Cale and Sujit discuss their current projects in Azure as 2024 comes to a close. They also cover a ton of AKS updates.

Semantic Kernel makes it easier for developers to build Azure Open AI applications that can also include SLMs like Phi-4.

Azure has many options to use File Shares and Volumes, and we walk through the process of figuring out which one is right for your needs.

Media file: https://azpodcast.blob.core.windows.net/episodes/Episode511.mp3

YouTube: https://youtu.be/dLfCJ6btKng

Resources:

Semantic Kernel - https://github.com/microsoft/semantic-kernel

• Journey with SK on OpenAI and AzureOpenAI
• Ollama (running SLM local) - https://github.com/ollama/ollama
• Ollamazure (running SLM that looks like Azure OpenAI) - https://github.com/sinedied/ollamazure
• PhiSilica - https://learn.microsoft.com/en-us/windows/ai/apis/phi-silica

File Shares: https://learn.microsoft.com/en-us/azure/storage/common/storage-introduction

Other updates:

Lots of AKS updates!

https://learn.microsoft.com/en-us/azure/aks/concepts-network-isolated

https://learn.microsoft.com/en-us/troubleshoot/azure/azure-kubernetes/availability-performance/container-image-pull-performance

https://learn.microsoft.com/en-us/azure/aks/imds-restriction

https://learn.microsoft.com/en-us/azure/aks/use-windows-gpu

https://azure.microsoft.com/en-us/updates/?id=471295

https://learn.microsoft.com/en-us/azure/backup/tutorial-restore-aks-backups-across-regions

https://learn.microsoft.com/en-us/azure/aks/app-routing-nginx-configuration?tabs=azurecli#control-the-default-nginx-ingress-controller-configuration-preview

https://learn.microsoft.com/en-us/azure/aks/automated-deployments

https://learn.microsoft.com/en-us/azure/aks/aks-extension-ghcopilot-plugins

https://learn.microsoft.com/en-us/azure/azure-monitor/containers/container-insights-logs-schema#kubernetes-metadata-and-logs-filtering

The team meets with Stephane Erbrech to talk about Azure Kubernetes Fleet and learns why it is necessary and also why it isn't the same thing as Azure Kubernetes Service (AKS).

Media file: https://azpodcast.blob.core.windows.net/episodes/Episode510.mp3

YouTube: https://youtu.be/h9rJNihhZPM

Resources:

The OSS bits powering Fleet Manager dataplane: https://github.com/Azure/fleet

The service doc: https://learn.microsoft.com/en-gb/azure/kubernetes-fleet/

Other updates:

https://www.linkedin.com/posts/cynthiakreng_nbafoundation-nbaallstar-nbafpitch-activity-7269908869684830209-MJlS

In this episode of the Azure Podcast, co-hosts, Sujit D'Mello, Evan Basalik, Russell Young, Cynthia Kreng, and Cale Teeter welcome special guest Thiago Almeida from the Azure Functions team. They dive into the details of the new Azure Functions Flex Consumption Plan, discussing its features, benefits, and how it fits into the broader landscape of Azure's compute options. Thiago shares insights on the programming model, pricing, and the improvements in cold start times, as well as the integration with networking features. This episode is a must-listen for anyone interested in serverless computing and Azure Functions.

Media file: https://azpodcast.blob.core.windows.net/episodes/Episode509.mp3

YouTube: https://youtu.be/WMdzxYy_tQ4

Resources:

Ignite news from Functions https://techcommunity.microsoft.com/blog/appsonazureblog/azure-functions-ignite-2024-update/4298757

Flex Consumption GA announcement https://techcommunity.microsoft.com/blog/appsonazureblog/azure-functions-flex-consumption-is-now-generally-available/4298778

Azure Functions cold start improvements https://techcommunity.microsoft.com/blog/appsonazureblog/our-latest-work-to-improve-azure-functions-cold-starts/4164500

Cale, Evan, Russell and Sujit discuss the recent updates in Azure and specially cover the Azure-related announcements in the recent Ignite 2024 conference.

Media file: https://azpodcast.blob.core.windows.net/episodes/Episode508.mp3

YouTube: https://youtu.be/DwSqEMjmcn0

Resources:

https://news.microsoft.com/ignite-2024-book-of-news/

https://learn.microsoft.com/en-us/azure/mysql/flexible-server/concepts-business-continuity

https://aka.ms/configure-pon

olling upgrades with MaxSurge for Virtual Machine Scale Sets - Azure Virtual Machine Scale Sets | Microsoft Learn

https://aka.ms/vmss-zonal-expansion

Azure Container Apps planned maintenance (preview) | Microsoft Learn

Release ProcDump-for-Mac 1.0 · microsoft/ProcDump-for-Mac

Workspaces in Azure API Management | Microsoft Learn

Protect APIs in API Management with Defender for APIs | Microsoft Learn

Gabriel Monteiro Nepomuceno, a Software Engineer and Cloud architect in Microsoft's ISE team, explains the concept of Azure Verified Modules (AVM). These include both resource and pattern modules. These modules ensure consistency and compliance with best practices, making it easier for users to deploy secure and well-architected solutions. They are also supported by Microsoft, so customers get long-term stability for their deployments.

