{"id":16504,"date":"2026-09-19T07:53:24","date_gmt":"2026-09-19T07:53:24","guid":{"rendered":"https:\/\/www.examlabs.com\/certification\/?p=16504"},"modified":"2026-09-19T07:53:24","modified_gmt":"2026-09-19T07:53:24","slug":"microsoft-dp-420-practice-test-questions-and-exam-dumps-part5-q81-100","status":"publish","type":"post","link":"https:\/\/www.examlabs.com\/certification\/microsoft-dp-420-practice-test-questions-and-exam-dumps-part5-q81-100\/","title":{"rendered":"Microsoft DP-420 Practice Test Questions and Exam Dumps Part5 Q81-100"},"content":{"rendered":"

 <\/p>\n

View Full <\/b>Microsoft DP-420 Exam Dumps<\/b><\/a> and Practice Test Dumps.<\/b><\/p>\n


\n<\/b>Q1. You are designing a Cosmos DB workload where one set of queries is optimized by customer ID and another independent set is optimized by product ID. No single partition key satisfies both access patterns. What should you consider?<\/b><\/p>\n

    \n
  1. <\/b> Use one constant partition key for all items.<\/span><\/li>\n
  2. <\/b> Use separate containers or duplicated data models optimized for the different access patterns.<\/span><\/li>\n
  3. <\/b> Disable partitioning.<\/span><\/li>\n
  4. <\/b> Use TTL to switch partition keys automatically.<\/span><\/li>\n<\/ol>\n

    Correct Answer: 2. Use separate containers or duplicated data models optimized for the different access patterns.<\/b><\/p>\n

    Explanation:<\/b> Azure Cosmos DB supports one partition-key definition per container, so a single container cannot independently optimize unrelated access patterns that require fundamentally different partition keys. When two important workloads require different distribution and routing strategies, separate containers or denormalized projections can be appropriate. Change feed processing can help keep duplicated projections synchronized. A constant partition key creates severe scalability limits, TTL controls expiration rather than partitioning, and scalable Cosmos DB containers depend on partitioning. The correct design should be driven by access patterns, transaction boundaries, storage growth, and throughput distribution.<\/span><\/p>\n

    Q2. You are creating a small development Cosmos DB account and want to minimize cost. Your Azure subscription has not previously used the Cosmos DB free tier. What should you evaluate during account creation?<\/b><\/p>\n

      \n
    1. <\/b> Continuous backup only.<\/span><\/li>\n
    2. <\/b> Multi-region writes.<\/span><\/li>\n
    3. <\/b> Strong consistency.<\/span><\/li>\n
    4. <\/b> Free tier eligibility.<\/span><\/li>\n<\/ol>\n

      Correct Answer: 4. Free tier eligibility.<\/b><\/p>\n

      Explanation:<\/b> Azure Cosmos DB free tier can provide a limited amount of throughput and storage at no charge for an eligible account, making it useful for development, learning, and smaller workloads. Eligibility and limits should be confirmed when the account is created because the option is subject to Azure subscription and account rules. Multi-region writes, strong consistency, and backup choices solve architectural requirements but do not by themselves minimize basic development costs. DP-420 candidates should understand how workload size, throughput mode, storage, global distribution, and available pricing options influence Cosmos DB cost planning.<\/span><\/p>\n

      Q3. You are using the Cosmos DB SDK in an application that performs thousands of concurrent independent operations. You want to increase client-side parallelism without creating a new Cosmos client for every operation. What should you do?<\/b><\/p>\n

        \n
      1. <\/b> Reuse the client and configure appropriate SDK parallelism or concurrency options.<\/span><\/li>\n
      2. <\/b> Create one client for every document.<\/span><\/li>\n
      3. <\/b> Disable all asynchronous operations.<\/span><\/li>\n
      4. <\/b> Route every operation through a stored procedure.<\/span><\/li>\n<\/ol>\n

        Correct Answer: 1. Reuse the client and configure appropriate SDK parallelism or concurrency options.<\/b><\/p>\n

