APEX Cloud Platform for Azure Expands Deployment Options
Wed, 07 Aug 2024 13:55:45 -0000
|Read Time: 0 minutes
The process of deploying and configuring IT infrastructure is a necessary evil.
Perhaps evil is too strong of a word, but I think it sometimes fits. The process of setting up the compute, storage, networking, and operating system that supports a new application can be complex and error prone. Unfortunately for IT infrastructure professionals, executives are not usually sympathetic to project delays no matter what the reason. As soon as the platform is purchased, business stakeholders become like the impatient kids on family vacation crying, “Are we there yet?” Their only concern is knowing when the new IT service will be ready to achieve their desired outcomes. Examples of these outcomes might include maximizing throughput and minimizing loss in a manufacturing plant, improving the customer experience in a retail store, or helping doctors analyze images more effectively for a rapid diagnosis in a healthcare setting.
To meet the increasing demands to deliver IT infrastructure within shrinking project timelines, a highly efficient deployment and configuration process should provide the four key benefits listed below.
Figure 1 Benefits of an efficient deployment process
- Automation – Reduce or eliminate manual tasks. This results in a repeatable, standardized, and reliable process that dramatically decreases the risk of human error and accelerates time to value. It also vastly improves deployment success rates.
- Validation – At defined stages in the process, verify that the configuration parameter values provided are fully supported and compliant with vendor best practices. The process should also include prerequisite checks and health checks to ensure that all environmental requirements are met. These tests prevent misconfiguration of critical settings that can lead to costly deployment failures, security breaches, and service disruption.
- Flexibility – Accommodate varied governance policies and IT staff skillsets by providing multiple approaches to deployment and configuration. Options include guided, wizard-driven graphical workflows and programmatic toolsets that leverage scripts or infrastructure as code (IaC).
- Monitoring – Easily track the progress of the deployment with real time execution status. Verbose logging should be readily available to help rapidly troubleshoot errors when they occur.
Today, Dell Technologies is pleased to announce the latest feature release of the Dell APEX Cloud Platform for Microsoft Azure. Enhancements to the APEX Cloud Platform Foundation Software deployment and cluster configuration automation top the list of new features that I will cover in this blog.
Choose your own cluster configuration adventure
Dell ProDeploy Services and certified Dell partners currently provide a white-glove deployment experience for the APEX Cloud Platform for Azure. You can shadow your deployment engineer to familiarize yourself with the process and learn about the implications of choosing between various configuration options. Coming soon, you will be able to install the platform yourself with the same tools used by Dell ProDeploy Services and partners. In this section, I provide an overview of the expanded deployment and cluster configuration experience.
In the previous releases of APEX Cloud Platform for Azure, the automated end-to-end deployment and cluster configuration process was a uniquely Dell experience. The deployment engineer leveraged the APEX Cloud Platform Foundation Software to progress through two stages:
- Provision Server OS – This stage was used to deploy new platforms and re-deploy existing platforms. Discover servers, install the Azure Stack HCI OS, update BIOS/firmware/drivers, setup host management networking, and bring up the Cloud Platform Manager VM.
- Deploy Cluster Configuration – Onboard the MC nodes to Azure Arc, and perform all cluster, storage, and networking configuration. This phase also included installing the APEX Cloud Platform extension and adding the cluster connection in Windows Admin Center.
The following illustration depicts the home page of the guided deployment wizard in v1.01 of the APEX Cloud Platform Foundation Software.
Figure 2 APEX Cloud Platform Foundation Software guided deployment wizard from the previous version
At each stage of the deployment process, the Dell automation performed validations and pre-checks to ensure configuration values were fully supported and all environmental requirements were met.
Figure 3 Validations performed during the Deploy Cluster Configuration stage
Beginning with Azure Stack HCI version 23H2, Microsoft transitioned to a cloud-based cluster configuration experience for all solutions in the Azure Stack HCI catalog. This new experience was designed to simplify cluster deployments at-scale across many locations. The following figure shows the first step in the Azure Stack HCI cluster configuration workflow in the Azure portal.
