AWS vs Azure
AWS and Azure are the two most prominent public cloud providers for enterprise infrastructure, platform services, and managed application deployment. Both offer broad portfolios across compute, storage, networking, databases, security, analytics, and AI, but they differ in ecosystem alignment, operational models, and common enterprise use cases.2
AWS is widely regarded as the broadest and most mature public cloud platform, with extensive service depth and a long lead in cloud-native tooling. Azure, by contrast, is especially strong in organizations already invested in Microsoft technologies such as Windows Server, Active Directory, Microsoft 365, and SQL Server, and it is often preferred for hybrid cloud strategies.2
A practical way to compare the two is to evaluate them across six dimensions:
- Service breadth and maturity
- Enterprise and Microsoft ecosystem integration
- Hybrid and multicloud capabilities
- Identity, governance, and security operations
- Pricing and licensing behavior
- Global infrastructure and workload fit3
In market terms, AWS remains the largest provider, while Azure is the closest large-scale competitor and is often described as growing rapidly in enterprise segments.2 However, market share alone should not determine platform selection; architecture requirements, internal expertise, governance standards, and application dependencies are usually more important than headline size.2
Footnotes
-
Azure vs AWS Pricing: Which Cloud Provider is Cheaper? - Comparative discussion of pricing models, storage examples, and market-share context. ↩ ↩2 ↩3
-
Azure vs AWS: Which Cloud Platform is Best in 2025? - Broad comparison of ecosystem fit, hybrid posture, performance, pricing, and enterprise use cases. ↩ ↩2 ↩3 ↩4
-
AWS vs Azure vs Google: Cloud Services Comparison - Overview of AWS and Azure strengths, service categories, and positioning in the cloud market. ↩ ↩2
-
Microsoft Azure vs AWS: A Comprehensive Comparison - Detailed comparison of compute, identity, security, performance, and hybrid characteristics. ↩ ↩2
-
What's the Difference Between AWS vs. Azure vs. Google Cloud? - High-level market and cloud-platform comparison useful for contextual positioning. ↩
Cloud Providers Compared: A Comprehensive Guide to AWS, Azure, and GCP
Core Decision Principle
The best choice is rarely the one with the most services overall; it is the one that minimizes architectural friction, governance risk, and long-term operating cost for your workloads.2
Footnotes
-
Azure vs AWS: Which Cloud Platform is Best in 2025? - Broad comparison of ecosystem fit, hybrid posture, performance, pricing, and enterprise use cases. ↩
-
Microsoft Azure vs AWS: A Comprehensive Comparison - Detailed comparison of compute, identity, security, performance, and hybrid characteristics. ↩
1. Platform philosophy and ecosystem position
AWS emphasizes breadth, modularity, and cloud-native service composition. Its portfolio includes mature offerings for virtual machines, object storage, containers, serverless, analytics, and infrastructure automation, making it attractive for teams building highly customized distributed systems.2
Azure emphasizes enterprise integration and operational continuity. It maps naturally to Microsoft-centric environments, especially where organizations already depend on Entra ID, Windows administration, Microsoft security tooling, or existing licensing agreements.2
A simplified comparison is shown below:
For many organizations, the decision is not “which cloud is objectively better?” but “which cloud reduces integration overhead?” A company running .NET applications, Active Directory, and SQL Server may find Azure more operationally coherent, while a team building event-driven microservices across many managed components may prefer AWS's service breadth and architectural patterns.2
Footnotes
-
AWS vs Azure vs Google: Cloud Services Comparison - Overview of AWS and Azure strengths, service categories, and positioning in the cloud market. ↩ ↩2
-
Compare AWS and Azure compute services - Azure Architecture Center - Official Microsoft mapping of AWS and Azure compute services and related infrastructure concepts. ↩
-
Azure vs AWS: Which Cloud Platform is Best in 2025? - Broad comparison of ecosystem fit, hybrid posture, performance, pricing, and enterprise use cases. ↩ ↩2
-
Microsoft Azure vs AWS: A Comprehensive Comparison - Detailed comparison of compute, identity, security, performance, and hybrid characteristics. ↩ ↩2
AWS tends to appeal to teams that want granular service choice, mature cloud-native patterns, and extensive managed infrastructure options such as EC2, S3, Lambda, EKS, and DynamoDB.2
Footnotes
-
AWS vs Azure vs Google: Cloud Services Comparison - Overview of AWS and Azure strengths, service categories, and positioning in the cloud market. ↩
-
Compare AWS and Azure compute services - Azure Architecture Center - Official Microsoft mapping of AWS and Azure compute services and related infrastructure concepts. ↩
2. Compute services: EC2 vs Azure Virtual Machines, Lambda vs Functions, EKS vs AKS
At the infrastructure level, AWS EC2 and Azure Virtual Machines provide broadly comparable IaaS capabilities. Both bill on a per-second basis for many VM scenarios and offer multiple instance or size families optimized for general purpose, memory, compute, storage, or GPU-intensive workloads.
