Intra-domain Traffic Engineering (TE) refers to the methods and practices used within a single administrative network domain (like a single ISP or enterprise network) to optimize the utilization of network resources and enhance overall network performance.
Key points about Intra-domain Traffic Engineering:
Scope:
- Operates within a single Autonomous System (AS) or administrative network.
- Does not involve cooperation or communication between different ISPs or domains.
Goals:
- Maximize efficient use of network resources (e.g., bandwidth, links, routers).
- Balance network load to avoid congestion.
- Minimize latency, jitter, and packet loss.
- Ensure reliability and robustness by adapting to link failures or network issues.
Methods and Technologies:
- Routing protocols adjustments (e.g., OSPF, IS-IS).
- Multiprotocol Label Switching (MPLS) for explicit route selection.
- Software-Defined Networking (SDN) for dynamic and programmable path management.
- Network monitoring and measurement tools for traffic load assessment and route optimization.
Limitations:
- Typically performed periodically (e.g., every few minutes or hours).
- Primarily optimized around aggregate traffic demands and utilization metrics rather than per-flow QoS demands.
- Does not handle end-to-end QoS across multiple domains or WANs (inter-domain).
Examples of Systems/Implementations:
- Google’s B4 WAN.
- Microsoft SWAN (Software-driven WAN).
- Facebook’s Express Backbone (Espresso).
Contrast with Inter-domain TE:
- Intra-domain TE: Within a single administrative domain, optimizing internal resource usage.
- Inter-domain TE: Across multiple domains (e.g., across different ISPs), optimizing path selection based on inter-domain routing protocols like BGP.