Figure 4 Deploy Azure Stack HCI in the Azure portal
The APEX Cloud Platform for Azure now supports choosing this Microsoft experience or continuing with the Dell experience to complete cluster configuration. Here are enhanced deployment stages using the APEX Cloud Platform Foundation Software v1.02:
- Provision Server OS – This stage is common to both the Dell and Microsoft cluster configuration experiences and is fully automated by Dell. It remains essentially unchanged from the previous release – allowing you to deploy new platforms and re-deploy existing platforms. In a new deployment scenario, the MC nodes arrive from the factory running a specialized bootstrap OS, and the Azure Stack HCI payload is staged on each node’s local storage. This Dell process automates the following steps:
- Discovery of the servers
- Installation of the Azure Stack HCI OS
- Updates of BIOS, firmware, and drivers
- Setup of host management networking
- Instantiation of the Cloud Platform Manager VM
- Register Servers – This stage also is common to both the Dell and Microsoft cluster configuration experiences and is fully automated by Dell. This new dedicated workflow onboards the MC nodes to Azure Arc. This stage is also known as Low Touch Provisioning (LTP) bootstrap. Communication from the software and MC nodes to Azure is via the Azure Resource Manager (ARM) API. The steps in this stage include:
- Use new and updated Dell validations to reduce the risk of failures. These include checks for correct networking configuration, existence of AD domain accounts, proper permissions on Azure resources, and prevention of hostname conflicts.
- Ensure installation of the Azure Arc Connected Machine Agent on all nodes.
- Confirm the status of Azure Arc extensions.
- Assign the required Azure subscription access roles to the nodes.
- Deploy Cluster Configuration – We strongly recommend continuing with the Dell APEX Cloud Platform Foundation Software to perform Azure Stack HCI cluster configuration. The Dell experience delivers end-to-end automation using the guided deployment wizard, which results in significant administrative time savings. Alternatively, you can choose the Microsoft experience and use the Azure portal or Azure Resource Manager (ARM) templates to configure Azure Stack HCI cluster, storage, and networking.
Figure 5 APEX Cloud Platform for Azure end-to-end deployment process
These updates to the deployment stages are reflected in the guided deployment wizard home page in the APEX Cloud Platform Foundation Software.
Figure 6 Updated guided deployment wizard for new release
Dell cluster configuration for the win
Having the flexibility to choose between the Dell and Microsoft cluster configuration experience is one of the benefits of APEX Cloud Platform for Azure. Both approaches help accelerate your time to value using a predictable and repeatable process. However, the Dell experience will continue to be the default and preferred deployment and cluster configuration methodology because of advanced automation and validation capabilities.
The reason the Dell cluster configuration experience saves time is because it automates more tasks. On average, Dell Engineering saved approximately one hour of time when deploying 2-node APEX Cloud Platform for Azure clusters while using the Dell cluster configuration experience as compared with the Microsoft experience. Since manual tasks are required on each cluster node, the time savings increased with larger cluster sizes. Greater levels of automation also increase reliability and decrease the risk of human errors. Here is a list of steps that are automated using Dell’s cluster configuration workflow but must be completed manually if using the Azure portal:
- Steps required on each node prior to cluster configuration:
- Rename the virtual network adapter (vNIC) to match the Network ATC intent name.
- Set NIC teaming.
- Disable unused NICs.
- Steps required after the deployment is complete:
- Install the APEX Cloud Platform extension in Windows Admin Center.
- Register Cloud Platform Manger cluster parameters on each node.
- Add the new cluster to Windows Admin Center.
- Re-enable unused NICs on each node.
The following screen is taken from the APEX Cloud Platform Foundation Software during the cluster configuration process. This software provides the ability to monitor the logging of the deployment operation in real-time. If there is a failure, you can click the icon in the upper-right corner of the screen to download the verbose logs for further investigation. Dell ProDeploy and ProSupport technicians can also remotely access the Cloud Platform Manager VM and cluster nodes for in-depth troubleshooting to reduce the mean time to resolution of any issues.
Figure 7 Download logs during Dell cluster configuration process
Another reason why the Dell cluster configuration experience saves time is due to the extensive validations. These detailed checks ensure that all environmental requirements are met, and they prevent misconfiguration of critical settings that can lead to costly deployment failures, security breaches, and service disruption. Microsoft’s cluster configuration experience in the Azure portal includes many prerequisite and health checks including the following:
- Create Azure resources like cluster resource and create secrets.
- Check external connectivity requirements.
- Check external active directory preparation.
- Check SBE health requirements.
- Check hardware requirements.
- Check network requirements.
- Check Log Collection and Remote Support requirements.
- Check Operating System requirements.
- Check Moc Stack requirements.
- Check Arc Integration requirements.
- Check cluster witness requirements.
The Dell experience includes these Microsoft checks as part of the cloud cluster deployment validator and adds more:
- Compatibility checks – Ensure all selected nodes have the same hardware (hardware symmetry) and that the hardware appears in the support matrix.
