ECI's Data Networking Division is among an elite group of companies that have successfully developed Internet-class routing capabilities, and its flagship ST-series Service Edge Routers contain routing as robust as that found in widely deployed core routers. Designed from the ground up with the additional routing intelligence required at the multi-service edge, the ST-series routers know where packets are coming from and where they are going, to provide multiple service levels and service level guarantees to customers to shape and groom bandwidth and combine multiple services on a single infrastructure.

Complete Internet Protocol Support

The ST-series includes a highly scalable, Internet-class routing engine designed specifically to meet the stringent edge requirements of the world's largest carrier and ISP networks. ST-series routing was developed in-house and includes a feature-rich suite of routing protocols including unicast routing (BGP-4, OSPF-TE, IS-IS-TE, RIPv2), multicast routing (MBGP, PIM-SM, IGMPv2) and MPLS switching (RSVP-TE, LDP). As a result, the ST-series interoperates seamlessly with current IP backbones to ensure reliable end-to-end service delivery. With the ST-series at the network edge, service providers can scale to meet customer demand without sacrificing features or reliability.

Highly Scalable Routing Design

The ST-series Route Control Processor (RCP) runs the ShadeTree System Software and the full suite of routing protocols. All forwarding is distributed across multiple (up to eight) Network Processor Blades (NPBs) that perform all packet processing, Layer 2/3 forwarding, IP filtering, accounting and link management for the system at wire speed. The NPBs support advanced services such as VPNs (which scale to over 4,000,000 VPN routes), where each customer has a unique forwarding table. By selectively distributing routes to each NPB, forwarding table capacity increases as cards are added to the system. This is in sharp contrast to routers that use a centralized forwarding table, which limits the number of customer-facing interfaces.

Since NPBs handle all per-customer link management, Layer 2 processing and performance monitoring functions, the ST-series RCP can focus on the task of scalable routing.

Scalability Performance

The proof is in the numbers, and ST-series scalability performance sets a new standard at the edge and even bests core routers, in many cases. For example, extensive in-house and independent testing of the ST-series found the following scalability results.

10,000,000 BGP RIB-In paths
1,000+ BGP peers
2,500+ OSPF instances
8,000 IP interfaces per line card
64,000 ATM interfaces per line card
64,000 VLAN interfaces per line card
64,000 Frame Relay interfaces per line card
1,000,000 global IPv4 routes per forwarding table
4,000,000 VPN IPv4 routes per system
16ms backbone fail-over for up to 512,000 routes

Efficient Memory Consumption

The ST-series was designed using an optimized memory consumption approach requiring less memory than other routers, and running with lower CPU utilization. For example, the ST200 supports 1,000+ EBGP sessions with minimal CPU impact, while other routers with shared processes will be strained at a much lower number of sessions. This results in dramatic increases in scalability, with a larger number of routers and peer sessions can be supported per device. This density is critical to achieving maximum profitability per customer.

Conclusion

Internet-class routing is required to offer scalable services at the edge of carrier IP/MPLS networks. ECI's ST-series is among a select group of truly carrier class devices with complete Internet Protocol support. Additionally, the ST-series uniquely offers and a highly distributed, efficient design that results in unmatched levels of routing scalability and performance.