:30:00+00:00
AMA recap: July 12th, 2019 https://edge.network/knowledge/network/ama-recap-july-12th-2019 Question 1

What has been the most challenging thing to create while setting up this new brand?

Ravi Chandwani: There have been a number of challenges in replacing DADI with the new Edge brand. A key area of focus has been bringing the different App/Services under a single brand style. The existing offering felt somewhat disparate as there were a number of different styles and approaches being used. We have been and continue to work with the team to create more uniformity across the brand. Ensuring there is a stylistic and aesthetic connection in all aspects of the brand. Finalizing the logo itself has also been challenging. Ensuring the logo resonates with the target audience and sits comfortably alongside competitors whilst also having standout and impact. It is important that it is both distinctive and aesthetically pleasing. Something I feel we have achieved.

Joseph Denne: I’d say that you’ve definitely achieved! Super happy with the work

Follow up: Hi Ravi. What did you do to backup this feeling to make sure it’s not just your feeling and effectively something affecting the brand image? How much third-party input was taken for the rebrand? Focus groups, interviews, other designers’ input?

Ravi Chandwani: We’ve spent a few months on the design itself and moved - sometimes frustratingly! - through a wide number of variations and iterations. This included a significant amount of third-party input, including from customers and others within the crypto industry.

Question 2

How long has you been working on the rebrand to Edge?

Joseph Denne: Since October last year! So, we’ve been moving towards it for quite a while - starting with the naming of the network, and then through the reduction in use of DADI in comms, finally getting to the point of being able to announce it, and then of course formally launching it next week.

Question 3

In your opinion, what makes the brand identity that your team has created unique?

Ravi Chandwani: It is difficult to pinpoint any brand out there that is completely unique. Generally, there will be a connection between brands in a similar space which is what ensures they can communicate effectively in the wider industry. However, the intention with the new brand was to create something which differentiates us from our competitors and demonstrates a fresh approach to this space.

Key points of the brand are a focus on clarity and simplicity. We are communicating some complex ideas in a clear and concise manner. This is done through copy, but also through the overall aesthetic. Reduced colour palette, white space and a focus on minimalism. Combined, these qualities should give us stand out from the crowd.

Question 4

Is there any specific message or emotion you wanted expressed when designing this new brand?

Ravi Chandwani: The message we want to convey is clarity and simplicity, ease of use and accessibility to a complex and cutting-edge technology. This is achieved through refining and minimizing the visual and textual content. Lots of white space, giving content room to breathe as well as lighter font weights and a reductive palette should help to communicate these ideas. The brand will continue to develop as more elements are added to the visual language. We are taking an iterative approach so we can be flexible enough to accommodate new developments whilst ensuring that everything retains brand recognition.

Question 5

Will this new style also be used for the products you make together with Blond and how will this collaboration be in the future?

Ravi Chandwani: The new brand will definitely feed into the new products developed with Blond. This will be through the product design as well as packaging. This has been part of the thought process in developing the new brand look and feel. We are excited to see how this develops into 3 dimensional objects in the future.

Question 6

Can you tell us about the reason for the rebrand?

Joseph Denne: DADI was right for us at the beginning of our journey: It communicated the ambition of the technology, and it was a memorable name. However, when addressing a market as part of the proposition as now stands it lacked clarity. It’s also worth acknowledging that it wasn’t universally liked too.

Question 7

How do you comment the modified picture showing $EDGE being #7 on CoinMarketCap with your investors down 90%?

Joseph Denne: Well, #8! Contributors purchased in Eth. $DADI is down ~62% in Eth since the token sale. Eth remains down 76.46% over the same period. I’m not trying to paint a rosy picture, but it’s worth calling out the facts. In a very difficult market, our token price has fared reasonably well. The illustration is just that. It’s playful and reflective of our ambition, nothing more.

Question 8

What does a clean start with the new brand mean for the team and the future?

Joseph Denne: Well, we see Edge as central to our go to market strategy. It’s all about marking the platform out as the leader in the field of edge computing and edge networking. It’s a more grown up brand, and one which clearly communicates the proposition. We believe that this clarity of purpose will help to carry us forward, both internally and with the broader market.

Question 9

Do you think it will ever be super intuitive to create a website on DADI for those who are not code savvy? Like a Wordpress website, for example.

