Blogs by Ashwin

  Introduction Containers allow you to run Azure AI services either in your local environment or within Azure. For instance, if your application relies on sensitive information stored in an on-premises SQL Server to access an Azure AI service, you can implement Azure AI services in containers on the same network. This ensures that your data remains within your local network instead of being transferred to the cloud. Furthermore, deploying Azure AI services in a local container will reduce latency between the service and your data, which can enhance performance. Understand Containers When you launch a software service, it needs to be placed in an environment that supplies the necessary hardware, operating system, and supporting runtime components required by the service. Azure AI services are offered as a cloud service, where the service software is maintained within an Azure data center that supplies the core runtime services, operating system, and hardware. Additionally,...

  Introduction Diagnostic logging allows you to collect detailed operational data for an Azure AI services resource, which can be utilized to assess service usage and resolve issues. Create Resources for Diagnostic Log Storage To collect diagnostic logs for an AI services resource, you must establish a destination for the log data. Azure Event Hubs can serve as a destination, allowing you to forward the data to a custom telemetry solution, and it can also connect directly to certain third-party solutions. However, in many instances, you will utilize one (or both) of the following types of resources within your Azure subscription: Azure Log Analytics - a service that enables you to query and visualize log data within the Azure portal. Azure Storage - a cloud-based data store that you can use to store log archives (which can be exported for analysis in other tools as needed). It is essential to generate these resources prior to setting up diagnostic logging for your AI service...

  Introduction Through the development of alert rules, Microsoft Azure offers resource alerting support. You can set up alerts and notifications for your resources based on events or metric thresholds by using alert rules. When an issue occurs, these alerts will make sure that the right personnel is informed. Alert Rules Choose the Azure AI services resource in the Azure portal, then build a new alert rule under the Alerts tab. You must provide the following in order to define the alert rule: The scope of the alert rule, in other words, the resource you want to monitor. A condition on which the alert is triggered. The specific trigger for the alert is based on a signal type, which can be Activity Log (an entry in the activity log created by an action performed on the resource, such as regenerating its subscription keys) or Metric (a metric threshold such as the number of errors exceeding 10 in an hour). Optional actions, such as sending an email to an administrator notifying t...

  Overview Azure AI services allow you to incorporate artificial intelligence into your applications and services. Monitoring Azure AI Services is essential for tracking usage, identifying trends, and recognizing and resolving issues. Azure AI services offer a cloud-driven platform for integrating artificial intelligence features into your applications. As with any software service, it is important to oversee AI services to manage expenses, recognize utilization patterns, and identify possible problems. Monitor Cost One significant advantage of utilizing cloud services is the ability to achieve cost savings by only covering expenses for the services you actually use. Certain resources within Azure AI services provide a free tier with usage limitations, making it beneficial for development and testing; in addition, there are one or more paid tiers that result in charges based on transactions. The exact billing rate varies depending on the type of resource. Plan costs for AI services...

  About Network security plays a crucial role in preventing unauthorized individuals from accessing the services you aim to protect. Restricting user visibility is always a smart strategy, as they cannot jeopardize what they cannot observe. Apply network access restrictions By default, all networks can access Azure AI services. Certain individual resources within AI services (like Azure AI Face service, Azure AI Vision, and others) can be set up to limit access to designated network addresses, whether those are public Internet addresses or addresses within virtual networks. When network restrictions are in place, a client attempting to connect from an unauthorized IP address will encounter an Access Denied error. Secure key Access With Azure Key Vault Using a key for authentication, you can create apps that use Azure AI services. But this implies that the key must be obtainable by the application code. One choice is to keep the key in a configuration file or environment variable wh...

  Token-based Authentication Certain AI systems support (or even require) token-based authentication when utilizing the REST interface. In these situations, the initial request for an authentication token, which has a 10-minute validity period, presents the subscription key. To confirm that the caller has been authenticated, the token must be presented in subsequent requests. Note- When using an SDK, the calls to obtain and present a token are handled for you by the SDK. Microsoft Entra ID Authentication You may give access to particular service principals or managed identities for apps and services operating in Azure due to the support of Azure AI services for Microsoft Entra ID authentication.  A cloud-based identification and access management system is called Microsoft Entra ID. There are different ways you can authenticate against Azure AI services using Microsoft Entra ID: Authenticate Using Service Principals The overall process to authenticate against Azure AI services...