Microsoft's own architecture guidance maps AWS and Azure compute services directly: EC2 corresponds to Azure Virtual Machines, EBS-style attached disk patterns map to Azure data disks, EC2 instance store maps to Azure temporary storage, and EFS maps to Azure Files.
For application runtime models:
- AWS Lambda corresponds broadly to Azure Functions for serverless event-driven execution.
- Amazon EKS corresponds to Azure Kubernetes Service (AKS) for managed Kubernetes orchestration.2
- AWS Auto Scaling parallels Azure Virtual Machine Scale Sets for elastic capacity control.
Two practical differences often matter:
- AWS is frequently described as offering a broader selection of instance families and deeper cloud-native infrastructure specialization.
- Azure is often operationally advantageous when VM identity, Windows administration, and enterprise access control must align tightly with Microsoft identity systems.2
Footnotes
-
Compare AWS and Azure compute services - Azure Architecture Center - Official Microsoft mapping of AWS and Azure compute services and related infrastructure concepts. ↩ ↩2 ↩3 ↩4
-
AWS vs Azure vs Google: Cloud Services Comparison - Overview of AWS and Azure strengths, service categories, and positioning in the cloud market. ↩ ↩2 ↩3
-
Azure vs AWS: Which Cloud Platform is Best in 2025? - Broad comparison of ecosystem fit, hybrid posture, performance, pricing, and enterprise use cases. ↩
-
Microsoft Azure vs AWS: A Comprehensive Comparison - Detailed comparison of compute, identity, security, performance, and hybrid characteristics. ↩
How to choose between AWS and Azure for compute workloads
- 1Step 1
Classify the application as VM-centric, containerized, batch, HPC, GPU, or event-driven serverless. This determines whether services like EC2, Azure Virtual Machines, EKS, AKS, Lambda, or Functions are the primary comparison points.2
Footnotes
-
AWS vs Azure vs Google: Cloud Services Comparison - Overview of AWS and Azure strengths, service categories, and positioning in the cloud market. ↩
-
Compare AWS and Azure compute services - Azure Architecture Center - Official Microsoft mapping of AWS and Azure compute services and related infrastructure concepts. ↩
-
- 2Step 2
Assess whether the workload depends on Windows Server, Active Directory, Microsoft licensing, or existing enterprise tooling. Strong Microsoft coupling often favors Azure, while highly cloud-native service composition often favors AWS.2
Footnotes
-
Azure vs AWS: Which Cloud Platform is Best in 2025? - Broad comparison of ecosystem fit, hybrid posture, performance, pricing, and enterprise use cases. ↩
-
Microsoft Azure vs AWS: A Comprehensive Comparison - Detailed comparison of compute, identity, security, performance, and hybrid characteristics. ↩
-
- 3Step 3
Compare autoscaling behavior, zone design, and operational controls. AWS uses Auto Scaling and multi-AZ deployment patterns, while Azure commonly uses Virtual Machine Scale Sets and Availability Sets or Zones.2
Footnotes
-
Microsoft Azure vs AWS: A Comprehensive Comparison - Detailed comparison of compute, identity, security, performance, and hybrid characteristics. ↩
-
Compare AWS and Azure compute services - Azure Architecture Center - Official Microsoft mapping of AWS and Azure compute services and related infrastructure concepts. ↩
-
- 4Step 4
Model on-demand, reserved, and spot usage patterns rather than comparing list prices only. The cheapest platform for a workload depends heavily on runtime duration, storage access, and software licensing.2
Footnotes
-
Azure vs AWS Pricing: Which Cloud Provider is Cheaper? - Comparative discussion of pricing models, storage examples, and market-share context. ↩
-
What's the Difference Between AWS vs. Azure vs. Google Cloud? - High-level market and cloud-platform comparison useful for contextual positioning. ↩
-
- 5Step 5
Pilot deployment, monitoring, IAM integration, backup, and incident response should be validated before a platform decision is finalized.2
Footnotes
-
Azure vs AWS: Which Cloud Platform is Best in 2025? - Broad comparison of ecosystem fit, hybrid posture, performance, pricing, and enterprise use cases. ↩
