You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: articles/cost-management-billing/savings-plan/purchase-recommendations.md
+14-11Lines changed: 14 additions & 11 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -21,13 +21,15 @@ The goal of our savings plan recommendation is to help you make the most cost-ef
21
21
22
22
We start by looking at your hourly and total on-demand usage costs incurred from savings plan-eligible resources in the last 7, 30, and 60 days. We determine what the optimal savings plan commitment would be for each of these hours - we apply the appropriate savings plan discounts to all your savings plan-eligible usage in each hour. We consider each one of these commitments a candidate for a savings plan recommendation. We then run hundreds of simulations using each of these candidates to determine what your total cost would be if you purchased a savings plan equal to the candidate.
23
23
24
+
The goal of these simulations is to compare each candidate's total cost ((hourly commitment * 24 hours * # of days in simulation period) + total on-demand cost incurred during the simulation period) to the actual total on-demand costs. Only candidates that result in net savings are eligible for consideration as actual recommendations. We take up to 10 of the best recommendations and present them to you. For each recommendation, we also calculate the percentage of your compute usage costs are now covered by this savings plan, and any other previously purchased reservations or savings plan. The recommendations with the greatest savings for one year and three years are the highlighted options.
25
+
24
26
Here's a video that explains how savings plan recommendations are generated.
The goal of these simulations is to compare each candidate's total cost ((hourly commitment * 24 hours * # of days in simulation period) + total on-demand cost incurred during the simulation period) to the actual total on-demand costs. Only candidates that result in net savings are eligible for consideration as actual recommendations. We take up to 10 of the best recommendations and present them to you. For each recommendation, we also calculate the percentage of your compute usage costs are now covered by this savings plan, and any other previously purchased reservations or savings plan. The recommendations with the greatest savings for one year and three years are the highlighted options.
29
31
30
-
To account for scenarios where there were significant reductions in your usage, including recently decommissioned services, we run more simulations using only the last three days of usage. The lower recommendation (between the three day and 30-day recommendations) is shared, even in situations where the 30-day recommendation might appear to provide greater savings. It gets done to ensure that we don't inadvertently recommend overcommitment based on stale data.
32
+
To account for scenarios where there were significant reductions in your usage, including recently decommissioned services, we run more simulations using only the last three days of usage. The lower recommendation (between the 3- and 30-day recommendations) is provided to you - even in situations where the 30-day recommendation might appear to provide greater savings. We do this ensure that stale date doesn't cause us to inadvertently recommend overcommitment.
31
33
32
34
Keep the following points in mind:
33
35
@@ -37,30 +39,31 @@ Keep the following points in mind:
37
39
38
40
## Recommendations in Azure Advisor
39
41
40
-
When available, a savings plan purchase recommendation can also be found in Azure Advisor. While we might generate up to 10 recommendations, Azure Advisor only surfaces the single three-year recommendation with the greatest savings for each billing subscription. Keep the following points in mind:
42
+
Azure Advisorsurfaces 1- and 3-year savings plan recommendations for each billing subscription. Keep the following points in mind:
41
43
42
-
- If you want to see recommendations for a one-year term or for other scopes, navigate to the savings plan purchase experience in Azure portal. For example, enrollment account, billing profile, resource groups, and so on. For more information, see [Permissions to buy an Azure savings plan](permission-buy-savings-plan.md).
44
+
- If you want to see recommendations for shared and resource group scopes, navigate to the savings plan purchase experience in Azure portal.
43
45
- Recommendations in Advisor currently only consider your last 30 days of usage.
44
-
- Recommendations in Advisor are only for three-year savings plans.
45
-
- If you recently purchased a savings plan or reserved instance, it can take up to five days for the purchases to affect your recommendations in Advisor and Azure portal.
46
+
- If you recently purchased a savings plan or reserved instance, it may take up to five days for the purchases to affect your recommendations in Advisor and Azure portal.
