Skip to content

Commit 49acc35

Browse files
vgs-msftnormesta
andauthored
Apply suggestions from code review
addressing changes from Norm's review. Co-authored-by: Norm Estabrook <[email protected]>
1 parent 3aac171 commit 49acc35

File tree

1 file changed

+36
-18
lines changed

1 file changed

+36
-18
lines changed

articles/storage/blobs/lifecycle-management-performance-characteristics.md

Lines changed: 36 additions & 18 deletions
Original file line numberDiff line numberDiff line change
@@ -13,42 +13,60 @@ ms.custom: references_regions, engagement-fy23
1313
# Lifecycle management performance characteristics
1414

1515

16-
Azure Blob Storage Lifecycle Management (LCM) helps you automate the transition of objects to lower-cost access tiers, or to delete objects at the end of their lifecycle to reduce storage costs. If a policy is active, Lifecycle Management periodically processes the objects in a storage account that match the filter set and rules specified in the lifecycle policy. The time required for a policy to complete processing a storage account depends on many factors. Factors like the number of objects to be evaluated and operated on, customer workloads on the account, availability of storage resources and more. In some cases, it can take multiple days to complete processing all the objects in the storage account. And a new LCM run begins only after the ongoing run completes. LCM is constantly optimizing for speed of execution while keeping these factors in mind.
16+
Azure Blob Storage lifecycle management helps you automate the transition of objects to lower-cost access tiers, or to delete objects at the end of their lifecycle to reduce storage costs. If a policy is active, Lifecycle Management periodically processes the objects in a storage account that match the filter set and rules specified in the lifecycle management policy. The time that is required for a policy to complete processing objects in a storage account depends on many factors such as the number of objects to be evaluated and operated on, the customer workloads on the account, the availability of storage resources and more. In some cases, it can take multiple days to finish processing all the objects in the storage account. A new lifecycle management policy run begins only after the ongoing run completes. Lifecycle management is constantly optimizing for speed of execution while keeping these factors in mind.
17+
18+
19+
## Factors influencing lifecycle management performance
20+
21+
There are many factors that influence the execution time that lifecycle management would take to process the objects in a single storage account.
1722

18-
## Factors Influencing Lifecycle Management Performance
19-
There are many factors that influence the execution time that Lifecycle Management would take to process the objects in a single storage account.
2023

2124
Lifecycle Management processes the subset of objects in the storage account that users scope by creating filter sets (prefix or file path) using lifecycle policies. If there's no specific scope or it's too broad, Lifecycle Management may need to process a large number of objects. This could increase the time required for the policy to complete. If a significant percentage of objects meet the policy conditions, it could also increase the processing time. This increase in processing time is especially true the first time the LCM policy is enabled on the storage account.
2225

23-
All requests made to storage account, including requests made by Lifecycle policy runs, accrue to the same limit on requests per second. As that request limit is approached, priority is given to requests made by customer workloads. The latency of processing objects also increases if the scalability and performance limits of the storage account are reached. This latency in LCM should be considered when deciding operations that need to be performed within specific time frames. Learn more about [Scalability and performance targets for standard storage accounts](../common/scalability-targets-standard-account.md).
26+
All requests that are made to a storage account, including requests that are made by lifecycle management policy runs, accrue to the same limit on requests per second. As that request limit is approached, priority is given to requests that are made by customer workloads. The latency of processing objects also increases if the scalability and performance limits of the storage account are reached. Consider this latency when deciding operations that need to be performed within specific time frames. Learn more about [Scalability and performance targets for standard storage accounts](../common/scalability-targets-standard-account.md).
27+
28+
29+
By prioritizing customer workloads, lifecycle management ensures that these workloads run with minimal to no interruptions. However, it can impact the rate of objects processed by lifecycle management. In such cases, it is possible that the rate of object creation and modification is higher than the rate at which lifecycle management can process the storage account.
2430

25-
By prioritizing customer workloads, Lifecycle Management ensures that these workloads run with minimal to no interruptions. However, it can impact the rate of objects processed by Lifecycle Management. In such cases, it is possible that the rate of object creation and modification is higher than the rate at which Lifecycle Management can process the storage account.
2631

2732
Policy conditions are assessed on each object only once during a policy run. In some cases, an object might meet the condition after it was already assessed by a run. Such objects are processed in subsequent runs.
2833

29-
Any of these conditions could cause Lifecycle Management to take multiple days to complete processing objects in a storage account. This performance characteristic of LCM could result in objects taking longer to process and lead to an increase in the storage capacity.
34+
Any of these conditions could cause lifecycle management to take multiple days to complete processing objects in a storage account. This performance characteristic of lifecycle management could result in objects taking longer to process and could lead to an increase in the storage capacity.
35+
36+
37+
## Best practices to improve lifecycle management performance
38+
39+
40+
### Narrow the scope of the lifecycle management policy
3041