-
AWS vs Azure vs Google: Cloud Services Comparison - Overview of AWS and Azure strengths, service categories, and positioning in the cloud market. ↩
-
3. Storage comparison: S3 vs Blob Storage, EBS vs Managed Disks, EFS vs Azure Files
Storage is one of the clearest areas of conceptual alignment. Microsoft Learn explicitly compares Amazon S3 with Azure Blob Storage as equivalent object storage services, EBS with Azure VM data disks, and EFS with Azure Files.2
Key mappings include:
| AWS | Azure | Primary use |
|---|---|---|
| Amazon S3 | Azure Blob Storage | Object storage for unstructured data, backup, archive, analytics |
| Amazon EBS | Azure Managed Disks / data disks | Block storage for VMs |
| Amazon EFS | Azure Files | Shared file storage |
| Glacier classes | Cool/Cold/archive-oriented Blob tiers | Low-cost archival retention2 |
Important implementation differences exist in access control and network governance. Microsoft notes that S3 access commonly relies on IAM roles and bucket policies, whereas Azure Blob Storage uses a layered model involving storage firewalls and related authorization mechanisms such as shared access signatures.
This leads to different administrative habits:
- AWS teams often think in terms of buckets, IAM policies, and pre-signed URLs.
- Azure teams often think in terms of storage accounts, RBAC, firewalls, and SAS tokens.
From a performance and durability perspective, both platforms offer highly scalable object storage and multiple redundancy options. In practice, architectural correctness—tiering, lifecycle rules, replication design, and access patterns—matters more than brand choice.2
Footnotes
-
Compare storage on Azure and AWS - Microsoft Learn - Official Microsoft comparison of S3, Blob Storage, EBS, EFS, storage access models, and replication concepts. ↩ ↩2 ↩3 ↩4 ↩5 ↩6 ↩7
-
Compare AWS and Azure compute services - Azure Architecture Center - Official Microsoft mapping of AWS and Azure compute services and related infrastructure concepts. ↩ ↩2 ↩3
-
Azure vs AWS Pricing: Which Cloud Provider is Cheaper? - Comparative discussion of pricing models, storage examples, and market-share context. ↩ ↩2
Illustrative object storage price points from a comparative industry source
Selected examples cited in a 2025 comparison; actual pricing varies by region, redundancy, retrieval, and transaction profile.
Footnotes
-
Azure vs AWS Pricing: Which Cloud Provider is Cheaper? - Comparative discussion of pricing models, storage examples, and market-share context. ↩
Pricing comparisons can be misleading
Raw per-GB or per-VM price comparisons are incomplete. Total cost depends on region, reserved commitments, data egress, request volume, replication policy, and existing Windows or SQL Server licenses.2
Footnotes
-
Azure vs AWS Pricing: Which Cloud Provider is Cheaper? - Comparative discussion of pricing models, storage examples, and market-share context. ↩
-
What's the Difference Between AWS vs. Azure vs. Google Cloud? - High-level market and cloud-platform comparison useful for contextual positioning. ↩
4. Identity, security, and compliance
Cloud security is governed by the shared responsibility model: both AWS and Azure secure the underlying platform, while customers remain responsible for configuration, identity, data handling, and workload hardening.
For identity and access management:
- AWS uses IAM for fine-grained policy-based access control and AWS Identity Center for centralized workforce access.2
- Azure uses Entra ID and Azure RBAC, which integrates deeply with Microsoft identity, Windows environments, and enterprise SSO patterns.2
For security operations:
- AWS provides services such as GuardDuty, KMS, and CloudTrail.2
- Azure provides Defender for Cloud, Sentinel, Key Vault, and strong integration with enterprise identity and policy workflows.2
Compliance breadth is strong on both platforms. Comparative sources note that both providers maintain over 100 certifications or compliance offerings, while Microsoft is often credited with especially broad regional and industry-specific compliance coverage.
