Advanced Routing
Organizations developing applications and services for global audience reach will require enterprise class network design and deployment. By utilizing a diverse IP transit portfolio and incorporating peering points into network design, application delivery can be optimized across multiple network paths with reduced latency. Advanced network design must also include levels of redundancy on the hardware side to allow for the use of failover routing protocols.
Altexxa Group believes strongly in the fundamentals of bandwidth availability and diverse networking. In addition to working with standard WAN deployments using industry standard practices, we also work in R&D labs to evaluate the needs of future networks. We believe advanced network deployments will not only offer high availability bandwidth and failover redundancy, but also work with software solutions to allow automation via outside orchestrator tied in to router operating systems.
- Enterprise class design with IP transit and peering networks
- Optimized application delivery and reducing network latency
- Incorporating failover and redundancy designs into deployments
- Allowing routing systems to be driven by software orchestrator
Considerations
When designing and deploying your enterprise class network, considerations must be made based upon certain key factors to determine which routing methods best apply. Altexxa Group primarily works with Cisco datacenter deployments and our choice of routing protocols will reflect this (EIGRP, HSRP), whereas deployments utilizing either a mix of vendors or one single vendor such as Juniper would require changes in logical design. WAN design considerations would include availability of IP transit and peering networks, and analyzing possible geographical risks along carrier route paths to avoid critical failures. Establishing a global presence and deploying private circuits to build your transit and peering network portfolio should become the common key component in all design considerations.
- Establishing long term vendor partnerships for protocol selection
- Diverse WAN routing path to avoid critical carrier failures
- Large scale deployment presence for optimal routing paths
Routing Protocols
Altexxa Group works primarily with BGP and OSPF for route distribution through the border and core network layers. For standard Cisco based datacenter deployments, we also work with EIGRP for its advanced metrics and efficient neighbour communication with RTP. Our core network level also utilizes VRRP for redundancy, or HSRP in Cisco specific deployments. Edge routing at remote sites utilizing a BSD based appliance use CARP for redundancy and load balancing features. Altexxa Group Labs is also working with GLBP to add further balancing between redundant clusters in Cisco based deployments. We employ the use of IPsec to protect communication between sites over a WAN.
- Dynamic BGP/OSPF deployments per industry common practice
- VRRP and HSRP for redundancy on core routing layer
- Remote office communication secured via IPsec transport
- EIGRP and GLBP seeing futures in advanced network deployments
Peering Networks
In addition to an IP transit carrier portfolio, developing a peering network at Internet Exchange hubs is beneficial to creating low latency networks for application content delivery. The geographical location of your target audience will generally dictate the ideal exchange points for optimal delivery. Establishing global peering agreements can also benefit CDN implementation by allowing for ideal routing paths between content mirrors and end users. The concept of peering networks is based upon the perception of the Internet as a global community with organizations working together in mutual benefit.
- Establish peering agreements at global hubs (SIX, NYIIX, LINX, Equinix, TORIX, etc)
- Reducing network latency by optimal routing paths
- Mutual benefit between peering networks and organizations
- Building upon the foundation of the global Internet community
IP Transit
Carrier selection for incorporating IP transit into your business model involves review of route diversity and backbone availability. We place heavy importance on long haul topology and native IPv6 support via dual stack networks. Modern network design will often involve Ethernet delivery, though availability of TDM connectivity via T1/DS3 circuits may offer additional benefit for remote office connectivity where reduced latency is required.
Altexxa Group believes a carrier portfolio should include a mix of both regional and Tier 1/2 providers, and also include diverse topology for optimized geographical coverage. This will allow locally driven organizations with optimal peering while providing advantages for global reach.
- Native IPv6 support for future IP allocation requirements
- Circuit delivery options according to organization needs
- Diverse carrier portfolio for optimal network latency