Distributed Systems & System Design
Find your next engineering question
Start with a subject, or narrow the library to the level and problem you want to practise.
26 questions
Clear filtersNetworking & Protocols · Staff
The server finished in 30 ms. Where did the other 370 ms go?Separating end-to-end client latency from instrumented server work, including requests that never produce an application span.
Networking & Protocols · Staff
We added six servers. Why did none get traffic?The difference between assigning transport connections and routing requests when HTTP/2 channels are long-lived.
Networking & Protocols · Staff
The link is 10 Gbps. Why is one file moving at 200 Mbps?Calculate BDP and distinguish receiver-window, congestion-window and application pacing limits on a long-RTT TCP flow.
Networking & Protocols · Staff
Only one percent of packets are lost. Why did API p99 explode?Diagnose how a small segment-loss rate can block ordered bytes, trigger retransmission timers and alter tail latency and congestion state.
Networking & Protocols · Staff
The server is idle. Why are new connections timing out?Differentiate incomplete SYN requests from completed connections awaiting accept, measure burst capacity and repair slow accept loops.
Networking & Protocols · Staff
All services are healthy. Who used the last outbound port?Calculate destination-specific NAT mapping occupancy and distinguish gateway port exhaustion from local ephemeral ports and bandwidth limits.
Networking & Protocols · Staff
Tiny requests work. Why do large responses hang over the VPN?Diagnose size-dependent VPN loss through tunnel overhead, MTU and MSS arithmetic, ICMP feedback and black-hole-safe probing.
Networking & Protocols · Principal
The load balancer says healthy. Why do 30 percent of requests fail?Detect partial backend failures invisible to liveness probes using client-outcome dimensions and safely bounded outlier ejection.
Networking & Protocols · Senior
The API is healthy. Why does the first request wait on DNS?Trace stub, recursive and authoritative lookups, including TTLs, CNAME chains and negative caching.
Networking & Protocols · Staff
We changed the DNS record. Why are 20% of users still on the old IP?Design a safe DNS-based migration despite existing resolver caches, HTTP connections, stale answers and rollback limitations.
Networking & Protocols · Senior
The server has a valid certificate. Why does the browser refuse HTTPS?Explain SAN hostname validation, intermediate and root trust chains, SNI and client-specific certificate policy.
Networking & Protocols · Staff
TLS 1.3 is faster. Should a payment API accept 0-RTT?Compare full TLS 1.2 and TLS 1.3 setup, PSK resumption, early-data replay and safe application policy.
Networking & Protocols · Staff
HTTP/2 multiplexes requests. Why can one lost packet stall them all?Distinguish HTTP application serialization, TCP stream-level head-of-line blocking and independent QUIC stream delivery.
Networking & Protocols · Staff
The phone switched from Wi-Fi to 5G. Why did the request survive?Explain address-independent QUIC connection IDs, path validation, anti-amplification and backend affinity across network changes.
Networking & Protocols · Senior
The database is healthy. Why can't the pod reach it?Trace DNS, pod source address, private routes, security policy and the return path without assuming NAT.
Networking & Protocols · Principal
We connected four regions. Why did one routing change affect all of them?Choose peering, transit gateways and service endpoints from a traffic matrix while limiting failures and privileges.
Networking & Protocols · Staff
Every server is healthy. Why did the internet lose the route to them?Separate BGP control-plane route withdrawals from healthy application servers, and diagnose internet reachability from outside.
Networking & Protocols · Staff
The IP address stayed the same. Why did the connection move to another city?Explain shared anycast IPs, routing catchments, and why stateful TCP connections can fail when the chosen site changes.
Networking & Protocols · Staff
We enabled IPv6. Why did only some customers lose the API?Diagnose failures after adding IPv6 AAAA records, including client fallback variation, NAT64, MTU and policy parity.
Networking & Protocols · Staff
We upgraded the inter-region link. Why is checkout still taking a second?Quantify the critical path from dependent WAN calls and move work to the data without weakening consistency.
Networking & Protocols · Senior
The browser uploaded the photo. Why is it not yet part of the account?Move upload bandwidth off the API without delegating final publication or authorization to the browser.
Networking & Protocols · Staff
A photo was uploaded once. Why did the image service generate 40,000 files?Bound transformation cardinality and secure decoding while using source and recipe identity in derivative cache keys.
Networking & Protocols · Staff
We put a CDN in front of storage. Why is the origin still receiving thousands of requests?Explain independent edge misses and how hierarchical caching reduces eligible repeated origin fetches.
Networking & Protocols · Staff
The origin is down. Should the CDN show yesterday's page or return an error?Derive cache freshness and extension windows and choose a business-safe permitted staleness contract.
Networking & Protocols · Principal
The edge lost its control plane. Why did it send users to a retired origin?Define policy-specific offline behavior using versioned, validated configuration snapshots and revocation limits.
Networking & Protocols · Staff
One software release is 2 GB. How do you deliver it to 100 million devices?Design signed content delivery, partial resume, rollout waves and post-install health for massive downloads.