fix broken links

Author: kaysieyu

articles/aks/best-practices-cost.md

@@ -23,8 +23,8 @@ In this article, you learn about:
 ### Evaluate SKU family
 It's important to evaluate the resource requirements of your application prior to deployment. Small development workloads have different infrastructure needs than large production ready workloads. While a combination of CPU, memory, and networking capacity configurations heavily influences the cost effectiveness of a SKU, consider the following VM types:
 
-- [**Azure Spot Virtual Machines**](/azure/virtual-machines/spot-vms.md) - [Spot node pools](https://learn.microsoft.com/en-us/azure/aks/spot-node-pool) are backed by Azure Spot Virtual machine scale sets and deployed to a single fault domain with no high availability or SLA guarantees. Spot VMs allow you to take advantage of unutilized Azure capacity with significant discounts (up to 90% as compared to pay-as-you-go prices). If Azure needs capacity back, the Azure infrastructure evicts the Spot nodes. _Best for dev/test environments, workloads that can handle interruptions such as batch processing jobs, and workloads with flexible execution time._
-- [**Ampere Altra Arm-based processors (ARM64)**](https://azure.microsoft.com/blog/now-in-preview-azure-virtual-machines-with-ampere-altra-armbased-processors/) - ARM64 VMs are power-efficient and cost effective but don't compromise on performance. With [AMR64 node pool support in AKS](./azure/aks/use-multiple-node-pools.md#add-an-arm64-node-pool), you can create ARM64 Ubuntu agent nodes and even mix Intel and ARM architecture nodes within a cluster. These ARM VMs are engineered to efficiently run dynamic, scalable workloads and can deliver up to 50% better price-performance than comparable x86-based VMs for scale-out workloads. _Best for web or application servers, open-source databases, cloud-native applications, gaming servers, and more._
+- [**Azure Spot Virtual Machines**](/azure/virtual-machines/spot-vms) - [Spot node pools](./spot-node-pool.md) are backed by Azure Spot Virtual machine scale sets and deployed to a single fault domain with no high availability or SLA guarantees. Spot VMs allow you to take advantage of unutilized Azure capacity with significant discounts (up to 90% as compared to pay-as-you-go prices). If Azure needs capacity back, the Azure infrastructure evicts the Spot nodes. _Best for dev/test environments, workloads that can handle interruptions such as batch processing jobs, and workloads with flexible execution time._
+- [**Ampere Altra Arm-based processors (ARM64)**](https://azure.microsoft.com/blog/now-in-preview-azure-virtual-machines-with-ampere-altra-armbased-processors/) - ARM64 VMs are power-efficient and cost effective but don't compromise on performance. With [AMR64 node pool support in AKS](./create-node-pools.md#arm64-node-pools), you can create ARM64 Ubuntu agent nodes and even mix Intel and ARM architecture nodes within a cluster. These ARM VMs are engineered to efficiently run dynamic, scalable workloads and can deliver up to 50% better price-performance than comparable x86-based VMs for scale-out workloads. _Best for web or application servers, open-source databases, cloud-native applications, gaming servers, and more._
 - [**GPU optimized SKUs**](/azure/virtual-machines/sizes) - Depending on the nature of your workload, consider using compute optimized, memory optimized, storage optimized, or even graphical processing unit (GPU) optimized VM SKUs. GPU VM sizes are specialized VMs that are available with single, multiple, and fractional GPUs. _[GPU-enabled Linux node pools on AKS](./gpu-cluster.md) are best for compute-intensive workloads like graphics rendering, large model training and inferencing._
 
 > [!NOTE]
@@ -35,7 +35,7 @@ It's important to evaluate the resource requirements of your application prior t
 Picking the right VM SKU, regions, number of nodes, and other configuration options can be difficult upfront. [Cluster preset configurations](./quotas-skus-regions.md#cluster-configuration-presets-in-the-azure-portal) in the Azure portal offloads this initial challenge by providing recommended configurations for different application environments that are cost-conscious and performant. The **Dev/Test** preset is best for developing new workloads or testing existing workloads. The **Production Economy** preset is best for serving production traffic in a cost-conscious way if your workloads can tolerate interruptions. Noncritical features are off by default and the preset values can be modified at any time.
 
 ### Consider multitenancy
-Clusters and infrastructure can be shared across many teams and business units within an organization, each of whom may operate one or many workloads. Shared infrastructure reduces cluster management overhead while also improving resource utilization and cluster binpacking. This has direct cost saving implications. To learn more about multitenancy on AKS and to determine if it's right for your organizational needs, see [AKS considerations for multitenancy](azure/architecture/guide/multitenant/service/aks).
+Clusters and infrastructure can be shared across many teams and business units within an organization, each of whom may operate one or many workloads. Shared infrastructure reduces cluster management overhead while also improving resource utilization and cluster binpacking. This has direct cost saving implications. To learn more about multitenancy on AKS and to determine if it's right for your organizational needs, see [AKS considerations for multitenancy](/azure/architecture/guide/multitenant/service/aks).
 
 
 ## Build cloud native applications
@@ -47,10 +47,10 @@ A lean container refers to optimizing the size and resource footprint of the con
 [Resource quotas](./operator-best-practices-scheduler.md#enforce-resource-quotas) provide a way to reserve and limit resources across a development team or project. Quotas are defined on a namespace and can set on compute resources, storage resources, and object counts. When you define resource quotas, individual namespaces are prevented from consuming more resources than allocated. This is particularly important for multi-tenant clusters where teams are sharing infrastructure.
 
 ### Use cluster start stop
-Small development and test clusters, when left unattended, can realize large amounts of unnecessary spending. Turn off clusters that don't need to run at all times using [cluster start and stop](./aks/start-stop-cluster.md?tabs=azure-cli). Doing so shuts down all system and user node pools so you aren’t paying for extra compute. All objects and cluster state will be maintained when you start the cluster again. 
+Small development and test clusters, when left unattended, can realize large amounts of unnecessary spending. Turn off clusters that don't need to run at all times using [cluster start and stop](./start-stop-cluster.md?tabs=azure-cli). Doing so shuts down all system and user node pools so you aren’t paying for extra compute. All objects and cluster state will be maintained when you start the cluster again. 
 
 ### Use capacity reservations
-Capacity reservations allow you to reserve compute capacity in an Azure region or Availability Zone for any duration of time. Reserved capacity will be available for immediate use until the reservation is deleted. [Associating an existing capacity reservation group to a node pool](./manage-node-pools.md#associate-capacity-reservation-groups-to-node-pools-preview) guarantees allocated capacity for your node pool and helps you avoid potential on-demand pricing spikes during periods of high compute demand.
+Capacity reservations allow you to reserve compute capacity in an Azure region or Availability Zone for any duration of time. Reserved capacity will be available for immediate use until the reservation is deleted. [Associating an existing capacity reservation group to a node pool](./manage-node-pools.md#associate-capacity-reservation-groups-to-node-pools) guarantees allocated capacity for your node pool and helps you avoid potential on-demand pricing spikes during periods of high compute demand.
 
 
 ## Monitor your environment and spend