Loading
Pick a focused question that fits your time, stack, and interview goal.
25 results across 1 active filter
Page 1 of 2
Explains physical connection reuse, pool partitioning, short ownership, exhaustion causes, diagnostics, and why increasing the pool is not the first fix.
Compares server-controlled output caching with HTTP cache semantics and defines safe keys, variation, invalidation, identity, and multi-instance behavior.
Builds a bounded dynamic query that returns a read model, applies only supplied predicates, and uses deterministic ordering.
Compares caching response-shaped read models with caching domain entities or raw database objects.
Explains how TTL choices balance freshness, load reduction, user expectations, and operational safety.
Implements single-flight request coalescing so one load runs per key while unrelated keys remain concurrent.
Compares cache-aside, read-through, and write-through by read and write ownership, freshness, failure behavior, invalidation, fallback, stampede control, and operational complexity.
Explains when local process memory is enough and when shared Redis-style caching is worth the extra dependency.
Collects a bounded EventPipe CPU trace and reads stack evidence without confusing hot methods, blocked time, native work, or sampling artifacts.
Explains when stale cache reads are acceptable, when they are dangerous, and how to make freshness visible.
Uses a bounded entity graph for a to-one association and proves the repair with Hibernate query statistics.
Builds stable Micrometer dimensions that answer operational questions without unbounded time series and cost.
Moves from user-visible latency to route, dependency, database, executor, JVM, and change evidence before selecting a fix.
Uses Hikari metrics, thread evidence, database activity, transaction scope, and query behavior to separate leaks from slow or excessive work.
Separates server request capacity from application executor capacity and finds blocking, queueing, rejection, or downstream causes.
Sizes per-instance database concurrency from workload and database capacity instead of equating more connections with more throughput.
Explains how IHttpClientFactory separates short-lived clients from pooled handlers and how concurrency, DNS, resilience, and DI lifetimes still require judgment.
Implements descending cursor pagination with a unique tie-breaker, bounded projection query, and next-cursor detection.
Explains delete-on-write, versioned keys, event-driven invalidation, and the race conditions around cache freshness.
Combines ThreadPool counters, repeated stack snapshots, and wait-event traces to distinguish blocked workers from CPU or dependency pressure.
Fetches multiple URLs with bounded concurrency, per-request and batch deadlines, cancellation, ordered results, and executor cleanup.
Explains cached value serialization, schema versioning, deploy compatibility, and safe handling of old payloads.
Designs cancellation-aware response streaming with bounded memory, correct resource ownership, and practical download behavior.
Uses live runtime and application counters to classify CPU, memory, GC, exception, request, and ThreadPool symptoms before deeper collection.