HCE-3700 Sample Questions & Answers
RAID, cache, and processing architecture ties with pool, tier, and workload-profile design for the top weighting, alongside storage virtualization, industry-standard and Hitachi performance tools, solution design, and troubleshooting strategy.
Launch the full HCE-3700 simulator →Showing 10 of 20 free samples.
- Question 1Advanced
Solution Design and Deployment Planning · Architectural Design Decisions
Case Study
A regional bank is consolidating its core banking and data warehouse applications onto a single Hitachi VSP 5600 array. The bank's IT department has provided detailed performance requirements and constraints.
Core Banking Application: This OLTP application runs 24/7 and is extremely latency-sensitive. It requires sub-500 microsecond read response times for 8KB random I/O blocks. During peak business hours, it generates approximately 250,000 read IOPS and 80,000 write IOPS. Data protection is critical, and any performance impact from a drive failure must be minimized.
Data Warehouse Application: This application runs large, complex queries nightly between 10 PM and 4 AM. The workload consists of large block (256KB) sequential reads, requiring a sustained throughput of at least 15 GB/s. Cost-effective capacity is a major concern for this application, as the dataset is several petabytes and growing.
Business Constraints: The bank wants to avoid application-level performance management and requires the storage array to handle workload contention automatically. They need a solution that can guarantee performance for the core banking application without manual intervention, even when the data warehouse jobs are running. The total budget for storage media is a significant constraint.
Which design provides the optimal solution to meet all performance, protection, and business requirements?
Show answer & explanation
Correct answer: C
This solution optimally addresses all requirements. 1) The all-FMD pool with RAID 5 provides excellent random read performance for the core banking OLTP workload while being more capacity-efficient than RAID 10. 2) The NL-SAS pool with RAID 6 offers the best balance of cost-effective capacity and high sequential read throughput for the data warehouse. 3) Using QoS is the key to meeting the business constraint of automatic contention management. It guarantees performance for the critical banking application and prevents the high-throughput warehouse jobs from consuming all array resources, ensuring service levels are met for both workloads without manual intervention.
- Question 2Intermediate
Performance Tools and Data Acquisition · Workload Simulation and Testing
A performance architect is using
vdbenchto validate the performance of a new VSP G370 array before it goes into production. The goal is to simulate an OLTP workload. Whichvdbenchparameter is most critical for accurately mimicking a typical OLTP I/O pattern?Show answer & explanation
Correct answer: C
OLTP (Online Transaction Processing) workloads, such as those from databases, are characterized by small, random I/O operations. The
vdbenchparameterseekpct=random(orseekpct=100) instructs the tool to perform I/O at random locations on the logical unit, which accurately simulates this behavior.seekpct=0would simulate a purely sequential workload, andrdpct=100would simulate a read-only workload, neither of which fully represents a typical OLTP profile that includes both reads and writes. - Question 3Advanced
Monitoring and Troubleshooting · Replication Performance Issues
A customer has a VSP G1000 with Global-Active Device (GAD) configured between two sites for a mission-critical application. They observe that write response times on the primary array are consistently 3-4 ms higher than read response times. The network link between the sites is a dedicated 10ms latency dark fiber connection. What is the most accurate explanation for the elevated write latency?
Show answer & explanation
Correct answer: C
Global-Active Device operates as a synchronous replication solution to provide an active-active storage cluster. For every write I/O from the host, the primary array must not only write the data to its own cache but also send it across the network link to the secondary array, which must then confirm the data has been written to its cache. Only after this remote confirmation is received can the primary array send the final acknowledgement back to the host. This round-trip network latency is therefore added to every write operation, explaining the higher response time compared to reads, which are served locally.
sequenceDiagram participant Host participant Primary Array participant Secondary Array Host->>Primary Array: Write Request Primary Array->>Primary Array: Write to Cache Primary Array->>Secondary Array: Replicate Write Secondary Array->>Secondary Array: Write to Cache Secondary Array-->>Primary Array: Acknowledge Write Primary Array-->>Host: Final ACK - Question 4Intermediate
Pools, Tiers and Workload Profiles · Dynamic Provisioning Concepts
When designing a Hitachi Dynamic Provisioning (HDP) pool, what is the primary performance trade-off of using a very small page size (e.g., 42 MB) versus a larger page size (e.g., 1 GB)?
Show answer & explanation
Correct answer: B
The page is the fundamental unit of allocation in an HDP pool. A smaller page size allows for more efficient use of capacity, as space is allocated in smaller chunks. However, this granularity comes at a cost: the system must manage a much larger number of metadata pointers, which increases the processing overhead on the controllers. Conversely, a larger page size reduces metadata overhead but can lead to wasted space if applications write small amounts of data, as the entire large page must be allocated.
- Question 5Intermediate
Storage Virtualization · Performance Optimization for Virtualization
Which of the following metrics, when observed in Hitachi Performance Monitor for a VMware datastore LUN, would most strongly indicate an I/O blender effect?
