What is the purpose of health check response caching and how does frequent health check polling itself impact application performance?

Health checks, if not designed carefully, can themselves become a performance concern under frequent polling (e.g., a Kubernetes liveness probe hitting the endpoint every few seconds) -- especially if a health check performs an expensive operation like a full database query on every invocation, so it's common to cache health check results for a short period or design checks to be lightweight (a simple connectivity ping rather than a full query) to avoid the diagnostic mechanism itself becoming a load concern.

builder.Services.AddHealthChecks()
    .AddCheck<DatabaseHealthCheck>("database", tags: new[] { "ready" });

// A well-designed health check avoids expensive operations:
public class DatabaseHealthCheck : IHealthCheck {
    public async Task<HealthCheckResult> CheckHealthAsync(HealthCheckContext context, CancellationToken ct) {
        await _db.Database.CanConnectAsync(ct);  // cheap connectivity check, not a full query
        return HealthCheckResult.Healthy();
    }
}