Arthur Mingard: There’s a big difference between wordpress.com and wordpress.org. And between either of those and Wix. Edge is all about network level services. This will eventually include a 1-click deploy marketplace for different technologies, which will lower the barrier to entry significantly.

What we’re doing with the web services under a new brand is much more closely aligned to a Wordpress, in that it is a SaaS proposition designed to lower the barrier to entry. In time we expect this to evolve further, with versions of the offering more closely aligned to the likes of wordpress.com.

Joseph Denne: That’s a yes. But it’s not our core focus in Edge right now.

Question 10

How does the DADI rebrand impact the relationship with the foundation?

Joseph Denne: It doesn’t. The DADI Foundation is also rebranding, and will be known as the Edge Initiative. The go live date for this isn’t confirmed yet, but we expect it to follow on quickly from the launch of Edge.

Question 11

Will DADI support the QUALCOMM Snapdragon Secure Processing Unit feature?

Arthur Mingard: This solution is similar to Intel SGX - we really support the introduction of this sort of feature to all chips on the market. We’ll be watching each version into maturity and adoption, and if they stand up to scrutiny, they’ll almost certainly find their way onto the roadmap.

Question 12

Do DADI customers know that they are migrating services to decentralised services/network? Is this something they like, does DADI advertise it proudly or is it only the cost-saving aspect that customers like? Could you tell a bit what makes customers move and how you gain new customers?

Joseph Denne: Yes, this isn’t something that we’re doing by stealth, although we are doing it slowly as we continue to test and improve the mainnet. And yes, we’re very proud of the technology. The strongest selling points are performance (speed of deployment, speed of delivery), security, transparency and price.

Question 13

Why is it that the response headers of thedecentralisedweb.com seem to return a DADI-host only the first time I load the page then after a refresh those response headers or no longer there?

Arthur Mingard: In CDN, once an object has been generated and moved in to the cache of a Gateway, the original node isn’t participating in the request, which is why the header changes. We are working on updates that will include a breakdown of all of the nodes that participated in a request, regardless of cache state.

Question 14

Have you considered onboarding users from other Crypto projects who need to run nodes, like Tierion?

Joseph Denne: We have been working with a handful of partners on this. So yes! I’m not familiar with Tierion, but will take a look.

Question 15

When can we use Edge like we use AWS. Obviously not all the features, but as a developer, if your offering is CDN, when can I just sign up and get going without any hand holding?

Joseph Denne: You’re correct in identifying this as a key milestone in the go to market. We are working the interfaces for the first services now, and expect to be moving them to live in Q4.

Question 16

It has now been over a year and a half since $DADI has been listed on a reputable exchange. Clearly, not enough has been done on this front since projects which have only launched in Q3/Q4 2018 have already been listed on the likes of Binance or Binance DEX (REN, Loki, Grin to name a few). When can we expect some positive developments on this front? It would be great for brand exposure and legitimacy.

Joseph Denne: We agree that broader reach and liquidity would be beneficial. And in fact, we see it as being essential as we build out the network. This isn’t something that has been ignored, but it’s also unfortunately not something that we can talk about at the moment.

Question 17

If I want to use DADI as my CDN provider, have you implemented some kind of DRM so that only authorized users can see the content and restrict authorized users from uploading content to Facebook or some other platform?

Arthur Mingard: User account management is something that typically exists within the application that makes use of a CDN. There are however plans for single use cryptographic keys within the solution.

Joseph Denne: So, for example, you could choose to lock down certain content and make it available only to authenticated users within your app. Those users would be given links (or the content would be embedded with) URLs which would expire after a set amount of time has passed.

Question 18

It is understandable not wanting to name certain clients using/interested in the network. However, I find this to not be an excuse for more clarity on network stats. It was mentioned that more of these will be available with the new website. Can you specify which stats will be supported?

Joseph Denne: We’ve started to provide more detail on this front and will continue to do so. We are working on client success stories and will be publishing ever deeper levels of data as we move forward and also as we continue to evolve the network explorer.

Question 19

When can we expect the updated roadmap and updated website (please specify what coming weeks mean)?

Joseph Denne: As I think I mentioned in my last weekly update, next week! (And if I didn’t, I meant to!)

Question 20

Self onboarding is one step into decentralization so I guess there will be no restriction and the free market will regulate the node count over time. Is this assumption correct?