        Explanation:<\/b> Cosmos DB SDK clients are designed to be reused, and high-throughput applications should generally use asynchronous operations and suitable concurrency settings rather than repeatedly creating clients. Proper client-side threading and parallelism can improve throughput while preserving efficient connection reuse. Creating many clients increases resource consumption and can cause connection-management problems. Stored procedures are limited to transactional logic within a logical partition and are not a general replacement for scalable client concurrency. Applications should also consider bulk support, connection mode, partition distribution, and available RU\/s when tuning parallel execution.<\/span><\/p>\n

        Q4. Documents in one container have different optional properties. You need a SQL query that returns items only when the <\/b>promotionCode<\/b> property exists. Which query capability is most appropriate?<\/b><\/p>\n

          \n
        1. <\/b> Spatial querying.<\/span><\/li>\n
        2. <\/b> Transactional Batch.<\/span><\/li>\n
        3. <\/b> A type or existence-checking function such as <\/span>IS_DEFINED<\/span>.<\/span><\/li>\n
        4. <\/b> Continuous backup.<\/span><\/li>\n<\/ol>\n

          Correct Answer: 3. A type or existence-checking function such as <\/b>IS_DEFINED<\/b>.<\/b><\/p>\n

          Explanation:<\/b> Azure Cosmos DB for NoSQL supports flexible JSON schemas, so properties can be present in some documents and missing from others. A function such as <\/span>IS_DEFINED<\/span> can test whether a property exists before additional logic is applied. This is useful for queries over variable-schema data and helps prevent assumptions that every item contains the same fields. Spatial functions operate on geographic data, Transactional Batch performs atomic operations within a logical partition, and continuous backup supports recovery. DP-420 candidates should be comfortable querying heterogeneous JSON documents using type-checking and existence functions.<\/span><\/p>\n

          Q5. You need a query that returns the total number of orders and the average order amount for a selected customer. Which Cosmos DB SQL capability should you use?<\/b><\/p>\n

            \n
          1. <\/b> Aggregate functions such as <\/span>COUNT<\/span> and <\/span>AVG<\/span>.<\/span><\/li>\n
          2. <\/b> TTL.<\/span><\/li>\n
          3. <\/b> Change feed leases.<\/span><\/li>\n
          4. <\/b> Manual failover.<\/span><\/li>\n<\/ol>\n

            Correct Answer: 1. Aggregate functions such as <\/b>COUNT<\/b> and <\/b>AVG<\/b>.<\/b><\/p>\n

            Explanation:<\/b> Cosmos DB for NoSQL SQL supports aggregate functions such as COUNT, SUM, AVG, MIN, and MAX. These functions allow server-side aggregation of matching documents and can reduce the need to retrieve every item and calculate totals in application code. The query should ideally include the customer partition key when possible so execution can be routed efficiently. TTL controls expiration, change feed leases coordinate processors, and failover changes regional roles. Aggregation queries can still consume significant request units, so developers should inspect query cost and indexes when operating over large datasets.<\/span><\/p>\n

            Q6. You operate a write-heavy Cosmos DB container. Most document properties are never used in filters, sorting, or projections that require indexing. Which indexing strategy should you evaluate?<\/b><\/p>\n

              \n
            1. <\/b> Index every possible property with multiple composite indexes.<\/span><\/li>\n
            2. <\/b> Add a unique key to every field.<\/span><\/li>\n
            3. <\/b> Use strong consistency to reduce index writes.<\/span><\/li>\n
            4. <\/b> Exclude unnecessary paths from the indexing policy.<\/span><\/li>\n<\/ol>\n

              Correct Answer: 4. Exclude unnecessary paths from the indexing policy.<\/b><\/p>\n

              Explanation:<\/b> Every indexed property adds index-maintenance work when documents are written or updated. In a write-heavy workload, excluding properties that are never queried can reduce indexing overhead and request-unit consumption. The indexing policy should still include paths required by important filters, ordering, or search patterns. Creating unnecessary composite indexes increases maintenance cost, while unique keys serve data-integrity requirements rather than general query optimization. Consistency level does not eliminate index maintenance. A well-designed write-heavy indexing policy balances efficient reads against the cost of maintaining indexes during frequent writes.<\/span><\/p>\n

