AZ-305 remains Microsoft’s advanced design exam for Azure solutions architects, and the current English skills were updated on April 17, 2026. The role is explicitly architectural: candidates advise stakeholders and translate business requirements into cloud and hybrid designs that align with the Azure Well-Architected Framework and the Cloud Adoption Framework for Azure. Microsoft expects advanced knowledge of networking, virtualization, identity, security, business continuity, disaster recovery, data platforms, governance, Azure administration, development, and DevOps.<\/p>\n
The current AZ-305<\/a> blueprint has four weighted areas: identity, governance, and monitoring at 25\u201330%; data storage at 20\u201325%; business continuity at 15\u201320%; and infrastructure solutions at 30\u201335%. A score of 700 or higher is required to pass. The exam is part of the requirements for the Microsoft Certified: Azure Solutions Architect Expert credential, which also requires the Azure Administrator Associate certification.<\/p>\n The first domain covers logging and monitoring, authentication and authorization, identity management, access to Azure and on-premises resources, secrets, certificates and keys, management groups, subscriptions, resource groups, tagging, compliance, and identity governance. These topics should be understood as one control plane around the workload rather than isolated services.<\/p>\n A review of Azure Monitor<\/a> can reinforce the monitoring side, while Azure Key Vault<\/a> helps connect secret and key management to application design. The architect should know which team owns which control and how evidence is routed for operations and compliance.<\/p>\n AZ-305 expects candidates to recommend a structure for management groups, subscriptions, resource groups, and resource tagging. Those choices affect policy inheritance, billing, ownership, deployment boundaries, and the blast radius of administrative changes.<\/p>\n The strongest design follows organizational and risk boundaries rather than arbitrary project size. A production workload, regulated data domain, or regional business unit may justify separate subscriptions or management-group policies, while resource groups should reflect lifecycle and deployment ownership.<\/p>\n The blueprint asks for authentication, identity management, authorization to Azure resources, authorization to on-premises resources, and management of secrets, certificates, and keys. That requires architects to distinguish human identity, workload identity, hybrid identity, role-based access, and application credentials.<\/p>\n Least privilege remains the design principle. A managed identity can remove embedded credentials for supported Azure workloads, while role assignments, Privileged Identity Management, and policy boundaries can reduce standing access. Hybrid applications may also need secure access to on-premises resources without collapsing identity boundaries.<\/p>\n The data-storage domain expects candidates to recommend a relational service, database tier, compute tier, scalability approach, and data-protection design. Azure SQL Database, Managed Instance, SQL Server on virtual machines, and other relational choices differ in compatibility, control, operations, scale, and high-availability behavior.<\/p>\n Architects should begin with workload semantics: compatibility requirements, transaction pattern, performance, regional needs, administration, and recovery. Moving to a more managed service can reduce operational burden, but some workloads still require infrastructure-level control.<\/p>\n The current exam asks candidates to recommend storage for semi-structured and unstructured data and to balance features, performance, cost, protection, and durability. That can involve Azure Storage, Data Lake patterns, Cosmos DB, managed file services, or other services depending on access behavior.<\/p>\n A Cosmos DB<\/a> review is useful for globally distributed and nonrelational scenarios. The key skill is matching consistency, latency, partitioning, scale, query pattern, and operational model to the business requirement rather than treating every nonrelational dataset the same way.<\/p>\n AZ-305 includes recommendations for data integration and data analysis. That means the architect needs to think about movement, orchestration, event streams, batch pipelines, analytical consumption, governance, and cost around data\u2014not only where it rests.<\/p>\n Integration decisions should state who owns the source, how changes are detected, how failures are recovered, and how downstream consumers know the data is current. A storage design without a reliable movement and consumption path is incomplete.<\/p>\n The third domain covers backup and disaster recovery for Azure and hybrid workloads, compute, databases, and unstructured data, plus high availability for compute and data services. Candidates should distinguish high availability from disaster recovery and backup from replication.<\/p>\n A review of Azure Backup<\/a> can support recovery concepts, but AZ-305 expects architects to start from recovery time and recovery point objectives. The appropriate architecture depends on how much downtime and data loss the business can tolerate.<\/p>\n The largest domain begins with compute choices. Candidates should recommend virtual-machine-based, container-based, serverless, and batch-processing solutions according to workload needs. Control, portability, scaling, startup behavior, platform limits, security, and team capability all matter.<\/p>\n The Azure compute<\/a> perspective helps candidates compare operating models. A technically capable service can still be the wrong architecture if the team cannot operate it or if a simpler managed alternative meets the same requirement with less risk.<\/p>\n The exam expects candidates to recommend messaging and event-driven architectures, API integration, application caching, configuration management, and automated deployment. This makes AZ-305 a distributed-application design exam as well as an infrastructure exam.<\/p>\n Azure Service Bus<\/a>, API Management<\/a>, and Azure Cache for Redis<\/a> illustrate different responsibilities: durable messaging, API governance, and low-latency caching. The architect should know which problem each layer solves and how failure at one layer affects the rest of the application.<\/p>\n The infrastructure domain also covers migration using the Cloud Adoption Framework, assessment of on-premises servers, data and applications, IaaS and PaaS migration, database migration, unstructured-data migration, internet connectivity, hybrid connectivity, network performance, network security, and load balancing or routing.<\/p>\n The current April 2026 update also reinforces that AZ-305 is about design recommendations rather than implementation syntax. The blueprint repeatedly uses the verb \u201crecommend,\u201d which means candidates should be able to compare alternatives and justify why one design better fits requirements. Knowing how to create a resource is useful, but the exam is aimed at architects who decide which resource pattern the implementation teams should use.