The central distinction is not that one cloud is inherently secure and the other is not. Rather:
- AWS often appeals to teams wanting granular policy control and modular security architecture.2
- Azure often appeals to enterprises wanting unified identity, compliance alignment, and integrated hybrid security workflows.
Footnotes
-
Azure Security vs. AWS Security: 7 Key Differences - Security-focused overview emphasizing shared responsibility and differences in cloud security tooling. ↩ ↩2 ↩3 ↩4
-
AWS vs Azure vs Google: Cloud Services Comparison - Overview of AWS and Azure strengths, service categories, and positioning in the cloud market. ↩
-
Microsoft Azure vs AWS: A Comprehensive Comparison - Detailed comparison of compute, identity, security, performance, and hybrid characteristics. ↩ ↩2 ↩3 ↩4 ↩5 ↩6 ↩7
-
Azure vs AWS: Which Cloud Platform is Best in 2025? - Broad comparison of ecosystem fit, hybrid posture, performance, pricing, and enterprise use cases. ↩
Security and governance questions
5. Hybrid cloud, enterprise integration, and migration pathways
Hybrid capability is one of the most important differentiators. Azure is widely recognized as particularly strong for hybrid architectures because it extends Microsoft identity, policy, management, and security models across on-premises and cloud environments. This matters for organizations modernizing gradually rather than performing a full cloud-native rebuild.
AWS also supports hybrid deployments through offerings such as AWS Outposts and related extensions, but comparative analyses frequently describe Azure as the more mature or natural fit for enterprises seeking integrated hybrid management.
This difference is partly architectural and partly organizational:
- Azure often fits enterprises that need to preserve compatibility with existing Microsoft systems during migration.2
- AWS often fits organizations that are more willing to re-architect around cloud-native operational patterns.
Migration strategy should therefore align with application state:
| Application condition | Likely fit |
|---|---|
| Legacy Windows workloads with Microsoft identity dependencies | Azure often has lower transition friction2 |
| Greenfield microservices and service-rich architectures | AWS often provides broader composability |
| Strict hybrid continuity with enterprise governance | Azure commonly has an advantage |
| Deep cloud-native experimentation across many managed services | AWS commonly has an advantage |
Footnotes
-
Microsoft Azure vs AWS: A Comprehensive Comparison - Detailed comparison of compute, identity, security, performance, and hybrid characteristics. ↩ ↩2 ↩3 ↩4 ↩5
-
Azure vs AWS: Which Cloud Platform is Best in 2025? - Broad comparison of ecosystem fit, hybrid posture, performance, pricing, and enterprise use cases. ↩ ↩2
-
AWS vs Azure vs Google: Cloud Services Comparison - Overview of AWS and Azure strengths, service categories, and positioning in the cloud market. ↩ ↩2 ↩3
A practical cloud decision roadmap
Portfolio Assessment
Phase 1Classify workloads by operating system, architecture style, data sensitivity, performance profile, and Microsoft dependency before comparing providers.2"
Footnotes
-
Azure vs AWS: Which Cloud Platform is Best in 2025? - Broad comparison of ecosystem fit, hybrid posture, performance, pricing, and enterprise use cases. ↩
-
AWS vs Azure vs Google: Cloud Services Comparison - Overview of AWS and Azure strengths, service categories, and positioning in the cloud market. ↩
Service Mapping
Phase 2Map EC2 to Azure Virtual Machines, S3 to Blob Storage, Lambda to Functions, and EKS to AKS to identify equivalent landing zones.2"
Footnotes
-
Compare AWS and Azure compute services - Azure Architecture Center - Official Microsoft mapping of AWS and Azure compute services and related infrastructure concepts. ↩
-
Compare storage on Azure and AWS - Microsoft Learn - Official Microsoft comparison of S3, Blob Storage, EBS, EFS, storage access models, and replication concepts. ↩
Security Design
Phase 3Define IAM or RBAC, encryption, logging, backup, and network isolation patterns under the shared responsibility model.2"
Footnotes
-
Compare storage on Azure and AWS - Microsoft Learn - Official Microsoft comparison of S3, Blob Storage, EBS, EFS, storage access models, and replication concepts. ↩