              Q7. A query scans many documents even though it returns only a small number of results. You want to determine whether the indexing policy could be improved. What should you review?<\/b><\/p>\n

                \n
              1. <\/b> Only the account’s Azure region.<\/span><\/li>\n
              2. <\/b> Query metrics and index utilization information.<\/span><\/li>\n
              3. <\/b> The backup retention period.<\/span><\/li>\n
              4. <\/b> Only the document TTL.<\/span><\/li>\n<\/ol>\n

                Correct Answer: 2. Query metrics and index utilization information.<\/b><\/p>\n

                Explanation:<\/b> Query metrics provide insight into execution behavior and request-unit consumption, helping identify queries that inspect substantially more data than they return. Index-related diagnostics can indicate whether the indexing policy supports the query efficiently or whether a different policy or composite index could help. Region selection, TTL, and backup retention do not explain poor query selectivity or index usage. Query tuning should combine metrics, partition routing, predicate design, and indexing analysis. A query that is logically correct can still be expensive if it fans out across partitions or cannot use the index effectively.<\/span><\/p>\n

                Q8. A multi-region Cosmos DB application experiences increased read latency from one Azure region. You need to determine whether replication behavior may be contributing. What should you monitor?<\/b><\/p>\n

                  \n
                1. <\/b> Only the number of unique keys.<\/span><\/li>\n
                2. <\/b> Only item TTL values.<\/span><\/li>\n
                3. <\/b> Replication and regional latency metrics.<\/span><\/li>\n
                4. <\/b> Only composite-index definitions.<\/span><\/li>\n<\/ol>\n

                  Correct Answer: 3. Replication and regional latency metrics.<\/b><\/p>\n

                  Explanation:<\/b> In globally distributed Cosmos DB accounts, latency can be influenced by regional connectivity, replication behavior, consistency settings, and application routing. Monitoring regional and replication-related metrics can help determine whether a particular region is experiencing delays or whether consistency requirements are affecting request behavior. Unique keys, TTL, and composite-index definitions do not explain regional replication latency. Troubleshooting should also consider preferred regions, SDK diagnostics, server-side latency, network latency, and failover configuration to distinguish database processing delays from client or network issues.<\/span><\/p>\n

                  Q9. Your organization does not require second-by-second point-in-time recovery but needs automatic backups at configured intervals. Which backup mode should you consider?<\/b><\/p>\n

                    \n
                  1. <\/b> Change feed only.<\/span><\/li>\n
                  2. <\/b> Integrated cache.<\/span><\/li>\n
                  3. <\/b> Analytical store.<\/span><\/li>\n
                  4. <\/b> Periodic backup.<\/span><\/li>\n<\/ol>\n

                    Correct Answer: 4. Periodic backup.<\/b><\/p>\n

                    Explanation:<\/b> Periodic backup creates backups according to a defined schedule and retention configuration. It can be appropriate when recovery requirements do not demand fine-grained point-in-time restoration. Continuous backup is the option used when point-in-time restore within the supported retention window is required. Change feed is an event-processing mechanism, integrated cache improves read performance, and analytical store supports analytical workloads. Backup strategy should be selected based on recovery-point objectives, recovery-time expectations, operational requirements, and the impact of accidental deletion or corruption on the business.<\/span><\/p>\n

                    Q10. You are configuring periodic backup for a Cosmos DB account. Which two settings are directly relevant to how often backups are taken and how long they are retained?<\/b><\/p>\n

                      \n
                    1. <\/b> Backup interval and retention period.<\/span><\/li>\n
                    2. <\/b> Partition key and unique key.<\/span><\/li>\n
                    3. <\/b> Session token and continuation token.<\/span><\/li>\n
                    4. <\/b> Conflict policy and TTL.<\/span><\/li>\n<\/ol>\n

