RFC 3439 (Internet Architectural Guidelines) Quiz

Q1: Which design choice best follows RFC 3439's Simplicity Principle?

Multiple Choice

A. Add every foreseeable feature to the shared core before deployment B. Prefer the simplest architecture that satisfies the required behavior C. Optimize each component independently even when it increases cross-component coupling

**Explanation:** RFC 3439 treats complexity as a primary obstacle to efficient scaling and a driver of CAPEX and OPEX. A: speculative feature richness expands interactions. B: the principle pushes architecture toward the simplest sufficient solution. C: local optimization can add complexity and tighter coupling.

Q2: What does the end-to-end argument imply about state in the Internet core?

Multiple Choice

A. The core must contain no state of any kind B. Every endpoint session should be replicated into the routing system C. Coarse core state may exist, but its interaction with endpoint state should be minimized

**Explanation:** A: the core necessarily maintains coarse-grained state such as routes. B: coupling routing to each endpoint session undermines simplicity and fate sharing. C: RFC 3439 emphasizes near-orthogonality between core state and endpoint communication state.

Q3: A configuration change for one tenant triggers a global recomputation across every gateway. Which principle identifies the concern?

Multiple Choice

A. The Amplification Principle B. The requirement for a universal interworking function C. The rule that every layer must repeat error recovery

**Explanation:** A: small perturbations can cause disproportionately large events at scale; RFC 3439 recommends making local changes have local effects. B: the RFC warns against universal interworking. C: duplicated functions across layers are an example of avoidable complexity, not a requirement.

Q4: Two individually safe recovery mechanisms repeatedly trigger each other and create a cascading outage. Which analysis fits best?

Multiple Choice

A. The mechanisms prove that more components always increase reliability B. Tight coupling created an unforeseen feature interaction C. The failure occurred because the payload was transported without modification

**Explanation:** A: component count and interaction can reduce stability. B: the Coupling Principle predicts increased interdependence and unforeseen interactions as systems grow. C: preserving payload is the Minimum Intervention guidance; it does not describe mutually triggering recovery loops.

Q5: Why can a smart-edge, simple-core architecture reduce the cost of adding a new service?

Multiple Choice

A. It requires every core node to implement the service before any user benefits B. It removes all control planes and forwarding state from the network C. Participating endpoints can adopt the service without upgrading the entire core

**Explanation:** A: that is the expensive smart-core model the RFC contrasts. B: a simple core still has routing and forwarding functions. C: distributing edge software to consenting users avoids a network-wide core upgrade.

Q6: How should “Layering Considered Harmful” be read in RFC 3439?

Multiple Choice

A. Layering has structural value, but excessive rigid layers can duplicate work and obstruct global optimization B. Every protocol must collapse application, transport, and link functions into one component C. More vertical layers always lower OPEX by isolating every decision

**Explanation:** A: the RFC acknowledges layering's conceptual benefits while warning about ordering constraints, hidden information, and repeated functions. B: it does not ban modular boundaries. C: its operational observation is that extra layering often increases complexity and OPEX.

Q7: A gateway can encapsulate a payload unchanged or translate every field into a universal internal schema. Which default follows the Principle of Minimum Intervention?

Multiple Choice

A. Translate every field so the gateway owns all application semantics B. Carry the payload as received when modification is not required C. Reject all encapsulation because it necessarily violates end-to-end design

**Explanation:** A: universal translation increases the scope and coupling of the intermediary. B: RFC 3439 recommends transporting payload without modification where possible. C: the RFC permits data-plane interworking and encapsulation, especially at the edge.

Q8: Two unlike networks must interoperate. Which boundary is preferred by RFC 3439?

Multiple Choice

A. Merge both control planes into one universal state machine in the core B. Replicate every application's policy into every transit node C. Restrict interworking mainly to data-plane encapsulation at the edge

**Explanation:** A: control-plane translation creates a large opportunity for failure. B: it tightly couples transit infrastructure to application state. C: the RFC recommends edge-based data-plane interworking and minimizing control-plane interworking.

Q9: Which reliability strategy is consistent with RFC 3439's “myth of five nines” discussion?

Multiple Choice

A. Design for system reliability without assuming every complex component must individually reach five nines B. Add mechanisms until every component independently claims perfect availability C. Ignore component failure because packet networks are automatically reliable

**Explanation:** A: system-level resilience can use comparatively unreliable components, and attempts to perfect every component may add human and process complexity. B: that can feed the robustness/complexity spiral. C: the argument changes how reliability is composed; it does not dismiss failures.

Q10: An Agent-to-Agent request must traverse seven mandatory brokers, translators, and policy engines. What is the strongest RFC 3439 review question?

Multiple Choice

The service works in a small test environment; the concern is whether the mandatory path remains stable and operable at scale.

A. Can each intermediary expose more optional features? B. Which components can be removed or made optional without losing required behavior? C. Can all intermediaries share one larger control-plane database?

**Explanation:** A: feature count is not evidence of deployability. B: the Architectural Component Proportionality Law ties complexity to component count and recommends minimizing components in the service-delivery path. C: a shared control plane can increase coupling and failure scope rather than reduce the path's obligations.

RFC 3439 Quiz

Simplicity in large-scale systems

References (URLs)

Q1: Which design choice best follows RFC 3439's Simplicity Principle?

Q2: What does the end-to-end argument imply about state in the Internet core?

Q3: A configuration change for one tenant triggers a global recomputation across every gateway. Which principle identifies the concern?

Q4: Two individually safe recovery mechanisms repeatedly trigger each other and create a cascading outage. Which analysis fits best?

Q5: Why can a smart-edge, simple-core architecture reduce the cost of adding a new service?

Q6: How should “Layering Considered Harmful” be read in RFC 3439?

Q7: A gateway can encapsulate a payload unchanged or translate every field into a universal internal schema. Which default follows the Principle of Minimum Intervention?

Q8: Two unlike networks must interoperate. Which boundary is preferred by RFC 3439?

Q9: Which reliability strategy is consistent with RFC 3439's “myth of five nines” discussion?

Q10: An Agent-to-Agent request must traverse seven mandatory brokers, translators, and policy engines. What is the strongest RFC 3439 review question?