<\/p>\n Logging and monitoring should be studied as architecture because routing logs, centralizing data, setting retention, and defining alert ownership all affect operations and compliance. An architect may recommend centralized Log Analytics workspaces, diagnostic settings, or service-specific monitoring patterns based on organizational boundaries and data sensitivity. The design should also state who investigates the signal and how long the evidence must remain available.<\/p>\n Identity governance is broader than Azure RBAC. It can include entitlement lifecycle, privileged access, reviews, group and application access, and hybrid identity considerations. The architect’s responsibility is to recommend a control model that gives users and workloads enough access to operate while limiting standing privilege and maintaining an auditable lifecycle.<\/p>\n Relational-data protection should be connected to service choice. Built-in backups, zone redundancy, geo-replication, failover groups, and point-in-time restore capabilities vary across platforms and tiers. The strongest recommendation starts from business recovery objectives and then chooses a service or tier that can meet them without adding unnecessary custom infrastructure.<\/p>\n For semi-structured and unstructured data, durability and protection are only part of the decision. Access protocol, namespace, latency, throughput, geographic distribution, lifecycle, analytics integration, and cost all matter. Architects should be able to explain why object storage, hierarchical namespace, files, or a globally distributed database is the better fit for the specific workload.<\/p>\n Data integration should also be viewed as a failure and ownership problem. Batch pipelines, event streams, queues, APIs, and replication create dependencies among systems. The architect should know where data is authoritative, how duplicates or late events are handled, and how downstream consumers recover when movement fails.<\/p>\n Business continuity questions often include both availability and recoverability. Zone-redundant compute can protect against a local failure, but accidental deletion or corruption may still require backup. Geo-replication can reduce regional recovery time, but it can introduce consistency, cost, or operational complexity. Each mechanism addresses a different failure class.<\/p>\n Compute selection is also influenced by deployment and organizational capability. Virtual machines offer broad compatibility and control, containers provide portability and orchestration options, serverless platforms reduce infrastructure ownership, and batch services optimize scheduled or parallel processing. Architects should include team skill, compliance, network integration, and release process in the recommendation.<\/p>\n Application architecture objectives make the exam highly cross-domain. A queue affects business continuity, APIs affect identity and governance, caching affects data consistency and performance, configuration services affect release safety, and automated deployment affects operational excellence. Candidates should expect scenarios where one supporting service changes several design properties at once.<\/p>\n Migration objectives are especially important for hybrid organizations. Assessment should identify application dependencies, database compatibility, network paths, identity, data volume, downtime tolerance, and whether IaaS or PaaS better fits the target operating model. A technically ideal PaaS destination may not be the first migration step if deadlines or compatibility make a staged approach safer.<\/p>\n One final scope distinction is certification context. AZ-305 is the required exam for the Azure Solutions Architect Expert certification, but the expert credential also requires the Azure Administrator Associate certification. That prerequisite structure reinforces the intended role: the architect should already understand how Azure resources are administered and can therefore focus this exam on cross-domain design choices, trade-offs, and stakeholder recommendations.<\/p>\n For final preparation, treat the blueprint as a set of recommendation problems. If you cannot explain why one Azure design is stronger than the nearest alternative under the stated constraints, return to the requirement rather than adding more service memorization.<\/p>\n A virtual network<\/a> foundation and Azure Load Balancer<\/a> context can support these objectives. Within the broader Azure Solutions Architect Expert<\/a> path, AZ-305 is ultimately about making cross-domain trade-offs that preserve business requirements across identity, data, continuity, infrastructure, and operations.<\/p>\n","protected":false},"excerpt":{"rendered":" AZ-305 remains Microsoft’s advanced design exam for Azure solutions architects, and the current English skills were updated on April 17, 2026. The role is explicitly architectural: candidates advise stakeholders and translate business requirements into cloud and hybrid designs that align with the Azure Well-Architected Framework and the Cloud Adoption Framework for Azure. Microsoft expects advanced […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":[],"categories":[1648,1647],"tags":[],"_links":{"self":[{"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/posts\/26265"}],"collection":[{"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/comments?post=26265"}],"version-history":[{"count":1,"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/posts\/26265\/revisions"}],"predecessor-version":[{"id":26266,"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/posts\/26265\/revisions\/26266"}],"wp:attachment":[{"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/media?parent=26265"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/categories?post=26265"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/tags?post=26265"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}Identity, governance, and monitoring form the control plane for the whole design<\/h3>\n
Management hierarchy and tagging make governance scalable<\/h3>\n
Identity design connects Microsoft Entra with Azure and hybrid access<\/h3>\n
Relational data design is a choice among engines, tiers, scale, and protection<\/h3>\n
Semi-structured and unstructured storage require access-pattern reasoning<\/h3>\n
Data integration and analysis extend storage into the wider platform<\/h3>\n
Business continuity is built around recovery objectives and failure domains<\/h3>\n
Compute design spans virtual machines, containers, serverless, and batch<\/h3>\n
Application architecture includes messaging, events, APIs, caching, configuration, and deployment<\/h3>\n
Migrations and network architecture complete the current scope<\/h3>\n