                      Correct Answer: 1. Backup interval and retention period.<\/b><\/p>\n

                      Explanation:<\/b> Periodic backup configuration includes settings that determine how frequently backups are created and how long those backups are retained. These values influence the available recovery points and should align with organizational recovery requirements. Partition keys govern data distribution, while unique keys enforce constraints. Session and continuation tokens relate to consistency and query pagination. Conflict policies and TTL address multi-region writes and item expiration. Backup configuration should be treated as part of the application’s resilience plan and validated through documented restore procedures rather than assumed to work only when an incident occurs.<\/span><\/p>\n

                      Q11. You are using continuous backup and need to restore a database after an application defect corrupted data. What should you identify before initiating the restore?<\/b><\/p>\n

                        \n
                      1. <\/b> The item’s integrated-cache age.<\/span><\/li>\n
                      2. <\/b> An appropriate restore point before the corruption occurred.<\/span><\/li>\n
                      3. <\/b> The client’s session token.<\/span><\/li>\n
                      4. <\/b> The current composite-index count.<\/span><\/li>\n<\/ol>\n

                        Correct Answer: 2. An appropriate restore point before the corruption occurred.<\/b><\/p>\n

                        Explanation:<\/b> Point-in-time restore requires selecting a restore point that predates the unwanted changes while minimizing loss of valid subsequent data. Operators should identify when corruption began and select an appropriate time within the supported continuous-backup retention window. Session tokens, integrated-cache age, and index counts do not determine recovery state. Recovery procedures should document how to locate valid restore points, restore databases or containers, validate restored data, and redirect applications when necessary. Choosing the wrong restore point can reintroduce corrupted data or unnecessarily lose later valid changes.<\/span><\/p>\n

                        Q12. A Fabric Warehouse contains historical analytical data replicated from Cosmos DB. You need to query the warehouse as it existed at an earlier point in time for investigation. Which capability should you use?<\/b><\/p>\n

                          \n
                        1. <\/b> TTL.<\/span><\/li>\n
                        2. <\/b> Manual regional failover.<\/span><\/li>\n
                        3. <\/b> Point reads.<\/span><\/li>\n
                        4. <\/b> Time travel in Microsoft Fabric Warehouse.<\/span><\/li>\n<\/ol>\n

                          Correct Answer: 4. Time travel in Microsoft Fabric Warehouse.<\/b><\/p>\n

                          Explanation:<\/b> Time travel in Fabric Warehouse allows supported analytical queries to access historical states of warehouse data, which can be valuable for auditing, investigation, and comparison. This is distinct from Cosmos DB operational point-in-time restore because time travel is an analytical warehouse capability rather than a transactional recovery operation. TTL deletes data automatically, manual failover changes regional write roles, and point reads retrieve current known items from Cosmos DB. DP-420 now includes analytical integration topics such as Fabric Mirroring and time-travel scenarios in Microsoft Fabric.<\/span><\/p>\n

                          Q13. A Spark job reads records from Cosmos DB, performs enrichment, and must write the transformed documents back to the transactional store. Which integration should you use?<\/b><\/p>\n

                            \n
                          1. <\/b> Only Azure Monitor.<\/span><\/li>\n
                          2. <\/b> A backup restore job.<\/span><\/li>\n
                          3. <\/b> Azure Cosmos DB Spark connector with write support.<\/span><\/li>\n
                          4. <\/b> A sensitivity label.<\/span><\/li>\n<\/ol>\n

                            Correct Answer: 3. Azure Cosmos DB Spark connector with write support.<\/b><\/p>\n

                            Explanation:<\/b> The Azure Cosmos DB Spark connector supports Spark-based interaction with the transactional store, including supported read and write operations. This makes it appropriate for distributed transformation jobs that must persist enriched or derived documents back to Cosmos DB. Azure Monitor is used for observability, backup restores recover data, and sensitivity labels are unrelated to transactional writes. When writing through Spark, the workload should still consider partition-key values, throughput, write concurrency, item sizes, and the potential impact on operational workloads.<\/span><\/p>\n

                            Q14. You need to delete one known item as efficiently as possible. You know its <\/b>id<\/b> and partition key value. Which operation should you use?<\/b><\/p>\n