Arthur Mingard: In the fullness of time, yes, that is correct. In the first instance self on-boarding will be gated on a first come first serve basis. The reason for this is that we need to monitor and assess performance as the network grows.

Question 21

Can we (as founding node holders) also join the self-onboarding? And how many raspberry PIXs will I be able to run following the launch? (I have a 5-device cluster set up already waiting idle)

Arthur Mingard: Yes, you can. How many devices you can run is limited only by your connectivity and the subsequent performance of the devices in resolving tasks in Gateways job queue.

Question 22

When will the first staking for Gateways and Stargates take place?

Joseph Denne: September for backbone rentals. With some potential for onboarding to start sooner.

Question 23

Will Stargates have the ability to censor specific sites, or will it be up to DADI only?

Arthur Mingard: The right to censorship isn’t something that we believe in as a team, however we have to respect the rule of law within those regions in which we operate. How this is handled within the network when fully decentralized is an open question, but we expect to be handled at Stargate level and on a region by region basis.

Question 24:

How is the Technology Partner programme going? Is this still a thing?

Joseph Denne: It is, yes, although it has taken a backseat to core development in the past few months. We expect to be picking it up again in earnest towards the back end of this year.

Question 25

Is the software on the FNs stable now? Those with original software have the same as the ones with new SD cards?

Arthur Mingard: The stability in the Founding Nodes relates to the update process rather than the node software itself. Because we need support for scale in the network, we need support for unattended updates. This is to support a broad array of devices and to remove the need for a high level of technical understanding to contribute. The auto update process has been one of the most complicated portions of development, other than securing the node software, and this has led to a number of issues with the update process. We see this as par for the course and given the cutting-edge nature of the platform it’s really not that surprising. This is also why we have a backbone of devices in the network: it allows for testing within the Founding Node programme without impacting live services.

Question 26

How could the network withstand an organized attack? (scrape the web for sites running in DADI network, DDOS all those sites simultaneous)

Arthur Mingard: Network scale is a big part of the answer here. But we’re also working with partners to build scrubbing centers into the backbone of the network, which will go a long way to protecting us from DDOS attacks.

Question 27

Does DADI have any international expansion plans?

Joseph Denne: Yes, these are being worked on. As soon as there’s something to share on this front, we will do so.

Question 28

I heard DADI was working on OEM hardware too. Does this involve design and production of PCB schemes and boards?

Joseph Denne: It does. We’re very, very early in this process and are not ready to announce anything on this front yet. Consider it an R&D line, which may or may not come to fruition.

Question 29

How did the network upgrade go? Can we expect a thorough report of the process?

Arthur Mingard: Well! But it’s ongoing and we have changes to telemetry still to make. This may necessitate a further update to Founding Nodes. And yes, I’ll be writing and providing a full update on the process soon!

Question 30

With the branding done, what will the next main challenge be for Ravi and his team?

Ravi Chandwani: The branding process is iterative, so we will continue to work on other aspects of the brand which as yet have not been seen. There are a number of strands which are being worked on, including iconography, illustrative elements as well as the product design. We will continue to develop these with a strong adherence to the overall brand aesthetic and tone. Some of this you will see with the launch of the new site, and some will be added over time.

Joseph Denne: It’s worth noting that we’ve made some additions to https://brand.edge.network as well, and that this will be kept up to date as we go.

Question 31

Can we have a whitelist for Telegram to allow admins to let trusted members of the community post links and images?

Joseph Denne: Yes! We’ve figured out a way to enable this in our bot, and will be rolling it out to our own channel - and to the other crypto projects that our bot power - next week.

]]>
Sat, 13 Jul 2019 15:30:00 +0000 5d29f6f6a26508082aa55d3a Bolaji Oyewole 2019-07-13T15:30:00+00:00
Weekly update: 12th July, 2019 https://edge.network/updates/announcements/weekly-update-12th-july-2019 My goodness it’s been hot this week ☀️ Averaging 30º in London… and it feels a hell of a lot hotter inside without air conditioning.

Here’s hoping the sun stays put through the weekend.

We had an AMA today (transcript here), so this update will be quick.

The network engineering team spent the week pushing a major update to mainnet live. There’s more to do and there have been a few hurdles, but in the main it went off smoothly.

Arthur will be posting an article about it early next week.

We are continuing work on the new site for edge,network, which is scheduled for live next week.

