Search the practice set
126 questions, 454 terms and 400 topics in 20 areas.
60 results for “Horizontal vs Vertical Scaling”
Horizontal vs Vertical Scaling
Adding more machines versus making one machine bigger — and the fact that vertical is underrated for stateful tiers.
Provisioned vs Serverless Capacity
Paying for a fixed database size continuously, versus paying for capacity consumed with automatic scaling — a crossover decision driven by duty cycle.
Figma's Postgres Sharding
Figma delayed sharding for years using replicas and vertical partitioning, then sharded Postgres horizontally without downtime using logical shards and a proxy layer.
Sharding
Splitting one dataset across multiple independent databases by a partition key, so that each holds a disjoint subset.
Active-Active vs Active-Passive
Whether all regions serve traffic simultaneously, or one serves while another waits to take over — a choice about which failure mode you would rather have.
Build vs Buy
The choice between developing a capability in-house and acquiring it, decided on differentiation and total cost rather than on feature lists.
Containment vs Eradication
Stopping an attacker's ongoing access versus removing their foothold entirely — sequential phases with different urgency and different risks of doing them wrong.
Cost vs Reliability Trade-off
The non-linear relationship between availability and spend, which makes each additional nine roughly an order of magnitude more expensive.
Delivery vs Maintainability
Choosing where to take deliberate shortcuts, based on which kinds of debt are cheap to repay and which compound.
Durability vs Availability
Two different storage guarantees — whether data survives, and whether it can be reached right now — routinely conflated because both are quoted in nines.
ETL vs ELT
Whether data is transformed before loading into the target or after it, which decides where the compute happens and how much raw history you keep.
Fail-Fast vs Fail-Safe
Whether a component should stop immediately on detecting a problem, or continue in a degraded but safe mode — a choice that depends entirely on which outcome is worse.
Instagram's Early Scaling
Instagram reached tens of millions of users on Django and PostgreSQL with a handful of engineers, by deliberately choosing boring technology and doing the simple thing first.
Layer 4 vs Layer 7 Load Balancing
Balancing on connection metadata (IP and port) versus on the content of the request (path, host, headers).
Managed vs Self-Managed
Trading control, portability and unit cost against the operational burden of running the thing yourself.
Monolith vs Microservices
A trade of deployment independence against distributed-systems complexity, decided by team topology far more often than by technology.
OLTP vs OLAP
Two workload shapes with opposite requirements — many small indexed transactions versus few large scans and aggregations — which is why they belong in different stores.
Operational vs Analytical Store
The separation between the store serving the application's transactions and the one serving reporting and analysis, and the mechanism connecting them.
Re-architect vs Rebuild
Restructuring an existing system incrementally versus writing a replacement from scratch — and the strong evidence that incremental wins.
SQL vs NoSQL
A choice driven by access patterns, consistency requirements and query flexibility — not by data volume, which is the reason usually given.
Security vs Usability
A trade-off that is usually resolved by varying the control with the value of the action, rather than by choosing a uniform level of friction.
Strong vs Eventual Consistency
A per-operation decision, not a per-system one: whether this specific read must reflect every completed write.
Synchronous vs Asynchronous Communication
Whether the caller waits for the callee's answer — decided by whether the caller's outcome depends on it, not by latency or taste.
Synchronous vs Asynchronous Replication
Whether a write is acknowledged only after a replica has it, trading write latency against the amount of data a failure can lose.
Target Tracking Scaling
An autoscaling policy that adds or removes capacity to hold a chosen metric near a target value, like a thermostat, rather than reacting to threshold breaches.
Zonal vs Regional Services
Whether a cloud resource lives in one availability zone or is inherently spread across several — a property that determines what a zone failure takes with it.
Autoscaling
Adding and removing capacity automatically in response to a demand signal, to track load without paying for peak all the time.
CQRS
Separating the model used to change state from the model used to read it, so each can be optimised independently.
Capacity Planning
Deciding in advance how much capacity will be needed, given growth, seasonality and failure scenarios, and ensuring it can be there in time.
Competing Consumers
Multiple identical consumers reading from one queue, so throughput scales with consumer count and work is distributed automatically.
Concurrency
The number of operations in progress at once — distinct from parallelism, which is how many are literally executing simultaneously.
Consumer Group
A set of consumers that cooperatively read one stream, with each partition assigned to exactly one member, so the group collectively processes every message once.
Dropbox's Move Off S3
Dropbox moved the majority of its file storage off Amazon S3 onto custom infrastructure, reporting savings that its S-1 filing put at roughly $75 million over two years.
Feature Parity Trap
The expectation that a replacement system must match every behaviour of the old one before it can be adopted, which is what makes rewrites never finish.
Hot Partition
One partition receiving disproportionate traffic, so the system saturates at a fraction of its aggregate capacity.
Layered Architecture
Organising code into horizontal layers — presentation, application, domain, data — where each layer may only call the one beneath it.
A multi-tenant SaaS product has outgrown one database. You must shard. How do you choose the partition key, and what makes this decision so expensive to get wrong?
What the interviewer is testing Whether you exhaust cheaper options first, and whether you understand that a shard key is close to irreversible. First: do not s
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
Design a URL shortener handling 100 million new links per month and 10 billion redirects. Where is the real difficulty?
What the interviewer is testing The classic warm up. What is being assessed is not whether you can shorten a URL — it is whether you do capacity arithmetic befo
Horizontal vs Vertical Scaling
Scale out for stateless, scale up first for stateful.
Build vs Buy
Differentiation versus table stakes, priced over five years.
Build vs Buy
Differentiation, five-year TCO, and the exit cost of each option.
Centralised vs Distributed
Shared platform leverage against team autonomy.
Cost vs Reliability
Each nine costing an order of magnitude, and pricing the failure instead.
Delivery vs Maintainability
Fast in the cheap places, careful in the expensive ones.
Functional vs Non-Functional
Behaviour versus quality of behaviour, and why only the second constrains structure.
Layer 4 vs Layer 7
Connection-level versus request-level balancing, and what each unlocks.
Managed vs Self-Managed
Trading control and unit cost against operational attention.
Monolith vs Microservices
A team-topology decision far more often than a technology one.
Orchestration vs Choreography
A coordinator that knows the flow, or services that react to events.
Performance vs Cost
Buying latency, and knowing what the last millisecond is worth.
Rebuild vs Re-architect
Why greenfield replacement fails, and the narrow cases where it does not.
Reliability vs Complexity
Mechanisms that add availability and add failure modes.
SQL vs NoSQL
Decided by access patterns and query flexibility, not by data volume.
Security vs Usability
Varying control by the value of the action rather than uniformly.
Serverless vs Containers
Spiky and event-driven versus sustained throughput.
Single vs Multi-Region
Driven by RTO, RPO and residency rather than by ambition.
Strong vs Eventual Consistency
A per-operation decision, resolved by what a stale read would cost.
Sync vs Async
Whether the caller's outcome depends on the callee's response.
Competing Consumers
Scaling throughput with instances, at the cost of ordering.