Kumulus Cloud

[morgueye4]

Project Abstract

This application serves as a follow-up grant of the previous work Futur Protocol. This current project, titled Kumulus Cloud, stands as an important component in the Relai Network, specifically functioning as its decentralized compute and storage layer.

In this iteration of the project, we aim to establish and validate the efficacy of a decentralized infrastructure for compute and storage. To achieve this, we are leveraging a marketplace model similar to that of the Akash Network protocol, which has demonstrated significant success in decentralized cloud services. The marketplace approach allows us to harness a wide array of distributed resources, thereby enhancing scalability, redundancy, and data sovereignty across the network.

Grant level

• Level 1: Up to $10,000, 2 approvals
• Level 2: Up to $30,000, 3 approvals
• Level 3: Unlimited, 5 approvals (for >$100k: Web3 Foundation Council approval)

Application Checklist

• The application template has been copied and aptly renamed (project_name.md).
• I have read the application guidelines.
• Payment details have been provided (Polkadot AssetHub (DOT, USDC & USDT) address in the application and bank details via email, if applicable).
• I understand that 30% of each milestone will be paid in vested DOT, to the Polkadot address listed in the application.
• I am aware that, in order to receive a grant, I (and the entity I represent) have to successfully complete a KYC/KYB check.
• The software delivered for this grant will be released under an open-source license specified in the application.
• The initial PR contains only one commit (squash and force-push if needed).
• The grant will only be announced once the first milestone has been accepted (see the announcement guidelines).
• I prefer the discussion of this application to take place in a private Element/Matrix channel. My username is: @_______:matrix.org (change the homeserver if you use a different one)

[semuelle]

Thanks for the application, @morgueye4, and welcome back. I have a couple of questions about the application and the project in general:

• In your last grant application, you listed the Fusion Protocol as the next step, but this one is focused on something else and lists the protocol again as future step. Why the change?
• How do Futur/Fusion and Kumulus relate to another?
• We have received several applications about decentralized compute in the past. Have you studied similar projects in our ecosystem (and related ecosystems) in terms of market fit and any reason why other projects might have been unsuccessful?
• Will you be relying on grants for the foreseeable future, or do you have a roadmap to self-sufficiency? Have you done any outreach towards or research on possible users/customers and market size?
• I found manuals on your blog to publish apps and books. What application does this refer to? Can I test this myself?

Feel free to amend the application with any data you might have.

[morgueye4]

Hi @semuelle thank you for the feedback and the question

• In your last grant application, you listed the Fusion Protocol as the next step, but this one is focused on something else and lists the protocol again as future step. Why the change?

Certainly, we initially planned to proceed with the Fusion Protocol MVP; however, upon further reflection, we realized that the compute/storage layer is essential component. Both Futur and Fusion Protocols will rely on it with backend services, functions, and other computational needs. Additionally, the Fusion Protocol will emphasize partnerships, particularly with Original Equipment Manufacturers (OEMs). We believe it would be advantageous to establish an initial compute/storage layer first. This approach allows us to refine Futur and also serves as a better foundation for building the Fusion Protocol.

• How do Futur/Fusion and Kumulus relate to another?

Futur and Fusion protocols are reliant on robust backend functionalities, including compute and storage capabilities, which will be supported by Kumulus.


• For Futur Protocol, a dependable infrastructure that can handle high-performance tasks effectively is crucial for various tasks such as processing app submissions and downloads on devices, among other use cases.


• The Fusion Protocol, on the other hand, is focused on establishing incentivized partnerships with OEMs, along with providing a marketplace and specialized tooling and infrastructure for devices that incorporate an adapted Futur store app, which varies based on the device type (e.g., phones, TVs, wearables). 

Also there is a significant requirement for a backend infrastructure capable of managing data from these devices. This could be analogous to services like iCloud or Google Drive, where users can store their data,..

Additionally integration with IPFS is still planned and Kumulus could be augmented with IPFS integration (via Crust and the like) for use cases where immediate speed is not a priority, such as in cold data storage, archiving scenarios.

This approach ensures that our backend solutions remain versatile and scalable, effectively addressing both the short term and mid term and future data management requirements of the Futur and Fusion protocols.

Furthermore, the architecture of Kumulus goes beyond serving only the Relai Network(Futur and Fusion) it is designed as a cloud computing marketplace and enables other projects to build and deploy their cloud-based applications on.

• We have received several applications about decentralized compute in the past. Have you studied similar projects in our ecosystem (and related ecosystems) in terms of market fit and any reason why other projects might have been unsuccessful?

