Static vs dynamic routing: when to use which

The decision in one paragraph

A static route is typed in by an administrator and never changes unless someone edits it; a dynamic route is learned from a routing protocol such as OSPF, EIGRP, RIP or BGP and updates itself when the topology changes. Use static routes where the path is fixed and small: a stub site with one exit, a default route to an ISP, a host route to one server. Use dynamic routing where there are many routers, several possible paths or frequent change. The exam rule: if the scenario says the network must adapt automatically to a failure, the answer is dynamic; if it says low overhead, a single path or tight control, the answer is static.

Scenario decision table

What static routing costs and gives

Static routes use no bandwidth and almost no CPU, they are predictable, and the administrator controls the exact path. The AD of a static route is 1 by default, so it beats every dynamic source for the same prefix (administrative distance explains why). The cost is manual work: every router needs routes in both directions, and when a link or next hop fails, a plain static route stays in the table as long as its next hop is reachable, so traffic goes into a black hole until someone fixes it.

Three static route types appear in the CCNA objectives: a default route (0.0.0.0/0, or ::/0 for IPv6), a network route to a remote subnet, and a host route with a /32 (or /128) mask. A fourth, the floating static route, is a static route with an AD above the primary source. For example, with OSPF as the primary path at AD 110:

ip route 0.0.0.0 0.0.0.0 198.51.100.1 130

The trailing 130 keeps the route out of the routing table while OSPF supplies a default route. If the OSPF route is withdrawn, the static route is installed.

What dynamic routing costs and gives

Dynamic protocols discover neighbors, exchange routes and recompute paths after a change, so they scale and recover without a human. The costs are CPU and memory on the router, protocol traffic, convergence time, and a larger configuration surface (a wrong neighbor setting or an unwanted route can spread through the whole network).

Exam questions rely on a short classification:

RIP is the standard example of a limit: the maximum usable metric is 15 hops, and 16 means unreachable (RFC 2453), so it does not suit large networks. OSPF scales through areas; the single-area case, with neighbor states and the DR/BDR election, is in OSPF explained.

A worked example: choosing for a small company

A company has a head office router HQ and two branch routers B1 and B2. Each branch has one link to HQ, and HQ has two ISP links.

• B1 and B2 each need exactly one route: a default route towards HQ. Static, two commands in total.
• HQ needs routes to the two branch LANs: static network routes, or OSPF if branches are added often.
• HQ to the ISPs: a primary static default route with the default AD of 1, and a floating static default with AD 250 to the second ISP. If the first ISP’s next hop becomes unreachable, the second takes over.

One caveat the exam also likes: a floating static route only works as a failover when the primary route actually disappears from the table. If the primary link stays “up” but the ISP is broken further away, a plain static route remains in the table. The usual fix is route tracking or a dynamic protocol, which is beyond the entry-level objectives.

Which exams test this

How exam questions are built around this choice

Questions rarely ask for a definition. They describe a network and a requirement and ask for the approach. The decisive words are “automatically” (dynamic), “a single path”, “minimal overhead” or “stub” (static), “backup only if” (floating static), and “between two ISPs” or “between organisations” (BGP). A common wrong answer is a dynamic protocol for a one-exit branch, which works but is more than the requirement asks for. Another is a static route with the wrong direction: a branch needs a route to HQ and HQ needs a route back to the branch, or return traffic is lost. Questions with an ip route line test whether you read the last number as AD, not as a metric. Subnet boundaries inside the route matter too; revisit subnetting step by step if the prefix arithmetic is slow.

Two practice questions

Go deeper

• RFC 2453, RIP Version 2, for the 15-hop limit
• Cisco: what is administrative distance, including floating static routes
• Related on this site: administrative distance and the routing table, NAT types, the CCNA IP routing guide and the Network+ implementation guide.