Search the practice set
126 questions, 454 terms and 400 topics in 20 areas.
52 results for “Load Testing”
Load Testing
Driving a system with realistic traffic at a target volume to verify it meets its performance targets before real users do.
Cognitive Load
The total amount a team must hold in its head to work effectively, and a real constraint on how many services or domains one team can own.
Contract Testing
Verifying that a provider satisfies the expectations each of its consumers actually relies on, without running all the services together.
Cross-Zone Load Balancing
Whether a load balancer node distributes traffic to targets in all zones or only its own — a setting that affects both balance and data transfer cost.
Layer 4 vs Layer 7 Load Balancing
Balancing on connection metadata (IP and port) versus on the content of the request (path, host, headers).
Load Shedding
Deliberately rejecting a portion of incoming work during overload so that the remainder can be served correctly.
Admission Control
Deciding at the edge whether to accept a request at all, based on current capacity, before any work is done on it.
Autoscaling
Adding and removing capacity automatically in response to a demand signal, to track load without paying for peak all the time.
Backpressure
A mechanism by which a component under load tells its callers to slow down, rather than accepting work it cannot complete.
Booking.com's Experimentation Platform
Booking.com runs over a thousand concurrent experiments and treats the ability to test any change safely as a platform capability rather than a product feature.
Brownout
Deliberately reducing the quality or completeness of every response under load, rather than serving some requests fully and rejecting others.
Cache Penetration
Repeated lookups for keys that do not exist, which miss the cache every time by definition and pass straight through to the store.
Cache Stampede
Many concurrent requests missing on the same expired key and all recomputing it simultaneously, converting one expiry into a load spike.
Chaos Engineering
Deliberately injecting failure into a system to discover, before an incident does, which of your resilience assumptions are false.
Client-Side Discovery
The caller queries the registry itself and chooses an instance, rather than sending to a stable address that something else resolves.
Connection Draining
The period during which a load balancer stops sending new requests to an instance while allowing in-flight ones to complete before it is removed.
DNS
The distributed directory that resolves names to addresses, and a surprisingly load-bearing part of most architectures.
Fan-Out
One incoming request causing many outgoing ones, which multiplies both load and tail latency.
Fitness Function
An automated check that an architectural characteristic still holds, run continuously rather than reviewed occasionally.
Game Day
A scheduled exercise in which a failure is deliberately introduced and the team responds as though it were real, to test the system and the response together.
Global Traffic Management
The layer that decides which region a given user reaches, using DNS, anycast or an edge network, and that performs regional failover.
LLM Evaluation
A repeatable measurement of whether an AI system's outputs are good enough, on cases that reflect the actual task.
Log-Based CDC
Capturing changes by reading the database's own write-ahead log, which sees every change with no load on the source and no application involvement.
Priority Queueing
Classifying requests by business importance so that overload sheds the least valuable work first rather than an arbitrary slice.
Quality Attribute Scenario
A structured, testable statement of a non-functional requirement: source, stimulus, environment, artefact, response, response measure.
Read Replica
A copy of a database that receives changes from the primary and serves read-only queries, spreading read load.
Redundancy
Having more instances of a component than the load requires, so that failures can be absorbed without loss of service.
Restore Drill
A scheduled, timed exercise of restoring from backup into a clean environment — the only thing that converts a backup from a hope into a control.
Scalability
The ability to handle growing load by adding resources, ideally with cost rising no faster than the load.
Test Pyramid
A distribution of tests weighted towards many fast unit tests, fewer integration tests, and very few slow end-to-end tests.
Thundering Herd
A large number of clients acting simultaneously because they were synchronised by a shared event, producing a spike that the steady-state design never sized for.
A serverless API works in testing and fails under load with connection errors. The database is at 5% CPU. Explain and fix.
The mechanism Serverless functions scale by creating independent execution environments , each with its own process and its own connection pool. Two hundred con
Peak trading day is six weeks away and expected to be four times normal traffic. What do you do in those six weeks?
What the interviewer is testing Whether you can run a readiness programme rather than just "add servers", and whether you know what fails at peak that does not
Your system handles 1,000 requests per second today. Marketing says a campaign will bring 10,000 next month. What breaks first, and how do you find out?
What the interviewer is testing Whether you reason about bottlenecks systematically or start adding servers. The shape of the answer Scaling is not uniform. Som
Your cluster fails over spuriously under load, but a real leader failure takes 45 seconds to detect. How do you resolve the tension?
The tension Detection time is set by the heartbeat interval times the failure threshold. Shorten it and you detect real failures faster and mistake slow but ali
Your service will exceed capacity by 30% during a known peak. Do you shed load or brown out, and how do you decide what goes first?
The decision Brown out first; shed only if that is not enough. They are complementary rather than alternatives, and brownout is strictly less harmful when it is
Every deployment produces a small spike of 502s that the team has learned to ignore. Fix it.
The cause In flight requests are being severed because the instance terminates before the load balancer has stopped sending to it, or before existing requests c
One customer's batch job saturates a shared service and degrades everyone. Rate limiting them fixes it, until the next customer does the same. What is the structural answer?
Why per customer rate limits keep failing A static limit is set from what that customer was doing, not from what the service can serve. It is reactive — you dis
TSB's 2018 core banking migration moved 1.9 million customers in a single weekend and failed publicly. What would you have required before approving that cutover?
The case, as publicly reported In April 2018 TSB migrated from a platform rented from Lloyds Banking Group to Proteo4UK, built by its parent Banco Sabadell. The
You need to ship a rewrite of the pricing engine. Same inputs, same expected outputs, completely new implementation. How do you release it?
What the interviewer is testing Whether you know that release strategy is part of architecture, and whether you reach for verification techniques beyond "test i
Load Testing
Realistic data, realistic mix, and a ramp rather than a step.
Contract Testing
Verifying what consumers actually rely on, without a shared environment.
DR Testing
Restore drills, timed against the stated RTO, into a clean environment.
Load Balancing
Distributing traffic, health checking, and removing failures from rotation.
Load Shedding
Rejecting some work deliberately so the rest can be served correctly.
Resilience Testing
Exercising retries, breakers and fallbacks that are otherwise never run.
Soak Testing
Long runs that surface leaks and slow degradation.
Stress Testing
Pushing past target to learn what breaks first and how it fails.
Testing Strategies
The pyramid, and the contract tests distributed systems add to it.
Capacity Modelling
Arithmetic before load tests, and headroom for failure as well as peak.
Game Days
Testing the response — runbooks, access, comms — not only the system.
Platform Teams
Reducing other teams' cognitive load, measured by adoption.