                              \n
                            1. <\/b> A cross-partition SQL query followed by deletion.<\/span><\/li>\n
                            2. <\/b> A point delete operation using the item ID and partition key.<\/span><\/li>\n
                            3. <\/b> A change feed processor.<\/span><\/li>\n
                            4. <\/b> A stored procedure that scans every partition.<\/span><\/li>\n<\/ol>\n

                              Correct Answer: 2. A point delete operation using the item ID and partition key.<\/b><\/p>\n

                              Explanation:<\/b> When both an item’s ID and partition key are known, an SDK point delete can target the exact item directly. This avoids an unnecessary query and reduces request-unit cost and latency. Cross-partition scanning is inefficient for a known item, while the change feed is intended for processing changes rather than initiating deletion. A stored procedure is constrained to a logical partition and would still be excessive for deleting one known item. Efficient Cosmos DB application design uses point operations whenever item identity and routing information are available.<\/span><\/p>\n

                              Q15. A container has a default TTL of one hour. One document must remain for 24 hours instead. What should you do?<\/b><\/p>\n

                                \n
                              1. <\/b> Set an item-level TTL value that overrides the container default.<\/span><\/li>\n
                              2. <\/b> Disable all indexing for the document.<\/span><\/li>\n
                              3. <\/b> Change the account consistency level.<\/span><\/li>\n
                              4. <\/b> Create another Azure region.<\/span><\/li>\n<\/ol>\n

                                Correct Answer: 1. Set an item-level TTL value that overrides the container default.<\/b><\/p>\n

                                Explanation:<\/b> When TTL is enabled on the container, individual items can specify their own TTL value to override the default expiration interval. This is useful when most items share a common lifecycle but certain records must remain longer or expire sooner. Indexing, consistency, and regional distribution do not control an individual document’s expiration time. Item-level TTL allows lifecycle behavior to remain data-driven without requiring custom cleanup jobs or separate containers solely for different expiration periods.<\/span><\/p>\n

                                Q16. Your application writes documents with several properties that have very high cardinality, but only two are used in queries. You want to optimize a custom indexing policy for write performance. What should you do?<\/b><\/p>\n

                                  \n
                                1. <\/b> Add all high-cardinality paths to multiple composite indexes.<\/span><\/li>\n
                                2. <\/b> Disable the partition key.<\/span><\/li>\n
                                3. <\/b> Include the required query paths and exclude unused paths where appropriate.<\/span><\/li>\n
                                4. <\/b> Use strong consistency.<\/span><\/li>\n<\/ol>\n

                                  Correct Answer: 3. Include the required query paths and exclude unused paths where appropriate.<\/b><\/p>\n

                                  Explanation:<\/b> A custom indexing policy can reduce write overhead by indexing only the properties needed for query patterns while excluding paths that are not queried. High-cardinality properties that are unnecessarily indexed can increase index maintenance cost during writes. However, removing an index from a path required by queries can make those queries unsupported or more expensive, so policies must be workload-driven. Partition keys and consistency settings address different architectural concerns. Index optimization should consider read frequency, write frequency, query predicates, ordering, and request-unit metrics.<\/span><\/p>\n

                                  Q17. A production account experiences a rise in HTTP 5xx responses. Which Azure Monitor information should you review first?<\/b><\/p>\n

                                    \n
                                  1. <\/b> Status-code and failure metrics together with server-side latency.<\/span><\/li>\n
                                  2. <\/b> Only the number of stored procedures.<\/span><\/li>\n
                                  3. <\/b> Only item TTL values.<\/span><\/li>\n
                                  4. <\/b> Only the unique-key policy.<\/span><\/li>\n<\/ol>\n

                                    Correct Answer: 1. Status-code and failure metrics together with server-side latency.<\/b><\/p>\n

                                    Explanation:<\/b> HTTP 5xx responses indicate server-side or service-related failures and should be investigated using failure metrics, status-code trends, server-side latency, resource logs, and SDK diagnostics. Reviewing these signals together can help determine whether failures are regional, transient, workload-specific, or associated with service processing. TTL values and unique-key policies do not explain general 5xx failures, and the number of stored procedures alone provides little diagnostic value. Production troubleshooting should correlate application logs with Cosmos DB metrics and diagnostics to establish both scope and likely cause.<\/span><\/p>\n