46
47
47
-
## Purchase recommendations in the Azure portal
48
+
## Recommendations in Azure portal
48
49
49
-
When available, up to 10 savings plan commitment recommendations can be found in the savings plan purchase experience in Azure portal. For more information, see [Permissions to buy an Azure savings plan](permission-buy-savings-plan.md). Each recommendation includes the commitment amount, the estimated savings percentage (off your current pay-as-you-go costs), and the percentage of your compute usage costs that would get covered by this and any other previously purchased savings plans and reservations.
50
+
Azure portal presents up to 10 savings plan commitment recommendations for each savings plan term and benefit scope. Each recommendation includes the commitment amount, the estimated savings percentage (off your current pay-as-you-go costs), and the percentage of your compute usage costs that would be covered by this savings plan, as well as any other previously purchased savings plans and reservations.
50
51
51
52
By default, the recommendations are for the entire billing scope (billing profile for MCA and enrollment account for EA). You can also view separate subscription and resource group-level recommendations by changing benefit application to one of those levels. We don't currently support management group-level recommendations.
52
53
54
+
Currently, all savings plan recommendations in Azure portal are based on a 30-day look back period.
55
+
53
56
Recommendations are term-specific, so you see the one-year or three-year recommendations at each level by toggling the term options.
54
57
55
-
The highlighted recommendation is projected to result in the greatest savings. The other values allow you to see how increasing or decreasing your commitment could affect both your savings. They also show how much of your total compute usage cost would get covered by savings plans or reservation commitments. When the commitment amount is increased, your savings might decline because you have lower utilization each hour. If you lower the commitment, your savings could also be reduced. In this case, although you have greater utilization, there are more hours where your savings plan doesn't fully cover your usage. Usage beyond your hourly commitment is charged at the more expensive pay-as-you-go rates.
58
+
The highlighted recommendation is projected to result in the greatest savings. The other values allow you to see how increasing or decreasing your commitment could affect both your savings. They also show how much of your total compute usage cost would be covered by savings plans or reservation commitments. When the commitment amount is increased, your savings might decline because you have lower utilization each hour. If you lower the commitment, your savings could also be reduced. In this case, although you have greater utilization, there are more hours where your savings plan doesn't fully cover your usage. Usage beyond your hourly commitment is charged at the more expensive pay-as-you-go rates.
56
59
57
-
## Purchase recommendations with REST API
60
+
## Recommendations API
58
61
59
62
For more information about retrieving savings plan commitment recommendations, see the saving plan [Benefit Recommendations API](/rest/api/cost-management/benefit-recommendations).
60
63
61
64
## Reservation trade in recommendations
62
65
63
-
When you trade one or more reservations for a savings plan, you're shifting the balance of your previous commitments to a new savings plan commitment. For example, if you have a one-year reservation with a value of $500, and halfway through the term you look to trade it for a savings plan, you will still have an outstanding commitment of about $250. The minimum hourly commitment must be at least equal to the outstanding amount divided by (24 * the term length in days).
66
+
When you [trade-in](reservation-trade-in.md) one or more reservations for a savings plan, you're shifting the balance of your previous commitments to a new savings plan commitment. For example, if you have a one-year reservation with a value of $500, and halfway through the term you look to trade it for a savings plan, you will still have an outstanding commitment of about $250. The minimum hourly commitment must be at least equal to the outstanding amount divided by (24 * the term length in days).
64
67
65
68
As part of the trade in, the outstanding commitment is automatically included in your new savings plan. We do it by dividing the outstanding commitment by the number of hours in the term of the new savings plan. For example, 24 times the term length in days. And by making the value the minimum hourly commitment you can make during as part of the trade-in. Using the previous example, the $250 amount would be converted into an hourly commitment of about $0.029 for a new one-year savings plan. If you're trading multiple reservations, the total outstanding commitment is used. You can choose to increase the value, but you can't decrease it.
0 commit comments