The team also made an significant update to API to store collection schemas in the database, rather than the filesystem. This reinstates endpoints for creating, modifying and deleting collections and includes backwards-compatibility with file-based collection schemas using something we call “collection seeds”. This will allow us to create interfaces for content modelling within our SaaS proposition.

They also removed the version portion of collection URLs in API, simplifying the URL used by data consumers.

A list component was added to our interfaces, allowing editors to insert, update and reorder entries within a field.

And they made a start on the migration of our existing documentation to reflect the change in focus as we roll out the Edge rebrand.

We made a number of updates to integrations with Signify (a fraud prevention system) and Nimble (a streaming media platform) for one of our customers.

A number of regional offices at one of our larger customers have expressed interest in migrating to the latest build of our platform. We’ll be following that up next week.

+We had a handful of meetings with prospective customers for the network

+We put a lot of flesh on the bones of our strategy for scaling the network effectively as part of the core approach for our go to market

+We delivered integration docs for CDN to a new property that will be moving to the network

And that’s it from me! I hope you enjoy your weekends 🙂

]]>
Fri, 12 Jul 2019 17:30:00 +0000 5d2c3f51f094f707c249e155 Joseph Denne 2019-07-12T17:30:00+00:00
Introducing Edge https://edge.network/updates/announcements/introducing-edge Edge represents the culmination of months of work, and synthesises everything that we know about edge computing, together with the core principles of our network in to a coherent and powerful brand.

The core components are presented below, and you can access a full introduction to the brand here: https://brand.edge.network

Logo concept

Image

Edge levels the playing field. It enables everyone to contribute, be part of, and access a network that exists all around us in the devices we use day-to-day. Edge simplifies the complex, and by decentralising the network creates a level of equality and functionality that surpasses any current offering.

Our logo is derived from these ideas, communicating them in a balanced, aesthetically engaging mark. It displays uniformity and simplicity. A perfect circle denotes equality, clarity, ease of use and accessibility. The circle is also a representation of the individual nodes which make up the network.

Reference to the ‘power symbol’ communicates the idea of democratising computational power, bringing it closer to the user utilising it in the devices that surround us. It’s green because it is always on.

The ‘e’ is edge.

Icon

Image

The Edge icon is the foundation of the logo. It is developed from a perfect circle conveying equality and balance. The mark reads as a lowercase ‘e’, whilst also referencing the power symbol.

Using a refined line weight it is both bold and elegant, and it is instantly recognisable The colour is striking and engaging and will ensure standout amongst the plethora of cryptocurrency brands.

The Edge icon is a standalone symbol used in isolation when the wordmark is not appropriate or relevant, for example as a currency symbol.

Wordmark

Image

The Edge wordmark is the primary logo for the brand. The wordmark is constructed using the ‘e’ icon. The ‘g’ is the same shape flipped and rotated. The ‘d’ is derived from the same circular shape. This creates an even, balanced mark which conveys the brand ethos in a clear and elegant way. The four linked circles can be seen as nodes, connected across the edge network.

]]>
Tue, 09 Jul 2019 16:30:00 +0000 5d2457bcf094f707c249e14c Ravi Chandwani 2019-07-09T16:30:00+00:00
Mainnet upgrade, July 2019 https://edge.network/knowledge/network/mainnet-upgrade-july-2019 Scale

Important in the networks growth is its ability to accept an increasing number of Hosts without compromising the speed in which vital coordination components calculate and synchronise topology.

Initial builds of the network manifest distributor operated on geographically spread Consul servers, adjacent to the Stargate services. Whilst decentralised in respect to their lack of a single point of failure, these nodes required consensus on all operations, including healthchecks, key/value updates and service metadata. Put simply, if a Host were to tell its nearest Consul server that it was up and healthy, the same data would need to be fully propagated before the write was considered successful. Operationally, this latency just doesn’t scale, so we started to see some failing health checks whilst telemetry painted a far more serene landscape.

One of the largest single changes in the upcoming release is a shift away from globally propagated service data in favour of a multi datacenter approach, more akin to your typical VPC. The services continue to register with their local Consul server, operating alongside the nearest Stargate, but that’s where the syncronisation stops. Consul offers a little flexibility in the way it connects to its peers, so whilst a multi datacenter setup does not share service data globally it does permit access by proxy – something the network relies upon on rare but vital occasions.

