Spheron Network: Unlocking Shared GPU Resources on a Decentralized Network

CHIRAG KUMARCHIRAG KUMAR
5 min read

Table of contents

Key Solutions

  1. Seamlessly Lease GPU Resources
    Tap into a vast global network of GPU providers to easily lease the computational power you need.

  2. Maximize Utilization and Monetization
    Turn idle GPU resources into a valuable asset, ensuring they are fully utilized and generating income.

  3. Boost Computational Capabilities
    Access additional GPU power for demanding tasks like machine learning, scientific simulations, and CGI rendering.

  4. Earn Passive Income as a GPU Supplier
    Monetize your idle GPU resources by offering them to those in need, creating a steady stream of passive income.

Challenges with Existing Solutions

Today, Existing solutions for accessing GPU resources typically involve centralized cloud providers or command-line interfaces (CLI) provided by specific vendors. While these solutions are effective, they come with limitations such as high costs, vendor lock-in, and complex configurations that can be difficult to manage, especially for users who need flexible and scalable resources

The Challenge with Traditional Solutions

  1. Centralized Cloud Providers
    Cloud providers like AWS, Google Cloud, and Azure offer robust GPU resources, but their centralized nature leads to higher costs, latency issues, and a reliance on a single provider. Users are often locked into specific ecosystems, making it difficult to switch providers or scale across multiple platforms.

  2. Vendor-Specific CLIs
    Command-line interfaces offered by GPU vendors provide direct access to GPU resources, but they typically require advanced technical knowledge to configure and manage. These solutions are often rigid, offering limited flexibility and making it challenging for users to optimize their resource utilization.

Spheron Network: Revolutionizing GPU Access

Spheron Network disrupts these traditional models by providing a decentralized alternative that connects users with a global network of GPU providers. This approach offers several key advantages:

  • Decentralization for Flexibility
    Unlike centralized cloud providers, Spheron leverages a decentralized network, allowing users to access GPU resources from multiple providers across the globe. This flexibility reduces costs, improves performance, and mitigates the risks associated with relying on a single provider.

  • User-Friendly Interface
    Spheron simplifies the process of leasing and managing GPU resources, offering a more intuitive interface compared to traditional CLIs. Users can easily browse, select, and deploy GPU resources without needing deep technical expertise.

  • Cost Efficiency and Scalability

    By decentralizing GPU access, Spheron enables more competitive pricing and greater scalability. Users can choose the most cost-effective and geographically optimal resources for their specific needs, making it easier to scale up or down based on demand.

  • Empowering GPU Suppliers with 💲💲
    For those with underutilized GPU hardware, Spheron provides an opportunity to monetize these resources. Suppliers can offer their GPUs to the network, ensuring maximum utilization and generating passive income.

What's new !!!

Spheron Network is rolling out some exciting updates that are sure to make your experience smoother and more powerful than ever. Here's what's new:

  1. Slark Node Enhancements

  2. CLI Improvements

  3. Provider App - Dashboard Updates

Slark Node Enhancements

Maintenance Mode Switching:
Slark Nodes will now automatically switch to maintenance mode if a provider fails the health check three times consecutively. This feature ensures that only healthy nodes remain active, improving overall network stability.
Unused Resources Challenges:
Slark Nodes will only challenge providers on resources not currently in use. This means more precise and relevant challenges, ensuring that providers are only queried about unutilized resources.
Inactive Provider Checks:
Implemented checks for providers in 'maintenance' (inactive) mode. If a provider remains inactive, their rewards will be slashed to maintain fairness and incentivize active participation.
Challenge Batch Processing:
Updated our approach to challenge batches, where each challenge is now independent of the number of CPU cores. This change simplifies challenge processing and improves efficiency.
Resource Weight Adjustment:
Adjusted resource weighting to give more importance to GPUs. CPU weight is now set at 20% compared to 80% for GPUs, reflecting the greater importance of GPUs even if CPU numbers are higher.
Claimed CPU/GPU:
Introduced the concept of claimed CPUs/GPUs in challenge batches so that providers can now see only the challenges for resources they have claimed on their dashboard, regardless of the actual number available.
Power Efficiency Metric:
This metric will be displayed on the provider dashboard to help providers understand their uptime and resource distribution accurately.

CLI Improvements

Improved Price Display:

  • Prices are now shown in base token value with decimal precision.

  • All prices are consistently displayed per hour for easier comparison.

Transaction Reliability:

  • Implemented a retry mechanism for bid transactions to improve success rates.

Deployment Cancellation:

  • Fixed an issue where deployments were not being properly canceled when transactions failed.

Provider App - Dashboard Updates

New Provider Registration State:
Added a new state to display when a provider is registered. This addition complements the existing active and inactive states, providing a clearer understanding of a provider's status.
Optimized Reward Calculation:
We've optimized the logic for calculating total accrued rewards, ensuring more accurate and faster reward tracking for providers.
Segregated GPU & CPU Challenges:
Challenges are now segregated by GPU and CPU. Previously, challenges were displayed in an aggregated format, but this update allows for more precise tracking and management of each type of resource.
Updated Withdrawal Fees:
Withdrawal fees for provider earnings have been updated to 10%. This adjustment aligns with our commitment to sustainable platform operations and provider support.

Stay updated and join the conversation on our social channels:

💡
Thanks for Reading!


We’re thrilled to have you as part of the Spheron Network community. Your support and feedback drive us to keep improving and innovating. If you have any thoughts, or suggestions, or just want to say hi, drop a comment below. Stay connected, and keep exploring the future of decentralized computing with us!

0
Subscribe to my newsletter

Read articles from CHIRAG KUMAR directly inside your inbox. Subscribe to the newsletter, and don't miss out.

spheron.network@spheron @spheron-cli @blockchain

Written by

CHIRAG KUMAR
CHIRAG KUMAR

Full stack Developer