- Network validator – Check connectivity between the Cloud Platform Manager VM and the hosts. Check connectivity between all nodes' networks and confirm that all network configurations are supported. Validate that all Network ATC settings values are valid and that the IP addressing information provided meets requirements.
- Windows Admin Center validator – Validate the user supplied Windows Admin Center FQDN and account are correct by establishing a remote connection.
- Cluster domain name validator – Check to make sure there are no hostname conflicts in AD and no hostname conflicts in the resource group in Azure.
- Azure resources validator – Check that the Key Vault has been created successfully. Also, check the permissions of the Service Principal on the subscriptions and resource group.
If any of these validations fail, you are presented with an explanation. The guided deployment wizard also identifies which step must be revisited to resolve the error. In the following screen shot, the certificates required for secure LDAP connectivity between the Cloud Platform Manager VM and Active Directory appear to be invalid. The UI indicates that you need to navigate to the Global Settings step to take further action. After the certificate issue is addressed, you would re-run the validation. Without this type of validation in place, your deployment would fail. It could be potentially time consuming to troubleshoot the issue during the deployment phase, which could contribute to project delays.
Figure 8 Validation failure example
Additional updates in the latest release
Enhancing the deployment and cluster configuration experience to support Microsoft LTP is just one major update in the latest APEX Cloud Platform for Azure release. The following is a list of additional features:
- Include support for Day 1 deployment of Azure Stack HCI version 23H2 with the 2405 baseline. Applying the latest baseline release ensures you are running in a fully secured and optimized Known Good State at initial deployment.
- Add support for environments with proxy servers for Day 1 deployment.
- Assign custom VLANs and IP addresses to the network adapter ports used for Storage Spaces Direct traffic instead of using the default subnet (10.71.x.x/24). This will help avoid IP conflicts on customer networks.
- Add support for the NVIDIA L40S GPUs for emerging AI use cases.
- Add support for NVIDIA ConnectX-6 10GBase-T NIC transceivers for greater flexibility in network connectivity.
- Update APEX Cloud Platform for Azure with PowerFlex Dell Validated Design to reflect support for PowerFlex software v4.6 and other scalability and networking advancements.
And coming later in August, you can include the new Azure Stack HCI OEM license as part of the purchase of the APEX Cloud Platform for Azure. This license grants you access to the latest version of Azure Stack HCI and Azure Kubernetes Service, along with unlimited containers and Windows Server 2022 Datacenter Edition VMs – all in a single perpetual license.
Conclusion
Indeed, setting up IT infrastructure can be a daunting task, and executives’ impatience is understandable. They are eager to see return on their technology investment as quickly as possible. Dell makes the necessary evils of infrastructure deployment and configuration, well, less evil. The APEX Cloud Platform Foundation Software includes the automation, validation, flexibility, and monitoring needed to ensure a reliable and repeatable Microsoft Azure Stack HCI cluster deployment every time.
Please be sure to check out the additional resources below to learn more about the other APEX Cloud Platform for Azure features that increase operational efficiency and drive innovation like:
- Full stack lifecycle management with continuously validated state assurance
- Event Monitoring for Dell APEX Cloud Platform for Microsoft Azure – Azure Monitor Insights for Azure Stack HCI single and multi-cluster workbooks created by Dell to visualize real-time hardware and software events.
- Rapid Azure Stack HCI cluster expansion.
- Add/replace/remove disks.
- Fully automated node repair/replace.
- Enhanced support through integration with Windows Admin Center – Log and telemetry uploads, call home, automated Support case creation, and remote access by Dell ProSupport technicians.
Resources
- Dell APEX Cloud Platform for Microsoft Azure Release Notes, Support Matrix, and other manuals and KBs
- Azure Stack HCI OEM license overview
- Info Hub white papers, blogs, and videos
- Design Guide – Consuming PowerFlex Block Storage from Dell APEX Cloud Platform for Microsoft Azure
- Implementation Guide – Consuming PowerFlex Block Storage from Dell APEX Cloud Platform for Microsoft Azure
- APEX Cloud Platform for Azure interactive demo
- YouTube playlist with educational and demo videos
- YouTube playlist for March 2024 release
- APEX Cloud Platform for Azure main product page
As always, please reach out to your Dell Technologies account team if you would like to have more in-depth discussions about the Dell APEX Cloud Platforms family. If you don’t currently have a Dell Technologies contact, we’re here to help on our corporate website.
Author: Michael Lamia, Engineering Technologist at Dell Technologies