Performance

When debugging the latency experienced by the single data center configuration the team discovered a number of performance improvements to the way service data is stored. Whilst the refactors were complex, the result can be simplified to two key areas:

Health

The first health checks consisted of periodic writes to a key/value directory on a per-device basis. A pruning process existed to remove services that exceeded the maximum TTL.

The latest iteration of the health check process now uses Consul’s inbuilt healthcheck methods, including GRPC standard health endpoints for Stargates and Gateways, and TTL checks for Hosts. This meant that we were able to completely remove the key/value data which had a significant reduction in latency due to the way the Consul propagates service health.

Metadata

Until recently all devices wrote information such as their current connections and build digest to Consul key/value. Whilst it appeared to be largely an efficient method, benchmarking showed us that storing this data within the core Service metadata reduced the number of Consul requests by other services. If, for example, a Host wanted to find the closest operational Gateway it would first need to make a call for all services of type ‘Gateway’, before individually requesting the KV metadata of each Gateway before an informed preference could be made. By moving this data to the Service itself, what could be a query of 1+n became simply a query of 1.

Security

ACL tokens were initially created to offer distinguishably different permissions to the three core services, with the Host application having the lowest level of trust, and the lowest staking. The Host token had read-only access to non-sensitive information, and a small amount of write access to its own part of the network manifest making it almost impossible to use it for evil.

A single ACL token for all Hosts does come with a downside, and one we discovered when migrating to multi DC on testnet. Firstly, Consul doesn’t allow ACL migration which poses a risk when embedding a single token into the Host. It also means ACL for a single machine cannot be revoked should an account be deactivated.

The next release will include a refactored method for retrieving ACL tokens to use with Consul, which required the introduction of Vault, an accompanying service which extends Consul ACL with policy templates which act as a blueprint for the onboarding mechanics ahead of self-onboarding.

What you need to do?

If you’re lucky enough to have a Founding Node in operation you’ll need to have it up and running between the 8th and the 12th July. During this time we’ll be issuing a unique token to the machine before we switch off the old ACL and migrate the network over. Don’t worry if you currently have a support ticket open or your device is disconnected under our instruction – we will reach out to you individually after the update and get you back up and running.

]]>
Sat, 06 Jul 2019 15:30:00 +0000 5d20bc9baa8a2c080bee2817 Arthur Mingard 2019-07-06T15:30:00+00:00
Weekly update: 05th July, 2019 https://edge.network/updates/announcements/weekly-update-05th-july-2019 Today’s update is going to be short and sweet, as we’re mostly head down building in support of the major update to mainnet next week.

The network team completed the mammoth task of configuring the deployment scripts for the update.

This includes the completion of the final parts of the network-sync scripts.

The network explorer was updated to patch the duplication issues that we have been seeing. A further build has been created with our new brand and fonts in place, ready to go live along with our new site.

We’ve completed and released updates to Agent and Host which are fully compatible with the new Vault ACL policies. This includes a refactored service package which dramatically reduces the amount of data to be written to Consul.

Stargate and Gateway have been upgraded with a new service package which fixes a bug with the service de-registration routine (which had caused orphaned services after reboot), and added vault configurations to the Jenkins deployment scripts.

API is being refactored to store collection schemas in the database, which will enable the creation and modification of collections directly within our interfaces, which is a critical step as we move towards a full SaaS proposition for our web services.

This work is now 80% done.

A list-type field for the interfaces is also in progress (see: https://github.com/dadi/publish/issues/389), which is in final stages of development and should be rolled in early next week.

Ravi and his team have been flat out on the Edge brand update, rolling content updates through the new site, applying the brand to the explorer and working through supporting collateral. It’s looking awesome.

I had planned to show you the work this week, but have decided to hold it back a few days to allow time to build out a page in support of the brand itself (rather than show you internal brand guidelines in documents).

This will be ready next week, and will be made available on brand.edge.network

We made significant progress migrating an existing customer’s data layer to API, which is a big step toward full migration to the network.

Tiipr is on target for being dev complete w/c 15th July, at which point we will start live testing in the network. The team there is expecting to go live with an MVP in August. I’ll let you know when it’s available to play with.

Our relationship with the PPA has led to a number of new leads. We have confirmed the dates and attendees for the product workshops that we will running with them in the coming months.

We started early stage discussions with a new partner, which should open up a good sales channel for network services.