                                    Q18. One logical partition receives most writes even though the overall container has sufficient RU\/s. What operational condition are you experiencing?<\/b><\/p>\n

                                      \n
                                    1. <\/b> Backup saturation.<\/span><\/li>\n
                                    2. <\/b> Session-token overflow.<\/span><\/li>\n
                                    3. <\/b> Index-free querying.<\/span><\/li>\n
                                    4. <\/b> A hot partition.<\/span><\/li>\n<\/ol>\n

                                      Correct Answer: 4. A hot partition.<\/b><\/p>\n

                                      Explanation:<\/b> A hot partition occurs when request-unit consumption is concentrated on a small subset of partition-key values or physical partitions. The account may have sufficient total throughput, yet requests targeting the overloaded partition can still be throttled. This typically indicates skewed access patterns or an unsuitable partition-key strategy. Monitoring per-partition throughput helps reveal the imbalance. Backup, session tokens, and indexing terminology do not describe this condition. Hot partitions are important because Cosmos DB scalability depends not only on total RU\/s but also on effective workload distribution.<\/span><\/p>\n

                                      Q19. You want an automated notification when server-side latency exceeds an acceptable threshold. Which Azure capability should you configure?<\/b><\/p>\n

                                        \n
                                      1. <\/b> Unique key policy.<\/span><\/li>\n
                                      2. <\/b> Azure Monitor alert.<\/span><\/li>\n
                                      3. <\/b> Transactional Batch.<\/span><\/li>\n
                                      4. <\/b> TTL.<\/span><\/li>\n<\/ol>\n

                                        Correct Answer: 2. Azure Monitor alert.<\/b><\/p>\n

                                        Explanation:<\/b> Azure Monitor alerts can evaluate Cosmos DB metrics such as server-side latency and trigger notifications or automated actions when configured conditions are met. Alert rules can use action groups to send email, invoke webhooks, trigger automation, or integrate with incident-management systems. Unique keys enforce data integrity, Transactional Batch provides atomic operations, and TTL controls expiration. Proactive alerts are essential for production monitoring because they allow operators to respond before sustained latency significantly affects end-user experience.<\/span><\/p>\n

                                        Q20. Your security team needs to investigate potentially unauthorized Cosmos DB operations over the previous week. Where should operational logs be sent for centralized querying?<\/b><\/p>\n

                                          \n
                                        1. <\/b> Integrated cache.<\/span><\/li>\n
                                        2. <\/b> Analytical store only.<\/span><\/li>\n
                                        3. <\/b> TTL history.<\/span><\/li>\n
                                        4. <\/b> A Log Analytics workspace through Azure Monitor diagnostic settings.<\/span><\/li>\n<\/ol>\n

                                          Correct Answer: 4. A Log Analytics workspace through Azure Monitor diagnostic settings.<\/b><\/p>\n

                                          Explanation:<\/b> Azure Monitor diagnostic settings can route Cosmos DB resource logs to a Log Analytics workspace for centralized investigation using Kusto Query Language. This supports security analysis, troubleshooting, auditing, and correlation with logs from other Azure resources. Integrated cache is a performance feature, analytical store supports analytics over application data, and TTL does not maintain a general security audit history. Centralized logging is particularly useful when investigating suspicious access patterns or failures across multiple resources and time periods.<\/span><\/p>\n

                                           <\/p>\n","protected":false},"excerpt":{"rendered":"

                                            View Full Microsoft DP-420 Exam Dumps and Practice Test Dumps. Q1. You are designing a Cosmos DB workload where one set of queries is optimized by customer ID and another independent set is optimized by product ID. No single partition key satisfies both access patterns. What should you consider? Use one constant partition key […]<\/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\/16504"}],"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=16504"}],"version-history":[{"count":1,"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/posts\/16504\/revisions"}],"predecessor-version":[{"id":16576,"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/posts\/16504\/revisions\/16576"}],"wp:attachment":[{"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/media?parent=16504"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/categories?post=16504"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.examlabs.com\/certification\/wp-json\/wp\/v2\/tags?post=16504"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}