Indeed, we have examined projects within polkadot ecosystem (focused on computing but not storage) such as DataGen Project. What we noticed is that employed approaches that may not be ideally suited for on-chain execution, particularly for complex computing tasks. The blockchain, while transformative, is not inherently designed for high-throughput computing due to its limitations in speed and scalability when handling such tasks.
It is worth noting that project was at PoC stage is not live and there might be changes in directions in the future.


Our approach differentiates by adopting a model similar to the Akash Network's decentralized cloud compute style leveraging containerized technologies, in our case focusing on Kata Containers. Kata Containers uniquely combine the security benefits of virtual machines (VMs) with the resource efficiency and scalability of containers. This hybrid approach addresses critical vulnerabilities and performance bottlenecks , thereby offering a more secure and efficient platform and more innovation could be brought by adding gradual decentralization on orchestration, RBAC, …

• Will you be relying on grants for the foreseeable future, or do you have a roadmap to self-sufficiency? Have you done any outreach towards or research on possible users/customers and market size?

Our project is currently in its initial stages of implementation. We are leveraging grants to kickstart these foundational pillars.
After successfully completing the initial phase, our plan is to grow solidify our community and enter active funding rounds this year.


The project targets a wide and varied audience.
Future Protocol:

• App developers : distribute applications across various platforms such as mobile devices, smart TVs, wearables, and any other devices supported by the Fusion protocol, requiring app stores.
• Companies : establish their own customized app stores through app store white labeling.
  Kumulus Cloud:
• Provides services for Relai Network, including Fusion and Future Protocol.
• Web2/3 projects having cloud computing needs (computing, storage, AI Inference, Training,…)

Fusion Protocol:

• OEMs: infrastructure for managing user data from Fusion-issued devices and other related aspects.
• Device buyers to purchasing devices on the Fusion marketplace chain
• Device users purchasing services hosted or provided via Fusion (for example a decentralized Netflix hosted onchain and served via Fusion Smart TV,…)
  Here is a report by messari about Depin Sector which is being more and more adopted and. The Kumulus cloud like Akash fall in this category.

Relai Network with it's services can be seen like a decentralized version of Google or Apple
with the following services/products:

• Play store / Apple store for mobile, wearables and TV
• Google Play console
• iPhone/Google Pixel phones (privacy and web3 focused)
• iCloud or Google Drive for Data storage
• Infrastructure as a service for OEM for users device data management
• Google Compute Engine (Kumulus Cloud)
• ...

• • I found manuals on your blog to publish apps and books. What application does this refer to? Can I test this myself?

The manual refers to the previous Futur protocol application and you can follow it to test the assets deployment.

[semuelle]

Thanks for the clarification, @morgueye4. To be honest, I am concerned about the long-term viability of the project, not because I doubt your ability or resolve, but because we have seen many attempts at building decentralized compute markets and none have been particularly successful. I assume it's just incredibly difficult to compete with the cheap centralized alternatives. Then, from the utility point of view, deliverables 2 and 3 will have little reuse value as they are very specific to your architecture and 1 is a fairly basic pallet. Anyway, I am going to share the application with the rest of the committee and they might disagree with me.

Just one more request: I don't see a Docker deliverable. Due to the complexity of your architecture, please add some kind of automation/wrapper so we don't have to follow a 30-step tutorial to test the deliverables.

[morgueye4]

Thanks @semuelle for your thoughtful feedback and allow me to try to bring more clarity.

Regarding your concern about the viability of decentralized compute markets,
I acknowledge the challenges posed by centralized alternatives .
However, it's crucial to note that our approach diverges from previous attempts in significant ways.

Unlike other projects, we have adopted a strategy similar to Akash, leveraging off-chain mechanisms for compute while utilizing on-chain processes solely for marketplace and management tasks. This approach, proven successful by Akash which is facing growth.

Furthermore, apart from other markets our project has a focus on addressing the needs of the African market, which is currently underserved (by centralized alternatives or even Akash ).

Just one more request: I don't see a Docker deliverable. Due to the complexity of your architecture, please add some kind of automation/wrapper so we don't have to follow a 30-step tutorial to test the deliverables.

Sure, I have updated the application to add the Docker to the deliverables.

[PieWol]

I sadly have to agree with @semuelle here. I'm also not in favor of this proposal.

[morgueye4]

Thank you @semuelle and @PieWol for the feedback.
I do understand ! Please feel free to close the PR anytime.

[takahser]

@morgueye4 thanks for your interest in our grants program.
I do agree with @semuelle and @PieWol that decentralized storage solutions & marketplaces are very challenging, and that none have really picked up so far. It's a very difficult problem to solve, and we already have a grant with moonsong labs that works on a decentralised storage solution. For there reasons, unfortunately I've decided not to add my approval here, neither.

I do understand ! Please feel free to close the PR anytime.

Understood, will close it. Thanks and feel free to apply again in the future.