> For the complete documentation index, see [llms.txt](https://doc.datagram.network/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://doc.datagram.network/documentation/whitepaper/3.-datagram-architecture/3.1.-the-datagram-node-network-and-fabric-networks.md).

# 3.1. The Datagram Node Network & Fabric Networks

Datagram operates as a highly scalable, low-latency network designed to serve as the foundational layer for
 DePIN applications. While it leverages the flexibility of a dedicated Avalanche L1 for tracking uptime and usage
 statistics, Datagram is not limited to a single blockchain ecosystem. Instead, it is designed to be usable to inform
 distributions for any blockchain network, ensuring businesses can integrate their decentralized infrastructure
 seamlessly.

The Datagram Node Network serves as the backbone of the Datagram ecosystem, providing a globally distributed
 infrastructure that supports both native and external DePIN projects, enabling seamless access to decentralized
 compute, bandwidth, and storage. Existing DePIN networks can integrate into Datagram to enhance their
 resource efficiency, while new DePIN projects can deploy instantly without needing to build infrastructure from
 scratch. Additionally, Web2 and Web3 businesses can leverage the network via API/SDK integrations to access
 scalable, decentralized services without managing blockchain complexities.

Underneath it are Fabric Networks, which represent independent DePIN networks that integrate with Datagram’s
 infrastructure. These networks leverage Datagram’s resources while maintaining their specialized
 operations, enabling scalable and interoperable decentralized services. These are connected via the Datagram
 Core Substrate (DCS), covered in Section 3.3, which acts as the connectivity layer that ensures seamless
 communication, security, and efficient resource allocation across the entire network.

Datagram Cores are distributed network nodes underpinning Datagram’s Fabric Networks, handling routing,
 validation, and data optimization. These Cores facilitate high-speed, reliable communication by efficiently
 processing and transmitting data across the network. They also act as decentralized infrastructure elements
 designed to route and manage traffic within a distributed system, boosting scalability, security, and overall
 network performance. These Cores operate as decentralized Beowulf clusters, which help ensure
 high-performance and fault-tolerant data pathways.

There are 5 types of Cores:

1. **Full Core:** The backbone of the Datagram network, Full Cores handle critical network functions, including
    routing and optimizing data flow. They contribute computing resources to maintain network efficiency
    and receive rewards for their role in securing and scaling the infrastructure. One must own a Core Token
    to be a Full Core.
2. **Partner Core:** Designed to provide additional computational support, Partner Cores assist with load
   \
   balancing during periods of high network demand. They help prevent congestion by redistributing traffic efficiently, ensuring smooth and uninterrupted service for all users.
3. **Device Core:** These Cores are based on IoT devices such as televisions, routers, sensors, or any other
   \
   device integrated into the system. They utilize the Datagram Core Substrate to provide load-balancing
    services during periods of downtime, contributing to network efficiency by sharing idle processing power.
4. **Hardened Core:** Designed for high-security or priority traffic, such as for government operations or B2B
    communications. They have heightened security, and though they serve the entire network, their
    primary focus is on handling sensitive data and communications.
5. **Consumer Core:** These are localized Cores initiated by users on their hardware. Consumer Cores
    temporarily host services and allow users to benefit from Datagram's infrastructure without requiring
    permanent deployment, offering flexibility and localized service management.

Furthermore, by utilizing a modular and application-specific blockchain approach, Datagram optimizes real-time
 data transmission for latency-sensitive applications such as decentralized video conferencing and AI-driven
 communications. The overall result of Datagram’s architecture is that it benefits from:

* **Customizability:** Unlike shared Layer 1 blockchains, Datagram’s Subnet is fully customizable, enabling
   enterprises to optimize network parameters to suit their specific needs. While gas fees for transactions
   within the Datagram ecosystem follow a standardized structure, enterprises can still control aspects
   such as payment models, resource allocation, and fee distribution within their own applications,
   ensuring cost efficiency and predictable expenses.
* **Independent Execution Thread:** Each fabric network operates with its own independent execution execution environment, preventing competition for resources and eliminating congestion from other networks.
* **Lower Transaction Costs:** By diverting traffic away from the Avalanche Primary Network and operating
   as a separate Subnet, Datagram significantly reduces gas fees while maintaining high-speed
   transactions.
* **High Throughput:** Datagram leverages Avalanche’s sub-second finality, near-instant transaction
  \
  settlement, and parallelized execution to support high-demand applications without congestion or
  \
  bottlenecks.
* **Validator Selectivity:** Datagram’s Subnet allows for controlled validator participation, ensuring optimal
   performance while maintaining decentralization.\\