Microsoft has added the ability to run Azure AI services on LUIS. Will limits (like 10k calls per month) be applicable for the LUIS container service as well or is it that there would be no limits with containers running ?
There is no limit as I am using S0 plan for Luis and there is no limit pay as you go.
find the description below for container billing.
The LUIS container sends billing information to Azure, using a Language Understanding resource on your Azure account.
Queries to the container are billed at the pricing tier of the Azure resource used for the .
Cognitive Services containers are not licensed to run without being connected to Azure for metering. Customers need enable the containers to communicate billing information with the metering service at all times. Cognitive Services containers do not send customer data (e.g., the image or text that is being analyzed) to Microsoft. The container reports usage about every 10 to 15 minutes.
for more information check the here
Related
On Azure's payment estimator, I can choose the number of hours that I want an app service to be active/paid for. As my app isn't required out of office hours or at weekends, I could reduce the number of hours and therefore the cost estimate considerably versus paying 24/7. I therefore used the estimator to come up with an (affordable for my organisation) estimate of having the services running for max of 500 hours a month.
What is not clear is how, now I have app services up and running on Azure, do I set the services to run on a schedule that matches the cost estimator? Is there an option or functionality hidden away somewhere to do this?
There is no option to suspend an azure app service plan. You can stop/start a web app running on a plan but that won't save any costs. I agree it is confusing that the pricing calculator makes it seem that you can suspend a web app plan.
The only thing I can think of is to scale down to a free tier plan during out-of-office/weekend hours and scale up to a paid plan but that is limited to some basic tiers (for example, since there are no slots in some lower tiers you will have problem if you want to use those). You will have to script this yourself.
The other option is to delete the whole app service plan and web app and create it / deploy again when needed. You can automate the creation using a bicep or ARM template.
References
https://learn.microsoft.com/en-us/answers/questions/278494/can-we-stop-azure-app-service-to-save-cost
Automate tier scaling
App Service unlike Azure VMs don't have the ability to pause billing. If the App Service exists, it is billing. Please see the below recommendations to control costs.
Optimize costs
At a basic level, App Service apps are charged by the App Service plan that hosts them. The costs associated with your App Service deployment depend on a few main factors:
Pricing tier Otherwise known as the SKU of the App Service plan. Higher tiers provide more CPU cores, memory, storage, or features, or combinations of them.
Instance count dedicated tiers (Basic and above) can be scaled out, and each scaled out instance accrues costs.
Stamp fee In the Isolated tier, a flat fee is accrued on your App Service environment, regardless of how many apps or worker instances are hosted.
An App Service plan can host more than one app. Depending on your deployment, you could save costs hosting more apps on one App Service plans (i.e. hosting your apps on fewer App Service plans).
Source: https://learn.microsoft.com/en-us/azure/app-service/overview-manage-costs
Is there any way to get Cost and Utilization(CPU, Memory, DTU) trend of azure resources by using azure cost management apis.
If you check this Azure Cost Management documentation, then you will find that -
The Azure Cost Management APIs provide the ability to explore cost and usage data by creating customized filters and expressions allowing you to answer consumption-related questions for your Azure resources.
But note that these APIs are currently available for Azure Enterprise customers. So you can use it if you have the Enterprise Subscription.
To solve your problem you can use the Azure Consumption APIs. These APIs support Enterprise Enrollments and also some Web Direct Subscriptions. The Azure Consumption APIs give you programmatic access to cost and usage data for your Azure resources.
You should use Balances API, to get a monthly summary of information on balances, new purchases, Azure Marketplace service charges, adjustments, and overage charges.
And use Budgets API, to to create either cost or usage budgets for resources, resource groups, or billing meters.
Use the Usage Details API to get charge and usage data for all Azure 1st party resources you have.
Currently Balances API and Budgets API is only for Enterprise Customers only.
I have been following the learning path for Microsoft Azure AI 900. In the second module, I have deployed my model as an endpoint. It says Container instances for compute type. How much will this cost me. Azure doesn't seem to show any pricing for this. Is this endpoint always active? If yes how much does it cost?