Show answer & explanation
Correct answer: B
The I/O blender effect occurs when a hypervisor combines I/O streams from multiple virtual machines onto a single datastore. Even if each individual VM is performing a large, sequential I/O pattern, the hypervisor interleaves these streams. The storage array sees the result as a single stream of small, highly random I/O. This defeats the array's prefetching and caching algorithms, leading to poor performance. Observing this pattern in Performance Monitor is a classic symptom of the I/O blender effect.
- Question 6Advanced
Monitoring and Troubleshooting · Problem Identification and Isolation
During a performance analysis of a VSP 5200, an administrator notices that the MPB (Multi-Processor Board) utilization is consistently over 90%, while front-end and back-end director utilization remains below 60%. The array is running several software-based features, including Universal Replicator (HUR) and thin provisioning (HDP). What does this pattern of utilization indicate?
Show answer & explanation
Correct answer: C
In the VSP architecture, the Multi-Processor Boards (MPBs) are responsible for executing the code for value-added software features like replication (HUR), thin provisioning (HDP), and data protection. The front-end and back-end directors handle the core I/O processing and data movement. When MPB utilization is the highest resource consumer, it indicates that the overhead from these software features is the primary performance bottleneck, rather than the physical I/O paths or storage media.
- Question 7IntermediateSelect 3
Performance Tools and Data Acquisition · Industry-Standard Performance Tools
A performance architect needs to gather end-to-end performance data for a critical application, correlating host-level metrics, SAN fabric statistics, and VSP array performance. Which THREE tools would be most appropriate for this task? (Select THREE)
Show answer & explanation
Correct answers: A, B, D
- Question 8Beginner
Storage Virtualization · Virtualized Storage Deployment
True or False: When virtualizing an external storage array behind a VSP 5000 series, the performance characteristics of the external array, including its cache size and media type, directly impact the performance experienced by hosts accessing the virtualized LUNs.
Show answer & explanation
Correct answer: A
True. While the VSP's front-end cache can absorb some I/O and improve performance, it cannot completely mask the underlying performance of the external array. I/O that results in a cache miss on the VSP must be serviced by the external system. Therefore, the latency, throughput, and IOPS capabilities of the back-end virtualized array remain a critical factor in the overall end-to-end performance.
- Question 9Advanced
Solution Design and Deployment Planning · Performance-Oriented Design
Case Study
A large e-commerce company is experiencing severe performance degradation on its order processing database during seasonal sales events. The database runs on a physical server connected to a Hitachi VSP G700. The IT team has been tasked with identifying the bottleneck and proposing a solution without purchasing a new array.
Current Environment: The database LUNs reside in an HDP pool composed entirely of 10K RPM SAS drives in a RAID 5 (8D+1P) configuration. The server is connected via two 16Gb Fibre Channel links. The VSP G700 also serves file storage and development environments from the same HDP pool.
Performance Analysis: During a sales event, Hitachi Performance Monitor shows the database LUNs experience average read latencies of 15ms. The HDP pool's back-end disks show 95% utilization. The VSP's front-end processors and cache utilization are moderate, with a read cache hit ratio of 30%. The development environment VMs occasionally run resource-intensive builds, causing further latency spikes.
Requirements: The solution must reduce read latency for the database to under 2ms during peak times. The solution must isolate the database's performance from the other workloads. No new storage arrays can be purchased, but adding drives to the existing array is permitted.
Which proposal is the most effective and targeted solution?
Show answer & explanation
Correct answer: B
This is the most comprehensive solution. The performance analysis clearly points to a back-end disk bottleneck (95% utilization on SAS drives) for a latency-sensitive workload. 1) Creating a new pool with FMDs directly addresses the latency requirement, as flash media provides sub-millisecond response times. 2) Placing this pool on a separate set of physical resources isolates the critical database from the 'noisy neighbor' workloads (file storage, dev VMs), fulfilling a key requirement. 3) Using QoS adds another layer of protection. Migrating to RAID 10 on SAS would help but not meet the <2ms goal, and upgrading FC links is pointless as the bottleneck is on the back end.
- Question 10Beginner
Monitoring and Troubleshooting · Problem Identification and Isolation
A storage administrator is analyzing a performance issue where a single host is experiencing high latency, but all other hosts connected to the same VSP G350 front-end ports are performing normally. The array-level performance metrics show no signs of stress. What is the most likely component to investigate first?
Show answer & explanation
Correct answer: C
When a performance problem is isolated to a single host and the shared storage resources (array processors, cache, disks) show no issue, the troubleshooting focus must shift to the components unique to that host's I/O path. This includes the host's HBA card, its cabling, the multipathing software configuration (e.g., HDLM, PowerPath), and OS-level settings like queue depth. A misconfiguration or fault in any of these components can cause latency for that host without affecting others.
Ready for the real thing?
The full HCE-3700 simulator has every exam-style question, timed mode, and instant scoring.