We confirmed our attendance at the Great British Business Show in November (http://www.greatbritishbusinessshow.co.uk/), where David Wilde will be speaking about Edge, blockchain and their applicability to Government.

Finally Paul met with a number of car manufactures and automative publishers at Goodwood. He spoke to them about our technology to explore the potential for future engagement.

And that’s it! Enjoy your weekends.

]]>
Fri, 05 Jul 2019 17:30:00 +0000 5d248ba4a26508082aa55d37 Joseph Denne 2019-07-05T17:30:00+00:00
Month in review: June 2019 https://edge.network/updates/announcements/month-in-review-june-2019 June has been another very busy month for us as we have been preparing for a major update to mainnet, working flat out on our rebrand and finalising our roadmaps for the next 12-18 months.

Year to date commits? 3,483 (+357 in June) 💥

DADI is being rebranded to Edge – a move designed to help growth in the network by focusing on core network services. The team is working to launch Edge in the coming weeks: a new site, an update to the explorer, a rename of our ticker to $EDGE, and a clear focus on network-level services.

We will be sharing the brand itself this week.

Aside from the news of our rebrand, highlights this month include our attendance of the PPA awards, our attendance of the Cannes Lions, preparations for the coming mainnnet update, and sustained ROI for network contributors.

We turn six years old on the 08th of July! Quite a milestone.

Contents

  1. Business update
  2. Network status
  3. Development report
  4. Highlighted customer stories
  5. The Foundation
  6. Ask us anything

Business update

We have around 47 active deals in our new business pipeline with a combined value of over 6.2 million USD.

We met with eight new prospects in May. These ranged from security specialists to startups, publishers to healthcare businesses.

We attended and presented at the PPA awards:

Image

It was a really positive evening for us – 20+ new conversations have been opened up on the back of the event.

We finished defining the content of the series of workflow Workshops that we’re running with the PPA, plus we’ve also confirmed a set of new product development sessions in September. These are designed to aid with customer engagement and new business.

We were also in Cannes for the Cannes Lions, where we picked up a number of promising leads.

Joseph met with David Wilde to plan an approach for local government. They also discussed the detail of his ongoing engagement with the project, covering speaking arrangements, articles and an approach for central government.

IP protection

We met with our patent lawyers for an update – there has been a lot of progress made, and we expect to have completed the process in the next few months.

This will be an important milestone for us as it means that we will be in a position to fully open source the network codebase.

Pursuing patents for the core architecture and components of the platform is important because it provides protection for the entire community by putting a level of barrier in place to competition, which should help to establish the Edge network as the protocol for edge computing.

We also submitted fourteen trademark applications related to Edge.

Roadmap updates

We released a preview of our roadmap update, covering some of the key updates to the Edge network for the next 12-18 months.You can access the preview here: https://dadi.cloud/en/updates/network/roadmap-updates/

The full version will be released with our new website.

In the news

The future of our web services

We have made huge strides in the definition of what will become a stand alone SaaS proposition for our web services. We are breaking out the web services in the network and will be offering them under a new brand and on a new domain. The reason for this is to allow us to focus the network on network-level services, better supporting its go to market. This approach also means that we don’t lose the momentum and investment made in the web services to date, and will present a stand alone entry point to use of the network (use of the web services under their new brand will deliver revenue directly back to the network).

This will be announced formally when the full branding work for the proposition has been completed. We expect this to be in July.

Earnings

Earnings in June returned an annualised 56% against the Founding Node proof of stake and we expect the uptrend to continue as more traffic moves in to the network.

Image

Earning reports and payouts for June will happen in the first week of July.

We have gone a long way to enhancing and validating pricing for core services. These will be published on our new site in July, and include the addition of DNS as a service. As we continue to grow network use focus will increasingly move to the optimisation of the economics of the platform, particularly the balance between price and demand, earnings and network growth.

Our goal is to establish a self contained ecosystem that ensures network capacity alongside competitive market pricing for customers of the network.

Hiring

Moses Skoda joined us as an engineer. He will be working closely with Eduardo on the future of our web services.

We are still looking for an experienced technology sales director. If this sounds like the right job for you, please get in touch.

Network status

Our mainnet continues to grow, with significant additional traffic pushed in to the network in June.