-
Azure Security vs. AWS Security: 7 Key Differences - Security-focused overview emphasizing shared responsibility and differences in cloud security tooling. ↩
Cost Modeling
Phase 4Model reserved, spot, storage access, egress, and license reuse rather than relying on list price snapshots.2"
Footnotes
-
Azure vs AWS Pricing: Which Cloud Provider is Cheaper? - Comparative discussion of pricing models, storage examples, and market-share context. ↩
-
What's the Difference Between AWS vs. Azure vs. Google Cloud? - High-level market and cloud-platform comparison useful for contextual positioning. ↩
Pilot and Optimize
Phase 5Run production-like pilots, compare operational burden, and optimize for reliability, governance, and skill alignment.2"
Footnotes
-
Azure vs AWS: Which Cloud Platform is Best in 2025? - Broad comparison of ecosystem fit, hybrid posture, performance, pricing, and enterprise use cases. ↩
-
AWS vs Azure vs Google: Cloud Services Comparison - Overview of AWS and Azure strengths, service categories, and positioning in the cloud market. ↩
6. Pricing and total cost of ownership
Both AWS and Azure use pay-as-you-go pricing, but cost outcomes depend far more on workload profile than on nominal list prices.2 Each provider also offers long-term commitment discounts and lower-cost interruptible capacity:
- AWS: Reserved-style savings and Spot usage patterns
- Azure: Reservations, Spot VMs, and Azure Hybrid Benefit for eligible Microsoft licenses2
Azure can become economically attractive when an organization already owns Windows Server or SQL Server licenses that qualify for Hybrid Benefit.2 AWS can be attractive where teams exploit instance-family variety, spot elasticity, or deeply optimized cloud-native architectures.2
A simple total-cost model can be expressed as:
The final term, “risk overhead,” is often underestimated. A platform that is slightly cheaper per VM but significantly harder for a team to govern, secure, or operate can become more expensive overall.2
Footnotes
-
Azure vs AWS Pricing: Which Cloud Provider is Cheaper? - Comparative discussion of pricing models, storage examples, and market-share context. ↩ ↩2 ↩3 ↩4 ↩5
-
What's the Difference Between AWS vs. Azure vs. Google Cloud? - High-level market and cloud-platform comparison useful for contextual positioning. ↩ ↩2
-
Microsoft Azure vs AWS: A Comprehensive Comparison - Detailed comparison of compute, identity, security, performance, and hybrid characteristics. ↩ ↩2
-
AWS vs Azure vs Google: Cloud Services Comparison - Overview of AWS and Azure strengths, service categories, and positioning in the cloud market. ↩
-
Azure vs AWS: Which Cloud Platform is Best in 2025? - Broad comparison of ecosystem fit, hybrid posture, performance, pricing, and enterprise use cases. ↩
Use TCO, not sticker price
If your organization already depends on Microsoft licensing and identity, Azure may lower total cost through operational and licensing efficiencies. If your team is highly cloud-native and automation-heavy, AWS may lower cost through service flexibility and design efficiency.3
Footnotes
-
Azure vs AWS Pricing: Which Cloud Provider is Cheaper? - Comparative discussion of pricing models, storage examples, and market-share context. ↩
-
Azure vs AWS: Which Cloud Platform is Best in 2025? - Broad comparison of ecosystem fit, hybrid posture, performance, pricing, and enterprise use cases. ↩
-
AWS vs Azure vs Google: Cloud Services Comparison - Overview of AWS and Azure strengths, service categories, and positioning in the cloud market. ↩
7. Global infrastructure, regions, and availability design
Both providers operate global infrastructure at large scale, but their design emphases differ. Comparative sources note that Azure has 60+ geographic locations, while AWS is often credited with more Availability Zones per region and a large edge network footprint.
This distinction matters for availability zones and disaster recovery design:
- AWS architecture often emphasizes multi-AZ patterns within a region as a foundational availability strategy.
- Azure supports availability zones as well, but is also commonly chosen for region-pair and hybrid continuity strategies, especially in enterprise governance contexts.
