Resolve High EFS I/O Latency for SAP Transport Directory
A company has deployed SAP on AWS with a shared file system using Amazon EFS for the /sapmnt directory. The EFS file system is mounted on all SAP application servers. Recently, users reported that the SAP system is slow when accessing reports that are stored as PDF files in the /sapmnt directory. The operations team checks the CloudWatch metrics and finds that the EFS file system's PercentIOLimit metric is consistently at 100% during peak hours. The file system is using the General Purpose performance mode. The team also notices that the BurstCreditBalance is decreasing over time. What action should be taken to resolve this performance issue?
Quick Answer
This scenario is describing EFS bursting throughput mode running out of headroom. In bursting mode, baseline throughput and the rate at which burst credits accumulate both scale with the size of the file system, roughly 50 MiB/s of baseline throughput per TB of storage, so a smaller file system earns credits slowly and has a low baseline, which is exactly why PercentIOLimit sits at 100% during peak hours and BurstCreditBalance keeps declining: the workload is consistently spending credits faster than the file system earns them. Increasing the size of the EFS file system raises both of those numbers at once, giving the file system a higher baseline throughput and a faster credit accumulation rate, which directly relieves the I/O ceiling causing the slowdown and lets the burst credit balance recover over time. The other options don't solve the underlying mechanic: lifecycle management only moves cold files to Infrequent Access storage, which doesn't improve I/O performance and can add latency; performance mode is fixed at file system creation and can't be changed afterward; and enabling Provisioned Throughput, while it guarantees a fixed level of throughput independent of credits, doesn't rebuild the burst credit balance the way increasing file system size does, so it doesn't match the specific credit-exhaustion pattern described here. The pattern to remember is that a consistently maxed-out PercentIOLimit paired with a declining BurstCreditBalance on a General Purpose EFS file system points to insufficient size for the workload's sustained throughput needs.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Increase the size of the EFS file system to improve performance.
Increasing the size of the EFS file system increases the baseline throughput (which scales at 50 MiB/s per TB of storage) and also increases the rate at which burst credits are earned, helping to reduce the percent I/O limit and rebuild the burst credit balance. Option A is wrong because lifecycle management only moves files to Infrequent Access, which does not improve I/O performance and may add latency. Option C is wrong because enabling Provisioned Throughput does not increase the burst credit balance; it provides a fixed throughput independent of credits, but the wording is misleading. Option D is wrong because the EFS performance mode cannot be changed after creation; it is set when the file system is created.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Enable EFS lifecycle management to move infrequently accessed files to Infrequent Access storage class.
Why it's wrong here
Incorrect. Enabling lifecycle management moves infrequently accessed files to the Infrequent Access storage class, which does not solve the I/O bottleneck and may increase latency.
- ✓
Increase the size of the EFS file system to improve performance.
Why this is correct
Correct. Increasing the file system size increases baseline throughput (50 MiB/s per TB) and burst credit accumulation, directly addressing the performance issue.
- ✗
Increase the burst credit balance by enabling throughput mode.
Why it's wrong here
Incorrect. Enabling Provisioned Throughput provides a fixed throughput but does not increase the burst credit balance; the option is misleading and does not accurately describe the solution.
- ✗
Switch the EFS performance mode to Max I/O to get higher throughput.
Why it's wrong here
Incorrect. The EFS performance mode cannot be changed after the file system is created; it is only selectable at creation time.
Visual reference
Quick reference
AWS S3 Storage Class Comparison
| Storage Class | Min Duration | Retrieval | Use Case |
|---|---|---|---|
| S3 Standard | None | Immediate | Frequently accessed data |
| S3 Standard-IA | 30 days | Immediate | Infrequent access, rapid retrieval |
| S3 One Zone-IA | 30 days | Immediate | Non-critical infrequent data |
| S3 Intelligent-Tiering | None | Immediate–hours | Unknown or changing access patterns |
| S3 Glacier Instant | 90 days | Milliseconds | Archive with instant retrieval |
| S3 Glacier Flexible | 90 days | Minutes–hours | Archive, flexible retrieval |
| S3 Glacier Deep Archive | 180 days | Hours | Long-term compliance archive |
Go deeper
Related to this question
About these practice questions
Courseiva writes every PAS-C01 question from scratch — 1,616 in total, each with an explanation and a wrong-answer breakdown. None are copied from real exams or dumps. Learn why practice questions differ from exam dumps →
Same concept, more angles
2 more ways this is tested on PAS-C01
These questions test the same concept from different angles. Work through them to make sure you can recognise it however the exam phrases it.
Variation 1. A company is running SAP ERP on AWS with a multi-AZ deployment using Amazon EFS for shared transport directories. The system administrator notices that the SAP transport directory mounted via EFS is experiencing high latency during peak hours. The EFS file system is using the Standard storage class. Which combination of steps should the administrator take to reduce latency? (Choose TWO.)
hard- A.Use EFS Bursting Throughput mode.
- ✓ B.Move the EFS file system to the One Zone storage class.
- C.Change the storage class to EFS Standard-IA.
- D.Enable provisioned throughput on the EFS file system.
- ✓ E.Switch the EFS performance mode to Max I/O.
Why B: Moving the EFS file system to the One Zone storage class places the data in the same Availability Zone as the EC2 instances, reducing network latency for the SAP transport directory. Option E is correct because switching the performance mode to Max I/O increases the parallel processing capability of the file system, which reduces latency during peak hours. Option A is incorrect because Bursting Throughput mode is the default and does not specifically address high latency; it provides a baseline throughput with bursting credits. Option C is incorrect because Standard-IA is an infrequent access storage class with higher retrieval latencies, unsuitable for a continuously used transport directory. Option D is incorrect because provisioned throughput increases throughput capacity but does not directly reduce latency; Max I/O performance mode is the intended solution for latency reduction.
Variation 2. A company runs SAP on AWS and uses a shared file system via Amazon EFS for transport files. Recently, the SAP system experienced slowness when importing transports. The CloudWatch metrics show high BurstCreditBalance for the EFS file system. What action should be taken to improve performance?
hard- ✓ A.Enable Max I/O performance mode on the EFS file system.
- B.Change the performance mode to General Purpose.
- C.Enable encryption at rest to improve data transfer speed.
- D.Increase the BurstCreditBalance by purchasing additional credits.
Why A: When BurstCreditBalance is high, it indicates that the file system is not using its burst credits, meaning it is operating at baseline throughput. To improve performance, enable Max I/O performance mode on the EFS file system, which provides higher throughput and IOPS. Option B is incorrect because changing to General Purpose mode does not increase throughput; it may even reduce performance. Option C is incorrect because encryption does not affect throughput. Option D is incorrect because BurstCreditBalance cannot be purchased; it is automatically accrued and consumed.
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This PAS-C01 practice question is part of Courseiva's free Amazon Web Services certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the PAS-C01 exam.