Cloud concept · Updated

Noisy Neighbor - A noisy neighbor is a workload or tenant whose use of shared resources degrades the performance or availability of other workloads or tenants.

In short

The noisy neighbor problem occurs when competing workloads share constrained resources such as CPU, memory, storage throughput, database connections, or service quotas. It matters in both provider infrastructure and application-level multi-tenancy because one tenant’s burst, expensive query, or background job can disrupt others unless resource consumption is controlled.

Cloud Concepts Hub →

6 key aspects of Noisy Neighbor

Noisy neighbor protection combines resource-aware isolation, tenant-level fairness, observability, and controlled overload handling.

01
Contention across shared resources

Interference can occur in compute, storage, networks, caches, databases, worker pools, and shared quotas. Diagnose the constrained resource rather than assuming every latency spike requires more CPU.

02
Isolation matched to the boundary

Pooled resources improve utilization, while siloed resources reduce interference at the boundaries they separate. Mixed architectures and deployment cells can limit impact without dedicating every component to every tenant.

03
Per-tenant fairness and admission control

Use tenant-aware rate limits, concurrency budgets, fair scheduling, and bounded queues to prevent one tenant from monopolizing capacity. Account for operation cost as well as request count.

04
Performance isolation differs from data isolation

Verified tenant context, row-level security, and tenant-specific encryption help enforce identity and data boundaries. They do not by themselves prevent authorized work from exhausting shared resources.

05
Tenant-level observability and testing

Correlate tenant latency and errors with resource saturation, queue depth, and throttling. Run controlled overload tests to verify that one tenant’s increased demand does not violate other tenants’ service objectives.

06
Capacity and overload management

Combine scaling and capacity headroom with quotas, timeouts, retry budgets, and load shedding. Define which work is rejected or deferred when demand exceeds safe capacity.

Cloud services for Noisy Neighbor

API Gateway can reduce incoming request pressure before it reaches workloads on EKS, EC2, or Lambda. Compute-level controls then constrain resource consumption, while application-level tenant budgets protect shared databases on RDS. These layers must be coordinated: increasing worker capacity can otherwise overwhelm database connections or I/O. Verified tenant context should accompany requests and background jobs so admission controls and telemetry attribute work correctly. When pooled controls cannot meet isolation objectives, selected tenants can move to separate compute resources and database instances, with shared downstream limits reviewed explicitly.

Amazon RDSAmazon RDS

Tenants sharing an RDS database instance can compete for CPU, memory, I/O, locks, and connections. Application connection budgets, engine-specific workload controls, and separate database instances for demanding tenants can reduce interference; separate schemas or row-level security alone do not reserve capacity.

Amazon EC2Amazon EC2

EC2 provides instance-level compute boundaries, while Dedicated Instances or Dedicated Hosts remove physical host sharing with other AWS accounts. These options do not eliminate contention among workloads inside an instance or isolate shared downstream dependencies.

AWS LambdaAWS Lambda

Lambda functions can compete for an account’s available regional concurrency. Reserved concurrency reserves capacity for a function and caps its concurrency, but tenants sharing that function still need application-level fairness controls.

Amazon API GatewayAmazon API Gateway

API Gateway throttling helps reduce backend overload, and REST API usage plans support per-client throttling using API keys. Usage-plan throttles and quotas are best-effort rather than hard ceilings; API keys are not an authentication mechanism, so strict tenant protection requires verified identity and backend enforcement.

What is Noisy Neighbor

3 min read

A noisy neighbor is a workload that consumes enough shared capacity to interfere with another workload. The interference may appear as higher latency, lower throughput, throttling, timeouts, or failures. It does not require malicious behavior: a legitimate bulk import, cache rebuild, or reporting query can create the problem. Neighbors can be separate customers on shared infrastructure, containers on the same node, applications using a common database, or tenants inside one application.

The first design task is to identify the actual contention boundary. CPU limits do not protect a shared database connection pool, and separate application instances do not isolate tenants that still use the same storage throughput or downstream API quota. Investigate CPU scheduling, memory pressure, disk latency, network bandwidth, locks, connection counts, queue depth, and account-level service limits. Track request latency and resource consumption by tenant where feasible; fleet-wide averages can hide severe degradation for a small group of tenants.

Multi-tenant architectures commonly use pooled, siloed, or mixed isolation patterns. Pooling places tenants on shared resources and usually improves utilization and reduces cost, but requires explicit fairness controls. Siloing assigns dedicated resources to a tenant and provides stronger isolation at the boundaries actually separated, at the cost of additional capacity and operational overhead. A mixed design might run most tenants in shared deployment cells while moving high-volume tenants to dedicated compute and databases. Each cell limits the number of tenants exposed to one overloaded resource.

Apply controls close to the resource being protected. Use per-tenant admission limits, bounded concurrency, database connection budgets, query timeouts, and fair scheduling. For example, a tenant submitting thousands of image-processing jobs should not occupy every worker while other tenants wait. Separate queues combined with tenant-aware worker allocation can preserve progress for smaller tenants; separate queues alone do not guarantee fairness if all workers still drain one queue first. Bound retries and use backoff with jitter to avoid amplifying overload.

Identity and data isolation support these controls but do not replace them. A trusted tenant identifier derived from verified identity claims, such as claims issued through Amazon Cognito, can drive authorization, metering, and tenant-specific admission decisions. Do not trust an arbitrary tenant identifier supplied by a client. Database row-level security restricts which rows a tenant may access, but does not stop an authorized query from consuming shared CPU or I/O. Tenant-specific encryption keys separate cryptographic access and key management; they do not reserve processing capacity or prevent resource contention.

Scaling adds capacity but is not a complete solution. Autoscaling reacts after demand changes, may encounter quotas, and may leave a shared database or external dependency as the bottleneck. Request-rate limits also need to account for request cost: ten expensive reports can consume more capacity than thousands of simple reads. Combine capacity headroom with workload-aware limits and explicit overload behavior, such as rejecting excess requests or deferring batch work.

Validate isolation by deliberately increasing one tenant’s load while measuring other tenants’ latency, errors, and throughput. Define acceptable cross-tenant impact and test both sustained demand and bursts. Dedicated hosts, separate namespaces, and separate accounts each isolate different concerns; none should be treated as a blanket guarantee that every downstream resource is independent. Effective protection comes from matching isolation and fairness mechanisms to every important shared bottleneck.