The mainnet maintained 100% uptime for the month, with the switching of devices in and out of production in rapid succession tested and shown to have no impact on performance, proving the architecture for Hosts. This means that we expect to be in a position to lower and potentially even remove the minimum uptime requirement for Hosts in future.

Image

Key stats

Stat Value Description
Uptime 100% The mainnet achieved 100% uptime in May
Nodes online 356 Node capacity remained steady. Additional Founding Nodes are planned for release in July/August
Network reach 54 countries There are Hosts online in 54 countries
Stargates 21 All non-earning and under the control of the core team
Gateways 18 All non-earning and under the control of the core team
Hosts 317 All earning and operated by the community
Compute units 1,268 Edge one units: low-level compute capacity for web-based processing (websites, APIs etc.)
Edge storage 9.49 TB Long term object storage
Edge cache 844.70 GB Short term caching

Development report

The network engineering team were flat out throughout June:

They completed most of the final components of network-sync, the service the network uses to persist device data now that the network manifest is ephemeral. This includes recording the duration of device sessions and metadata, which is important for payout calculations in the next phase of development.

They have also been closing out final changes to Vault ACL, finishing the month with an ACL onboard script that’ll be used to automate the founding node migration away from the current ACL implementation.

API’s test suite has been extended even further. All of the 15,000+ database operations generated from the ~1,000 unit tests are now stored as a snapshot and replayed against each data connector module. you can see the new testing module at https://github.com/dadi/test-tools and an example of it in action at https://travis-ci.org/dadi/api-mongodb/jobs/551251808

Our code quality tools have been improved with consistent linting and code formatting (https://github.com/dadi/code-standards) now applied to Publish and API.

The team have built and released automated deployment for Gateways on test.network and have also introduced the first full build of Vault, a vital new component for the next iteration of network ACL. Account registration for ACL was also progressed. This is the last core component required before we start the much anticipated mainnet release.

A bunch of work has been done to simplify the way that ACL is passed between Agent and Host, which makes the self onboarding process faster, more lightweight and more secure.

We have been hard at work defining and building out the Edge DNS layer, and completed the addition of support for AAAA, CNAME and TXT records.

We completed and successfully deployed auto SSL generation and distribution to the network, so SSL is enabled by default and is now free for all network customers. You can read about this here: https://dadi.cloud/en/knowledge/network/edge-ssl/

API update

API version 5 was released! Release notes as follows: https://github.com/dadi/api/releases/tag/v5.0.0 And documentation can be found here: https://github.com/dadi/api/releases/tag/v5.0.0

Publish update

We’re around 95% through a conversion of Publish to React. The approach here is designed to better align our interfaces to one of the webs most active communities. It also opens up a lot of new functionality for our interfaces.

A network ready build of Publish is also in testing.

Highlighted customer stories

Ecohustler

Image

Ecohustler went viral. One of their articles (https://ecohustler.com/article/one-corporation-to-pollute-them-all/) was picked up by just about every social platform you can imagine.

At peak the site was receiving 1k+ concurrent connections. That’s an increase in traffic of 16,566%. Pretty significant!

Uptime during the stampede? 100%

The Foundation

Our charitable foundation attended Rightscon in Tunis, where they met with other organisations and charities to discuss how technology can be leveraged to properly support human rights around the world.

They are continuing to develop a series of training sessions for human rights organisations. If you would like to volunteer to help them deliver the pilot series, please reach out to Jennifer - the CEO of the DADI Foundation - directly on: jennifer@dadi.foundation

They will need half a day of your time. You can be based anywhere in the world, and you’ll earn a warm sense of well-being knowing that you’re helping to make the world a better place 🙏

You can read more about the foundation here: https://dadi.foundation

Ask us anything

The core team is active across all of our community channels, and we operate on the basis of as full a level of transparency as possible. You can always reach out to us. If you don’t get a reply to a question right away, our community moderators will add it to their backlog for the next AMA session.

On which, you can read the transcript of our last AMA here: https://dadi.cloud/en/knowledge/network/ama-recap-june-07th-2019/

Our next AMA will take place on Friday the 12th of July.

The session kicks off at 2pm UTC on Telegram, Reddit and Discord: get you questions in now.

Weekly updates

I write updates for our community covering all aspects of our work. You can access the updates for June here:

]]>
Sun, 30 Jun 2019 17:00:00 +0000 5d18de108896c3081cdd42a1 Joseph Denne 2019-06-30T17:00:00+00:00