Media file: https://azpodcast.blob.core.windows.net/episodes/Episode507.mp3

YouTube: https://youtu.be/BB5HQ_VkHG0

Resources:

https://azure.github.io/Azure-Verified-Modules/

Other updates:

Empower innovation and transformation through AI and data-driven insights on Azure

https://azure.microsoft.com/en-us/blog/empower-innovation-and-transformation-through-ai-and-data-driven-insights-on-azure/

How higher education is reimagining student experiences with Azure OpenAI Service

https://azure.microsoft.com/en-us/blog/how-higher-education-is-reimagining-student-experiences-with-azure-openai-service/

https://azure.microsoft.com/en-us/updates/v2/Azure-Cobalt-100-Arm-based-Virtual-Machines

https://azure.microsoft.com/en-us/updates/v2/VM-watch-on-Azure-VMs

https://azure.microsoft.com/en-us/updates/v2/Ermetro-ga-announcement

https://azure.microsoft.com/en-us/updates/v2/ExpressRoute-guided-configuration-experience

https://azure.microsoft.com/en-us/updates/v2/Simulate-everyday-interactions-with-your-GenAI-App-Public-Preview

The team meets up with Vaclav Jirovsky and Cody Hosterman from Pure Storage to learn about Pure's CBS solution and why Azure customers should consider using it.

Media file: https://azpodcast.blob.core.windows.net/episodes/Episode506.mp3

YouTube: https://youtu.be/wCbd5vdcBPQ

Resources:

https://www.purestorage.com/products/cloud-block-storage/cbs.html

https://support.purestorage.com/bundle/m_cbs_for_azure/page/Pure_Cloud_Block_Store/CBS_for_Azure/topics/topic/t_cbs_for_azure_overview.html

Other updates:
https://powerplatformconf.com/

Willie Ahlers, Principal Architect, joins Cale, Cynthia and Russell to discuss the challenges and processes to get value from AI projects and make them successful in a live environment.

Media file: https://azpodcast.blob.core.windows.net/episodes/Episode505.mp3

YouTube: https://youtu.be/1GsuH-YylP0

• MAIVA - Microsoft AI Value Accelerator Playbook
• Data Science Toolkit - a collection of proven ML and AI implementation accelerators

Other updates:

• Public Preview: Expanded GenAI Gateway capabilities in Azure API Management
• Azure AI Model Inference API
• Public Preview: Low-Code RAG ingestion with built-in document parsing and chunking in Logic Apps Standard
• Public Preview – Templates Support in Azure Logic Apps Standard
• Public Preview – Out of box monitoring dashboards for Logic Apps Standard
• Generally Available: Java on Azure Container Apps
• Generally Available: Azure Application Gateway v2 Basic SKU
• Introducing o1: OpenAI’s new reasoning model series for developers and enterprises on Azure
• Advanced Container Networking Services: Enhancing security and observability in AKS

Sadaf Khan joins Evan and Russell to explain and talk about Service Reliability Engineering in the Azure engineering group.

Media file: https://azpodcast.blob.core.windows.net/episodes/Episode504.mp3

YouTube: https://www.youtube.com/watch?v=QNGdTnb1W90&t=1684s

• Public Preview: Customer managed planned failover for Azure Storage
• Public Preview: Instance Mix on Virtual Machine Scale Sets
• Generally Available: Workspaces in Azure API Management
• Generally Available: Azure NetApp Files storage with cool access for all service levels
• Generally Available: Larger Enterprise tier cache instances for Azure Cache for Redis
• Generally Available: Azure Red Hat OpenShift Now Supports Clusters Up to 250 Nodes

Key Topics:

• Azure Reliability SRE: Evan introduced the episode's focus on Azure reliability SRE and mentioned a special guest, Sadaf, who would provide insights on the topic. 0:19
• Azure Storage Public Preview Feature: Russell discussed a new public preview feature for Azure storage that allows customers to manage planned failovers, enhancing the service's reliability. 1:10
• Virtual Machine Scale Set Update: Russell highlighted an update to virtual machine scale sets that allows mixing different instances, improving flexibility and scalability. 1:38
• Azure API Management Workspace: Russell introduced a new feature in Azure API management that enables teams to have more autonomy in managing and publishing APIs. 2:08
• NetApp Files Storage Update: Russell mentioned the general availability of cool access for NetApp files storage, allowing for more cost-effective data storage based on access patterns. 2:40
• Redis Cache Update: Russell discussed a new tier for Redis Cache that supports larger enterprises with increased memory and compute capabilities. 3:02
• Azure Red Hat Openshift Update: Russell shared an update on Azure Red Hat Openshift, which now supports up to 250 nodes, significantly increasing scalability. 3:29
• SRE Role and Impact: Sadaf explained the role of SRE in improving service reliability and quality, detailing their engagement model with various Azure services. 4:52
• SRE Engagement and Resistance: Sadaf shared insights on the initial resistance faced from service teams during SRE engagements and how trust is built over time to allow for more impactful changes. 7:49
• SRE's Approach to Service Improvement: Sadaf outlined the SRE team's structured approach to service improvement, focusing on fundamentals, service health, operational efficiency, and scalability. 10:51
• AI Initiatives in SRE: Sadaf discussed the SRE team's initiatives in leveraging AI to analyze incident data and generate insights, aiming to reduce the cognitive load on engineers. 30:27