The practical implication is that architects should compare not only “how many regions exist,” but also:
- Which regions are approved for the organization
- Whether zones are available in the required geography
- Data residency and replication implications
- Edge connectivity and latency requirements
- Disaster recovery topology and failover automation
Footnotes
-
Microsoft Azure vs AWS: A Comprehensive Comparison - Detailed comparison of compute, identity, security, performance, and hybrid characteristics. ↩ ↩2 ↩3 ↩4
When to choose AWS vs Azure
8. Final synthesis
AWS and Azure are both highly capable enterprise cloud platforms. The most defensible academic conclusion is not that one universally dominates the other, but that each embodies a different optimization strategy.3
- AWS optimizes for breadth, maturity, modularity, and cloud-native architectural flexibility.2
- Azure optimizes for enterprise integration, Microsoft ecosystem alignment, and hybrid operational continuity.2
Therefore:
- Choose AWS when service breadth, composability, and cloud-native depth are the primary requirements.
- Choose Azure when Microsoft alignment, identity coherence, and hybrid governance are central constraints.2
- Choose either only after validating real workload behavior, cost, and operating model through pilots rather than marketing comparisons.2
Footnotes
-
Azure vs AWS: Which Cloud Platform is Best in 2025? - Broad comparison of ecosystem fit, hybrid posture, performance, pricing, and enterprise use cases. ↩ ↩2 ↩3
-
AWS vs Azure vs Google: Cloud Services Comparison - Overview of AWS and Azure strengths, service categories, and positioning in the cloud market. ↩ ↩2 ↩3
-
Microsoft Azure vs AWS: A Comprehensive Comparison - Detailed comparison of compute, identity, security, performance, and hybrid characteristics. ↩ ↩2 ↩3
-
Compare AWS and Azure compute services - Azure Architecture Center - Official Microsoft mapping of AWS and Azure compute services and related infrastructure concepts. ↩
-
Azure vs AWS Pricing: Which Cloud Provider is Cheaper? - Comparative discussion of pricing models, storage examples, and market-share context. ↩
-
What's the Difference Between AWS vs. Azure vs. Google Cloud? - High-level market and cloud-platform comparison useful for contextual positioning. ↩
Knowledge Check
Which pairing correctly matches equivalent object storage services?
Explore Related Topics
AI vs Human Teachers: A Comprehensive Analysis
The module examines AI versus human teachers, advocating a hybrid approach where AI automates routine, personalized tasks while teachers supply emotional, mentorship, and critical‑thinking support.
- AI provides 24/7 availability, adaptive personalization, instant objective feedback, and scalability, freeing ~10 hrs/week of teacher workload.
- Human teachers contribute empathy, mentorship, cultural interpretation, ethical judgment, and social modeling—capabilities AI cannot replicate.
- Studies show AI use raises engagement (, ) but excessive reliance harms critical‑thinking skills.
- Optimal effectiveness combines AI efficiency with human depth: .
PostgreSQL vs MySQL: A Comprehensive Comparison
This course compares PostgreSQL and MySQL across architecture, types, performance, and selection guidance.
- PostgreSQL is an ORDBMS with process‑per‑connection, full ACID, advanced planner, and extensions (JSONB, PostGIS).
- MySQL uses thread‑per‑connection, pluggable engines (InnoDB, MyISAM), shines on read‑heavy simple workloads; ACID only with InnoDB/NDB.
- Benchmarks show PostgreSQL 9–13× faster on point SELECTs and 10–15× lower latency under concurrent read/write; INSERT speeds are similar.
- PostgreSQL’s JSONB stores binary data with GIN indexing and rich operators; MySQL’s JSON lacks such indexes.
- Choose PostgreSQL for complex queries, strict ACID, advanced types, and extensibility; choose MySQL for quick setup, read‑only workloads, and existing expertise.
Amazon Prime Video: From Serverless Microservices to Monolithic Architecture — A System Design Case Study
Amazon Prime Video migrated its Video Quality Analysis pipeline from a 30‑service serverless micro‑architecture to a single container on ECS/Fargate, cutting total infrastructure cost by ≈ 90% and boosting latency and throughput.
- Serverless design incurred high orchestration (), S3 data‑transfer (), and DynamoDB state‑sync costs.
- Collapsing the pipeline into one monolithic container removed Step Functions, S3, and DynamoDB overhead, achieving massive cost savings.
- In‑process communication replaced network hops, lowering latency from hundreds of milliseconds to near‑zero and increasing throughput.
- Scaling is done horizontally via ECS task scaling, preserving elasticity while sacrificing independent deployability, which was unused for this tightly‑coupled pipeline.