31-
## Best Practices to Improve Lifecycle Management Performance
42+
In cases where lifecycle management is taking a long time to complete a run, consider applying filter sets to narrow the scope of the search and evaluation. You can do this by adding prefixes and/or blob Index tags while authoring the lifecycle management policy rules. A narrowed scope enables lifecycle management to optimize the operations.
3243

33-
### Narrow the Scope of the Lifecycle Policy
34-
In cases where Lifecycle Management is taking a long time to complete a run, it's advisable to apply filter sets to narrow the scope of the search and evaluation. This can be done by adding Prefixes and/or Blob Index tags while authoring the lifecycle rules. A narrowed scope enables Lifecycle Management to optimize the operations.
3544
> [!TIP]
36-
> Use [Azure Storage Copilot](/azure/copilot/improve-storage-accounts#reduce-storage-costs) to help configure a Lifecycle Management policy.
45+
> Use [Azure Storage Copilot](/azure/copilot/improve-storage-accounts#reduce-storage-costs) to help configure a lifecycle management policy.
3746
38-
### Optimize for Storage and Transactions costs
3947

40-
It may be more cost effective for the small objects to stay in their current tier, than paying transaction costs to move them to cheaper storage tiers. Avoid moving small files to lower tiers, unless required. [Learn more](access-tiers-best-practices.md) about choosing the right storage tier.
48+
### Optimize for storage and transactions costs
49+
50+
51+
It might be more cost effective for the small objects to stay in their current tier, instead of paying transaction costs to move them to cheaper storage tiers. Avoid moving small files to lower tiers, unless you must. [Learn more](access-tiers-best-practices.md) about choosing the right storage tier.
52+
4153
> [!TIP]
4254
> [Azure Storage Actions](../../storage-actions/overview.md) supports size-based object targeting, and tiering/deleting operations.
4355
44-
### Set appropriate Time-based Rules
45-
Avoid policy conditions using a short duration between object Creation/Modification/Last Access time and the intended operation by the policy. Lifecycle Management can take up to 24 hours to begin processing once the prior run is completed. Policy changes and updates can also take up to 24 hours to go into effect. This includes deleting all the rules to make a policy inactive. Policies that take multiple days to complete may not operate on objects evaluated earlier in the run but meet the conditions over the run period.
56+
### Set appropriate time-based rules
57+
58+
Avoid policy conditions that use a short duration between object creation, modification or last access time, and the intended operation by the policy. Lifecycle management can take up to 24 hours to begin processing after the prior run is completed. Policy changes and updates can also take up to 24 hours to go into effect. This includes deleting all the rules to make a policy inactive. Policies that take multiple days to complete might not operate on objects that were evaluated earlier in the run even though they meet the conditions over the run period.
59+
60+
61+
### Be aware of scalability and performance limits
62+
63+
The request rate and bandwidth of your storage account depend on object size, access patterns, and workload type. lifecycle management might experience a slower rate of processing objects during high traffic workloads. If you consistently notice the storage account reaching the account limits and a slowdown in lifecycle management processing, request an increase in account limits. Lifecycle management performance might improve based on the resource allocation and by workload prioritization. To request an increase contact [Azure Support](https://azure.microsoft.com/support/faq/) .
64+
65+
66+
### Set up troubleshooting and monitoring
4667

47-
### Be Aware of Scalability and Performance limits
48-
The request rate and bandwidth of your storage account depend on object size, access patterns, and workload type. Lifecycle Management may experience a slower rate of processing objects during high traffic workloads. If you consistently notice the storage account reaching the account limits and a slowdown in Lifecycle management processing, request an increase in account limits. Lifecycle management performance may improve based on the resource allocation by workload prioritization. To request an increase contact [Azure Support](https://azure.microsoft.com/support/faq/) .
68+
You should periodically evaluate the performance of your lifecycle management policy. Set up [Event Grid Notifications](../../event-grid/blob-event-quickstart-portal.md) to get notified on when a lifecycle management policy run completed. Use storage resource logs in [Azure Monitor](monitor-blob-storage.md) to dive into run details.
4969

50-
### Set up Troubleshooting and Monitoring
51-
Periodically evaluating the performance of your Lifecycle Management Policy is recommended. Set up [Event Grid Notifications](../../event-grid/blob-event-quickstart-portal.md) to get notified on when a Lifecycle Management run completed. Use Storage Logs in [Azure Monitor](monitor-blob-storage.md) to dive into run details.
5270

5371
## Next Steps
5472
- [Blob Storage FAQ](storage-blob-faq.yml)

0 commit comments

Comments
 (0)