1Z0-050 · Question #76
Which three statements correctly describe the features of the I/O calibration process? (Choose three.)
The correct answer is A. Only one I/O calibration process can run at a time. D. It can be used to estimate the maximum number of I/Os and maximum latency time for the system. E. The latency time is computed only when the TIMED_STATISTICS initialization parameter is set. See the full explanation below for the reasoning.
Question
Which three statements correctly describe the features of the I/O calibration process? (Choose three.)
Options
- AOnly one I/O calibration process can run at a time.
- BIt automates the resource allocation for the Automated Maintenance Tasks.
- CIt improves the performance of the performance-critical sessions while running.
- DIt can be used to estimate the maximum number of I/Os and maximum latency time for the system.
- EThe latency time is computed only when the TIMED_STATISTICS initialization parameter is set
How the community answered
(37 responses)- A70% (26)
- B11% (4)
- C19% (7)
Community Discussion
8
Prof. SaraApr 12, 2026The correct answers are A, D, and E. Oracle enforces a singleton rule on the CALIBRATE_IO procedure, meaning only one calibration job can execute at a time, so A is locked in. D is correct because the whole point of the process is to profile your storage subsystem and return max_iops and latency benchmarks that the Resource Manager can then use for planning, so think of it as a diagnostic ruler for your I/O ceiling. E catches a lot of people off guard, but latency computation depends on TIMED_STATISTICS being enabled, which is the same parameter that governs wait event timing throughout the database, so if that parameter is off, you get throughput numbers without the latency piece. B is a distractor that conflates I/O calibration with the Automated Maintenance Task framework, and C is backwards since calibration floods the system with I/O stress tests, which actually hurts concurrent workloads rather than helping them.
Lena V.Apr 12, 2026The latency point on TIMED_STATISTICS is the one I see trip up the most test-takers, so worth adding that if TIMED_STATISTICS is off you still get a result but the latency column comes back NULL, not an error, which is exactly the kind of silent failure Oracle loves to hide in distractors.
Lena V.May 15, 2026Saw this exact question on my exam last year and I almost got burned by C, because it sounds reasonable until you remember that calibration hammers your I/O subsystem with a synthetic workload, so it is the opposite of friendly to concurrent sessions. A is locked in because Oracle simply will not let you run two calibration jobs simultaneously, the procedure enforces that as a hard constraint. D is the whole point of DBMS_RESOURCE_MANAGER.CALIBRATE_IO, it gives you estimated max IOPS and max latency so you can feed real numbers into the Resource Manager plan instead of guessing. E is the one that trips people up, and it tripped me too at first, but latency is a timing measurement and Oracle will not compute it unless TIMED_STATISTICS is enabled, same logic that governs wait event timing across the board. I flagged the question, came back to it, crossed out B because calibration has nothing to do with Automated Maintenance Task scheduling, crossed out C for the reason above, and walked out with the right three.
Mei-Ling H.Apr 30, 2026Almost chose C, but calibration hurts performance while running, not improves it.
Prof. SaraMay 2, 2026Exactly right, and the sharper way to frame it for exam day is that calibration trades runtime throughput for measurement accuracy, so any answer implying a performance gain should be an instant discard.
Samuel O.May 9, 2026A, D, E are right. Only one calibration runs at a time, it measures max I/O throughput and latency, and TIMED_STATISTICS must be on for latency numbers to appear.
Yusuf A.Apr 22, 2026Okay so I been going back and forth on this one but I am pretty sure B is one of the three correct answers. My senior showed me the DBMS_RESOURCE_MANAGER.CALIBRATE_IO procedure last week and he walked me through how after you run it Oracle uses those calibration results to actually inform how it schedules the Automated Maintenance Tasks, because those tasks need to know the I/O capacity before they can allocate resources properly. The whole point of calibration is to give the database a baseline so it can make smart decisions automatically, and Automated Maintenance Tasks are a big part of what runs on its own in the background. I also picked A and D to go with it since those felt pretty solid on their own, but I honestly think people are sleeping on B because they confuse "calibration" with just a measurement tool when it feeds directly into the automation layer. Anyone else go with B here or am I reading too much into what my senior was showing me?
Samuel O.Apr 22, 2026Yusuf, your instinct on A and D is solid but the calibration procedure you are thinking of feeds into the Resource Manager I/O limits for consumer groups, not the scheduling layer for Automated Maintenance Tasks, so B does not hold up under exam scrutiny. The third correct answer is E, and once you trace how CALIBRATE_IO actually surfaces its output through MAX_IOPS and MAX_MBPS parameters rather than any maintenance task dispatcher, it clicks pretty fast.