The price depends on the number of vCPU and GBs of memory requested for the container group. You are charged based on the vCPU request for your container group rounded up to the nearest whole number for the duration (measured in seconds) your instance is running. You are also charged for the GB request for your container group rounded up to the nearest tenths place for the duration (measured in seconds) your container group is running. There is an additional charge of $0.000012 per vCPU second for Windows software duration on Windows container groups. Check here Pricing - Container Instances | Microsoft Azure for details
After Deployed the Azure Machine Learning managed online endpoint (preview).
Have at least Billing Reader access on the subscription where the endpoint is deployed
To know the costs estimation
In the Azure portal, Go to your subscription
Select Cost Analysis for your subscription.
Create a filter to scope data to your Azure Machine learning workspace resource:
At the top navigation bar, select Add filter.
In the first filter dropdown, select Resource for the filter type.
In the second filter dropdown, select your Azure Machine Learning workspace.
Create a tag filter to show your managed online endpoint and/or managed online deployment:
Select Add filter > Tag > azuremlendpoint: "< your endpoint name>"
Select Add filter > Tag > azuremldeployment: "< your deployment name>".
Refer here for more detailed steps
If I create an Azure cloud services (implemented using .NET Core in my case), is there a way to get an estimation of the resources needed for the service BEFORE (not analyzing it after the fact) it is deployed to the cluster? By resources I mean number of cpus needed, memory being used etc.
Before deployment any application We can use Azure pricing calculator to analysis our uses and cost as per our requirement .
And alternatively based on the MS DOC we can review estimated costs according to our requirement tool in the Azure portal:
When you create an App Service app or an App Service plan, you can see
the estimated costs.
To create an app and view the estimated price:
On the create page, scroll down to App Service plan, and click Create new.
Specify a name and click OK.
Next to Sku and size, click Change size.
Review the estimated price shown in the summary. The following screenshot is an example and doesn't reflect current pricing.
For more information you can refer this MS DOC: Sizes for Cloud Services
I'm trying to understand the correct way when hosting a web service using Windows Azure.
After reading some of the documentation available, I have reached these lines:
Windows Azure takes the following actions if a subscription's resource usage quotas are exceeded in a quota interval (24 hours):
Data Out - when this quota is exceeded, Windows Azure stops all web sites for a subscription which are configured to run in Shared mode for the remainder of the current quota interval. Windows Azure will start the web sites at the beginning of the next quota interval.
CPU Time - when this quota is exceeded, Windows Azure stops all web sites for a subscription which are configured to run in Shared mode for the remainder of the current quota interval. Windows Azure will start the web sites at the beginning of the next quota interval.
I was always under the impression that using cloud solution will prevent such events, as I really don't know a head of time what needs my web service will have, and that the cloud will provide the resources as needed (and off-course I will be charged for them) -
is that assumption is wrong?
EDIT
I found this great post that really explains Azure perfectly
Scott Hanselman - my own Q&A about Azure Websites and Pricing
If you are hosting the Windows Azure Website in the Shared mode, although you are paying, there are certain quotas that are in place because in the background you are basically sharing the resources with other websites which are hosted on the same Virtual Machine.
If you are hosting using the Standard mode, then you no longer have quotas and you will not experience this issue. As an added bonus, you can now setup Autoscale to automatically scale out your website under load.
Azure provides you different scalability levels according to the method of hosting you pick. For example if you host your web service on an azure web site you can't scale to thousands of servers. If you host your web services in a cloud service you can scale much further.
In Azure the scalability does not always happen transparently. In the case of a web service your choices are "azure web sites", "azure mobile services" and "azure cloud services". None of these will provide transparent scalability. You will need to define how you want scalability to be processed by azure. Most of the time you can do it in your azure management portal and define "Auto-Scaling" based on your pre-defined metrics as in "total amount of memory used" or "compute power used". Azure helps you gather metrics from a distributed environment, define scaling rules and scale without worrying about the underlying infrastructure but you will need to glue these pieces together as it defines how much you will get billed as well.
Hope this makes sense.