Automation: Key to keeping server governance and compliance in lockstep

[ComputeExperts]

There are two significant benefits to using policy automation. First, you know that the configuration is the same for all servers or locations where the policy is enforced. The second benefit is scale. HPE for GreenLake for Compute Ops Management allows you to express a policy, or desired state, that is enforced on all your servers. Read on to learn more.

Managing servers in a datacenter can be challenging. If you scatter those servers worldwide, you significantly increase complexity. Even if you have an experienced, effective team managing those servers, differences or errors in settings can be introduced as they are individually controlled.

One of the best ways to address the variations that stem from server management sprawl is to manage servers through policy automation. This means, rather than an individual performing configuration tasks on a per-server basis, a policy can be configured that represents exactly what the IT manager wants to be configured. At this point, the software ensures that the configuration policy is correctly carried out across all applicable servers. Setting a policy reduces the need for user interaction – and leaves no place for security gaps or errors to be injected.

Replicable, scalable, and secure

There are two significant benefits to using policy automation. First, you know that the configuration is the same for all servers or locations where the policy is enforced. The second benefit is scale. If you are manually configuring servers, you need to add more people and hours to your timeline as you add more servers. With policy automation, the same policy can be implemented through an automated task, and can be applied to virtually unlimited numbers of servers.

HPE for GreenLake for Compute Ops Management allows customers to express a policy, or desired state that is enforced on servers throughout an environment. You can define a policy, and then use that policy to configure a single or multiple groups of servers, ensuring that the same configuration is maintained across all, or a select number, of the servers.

There are many policy types supported by Compute Ops Management today that will continue to grow as the portfolio evolves. For example, you can determine a policy around firmware baselines. That defines what collection of firmware components should a server, or collection of servers, be running. All you need to do to establish this new policy is to configure a firmware baseline setting and attach that setting to your target server group.

Figure 1. View of selection menu for setting firmware baseline for a group of serversManual updates and compliance guidelines

What happens if someone manually updates the firmware on a server? Compute Ops Management can detect this case and report what firmware is out of compliance with the stated policy.

Figure 2. A compliance view of a server showing what is currently running on the server, and what is expected to be running on the server

Another example of automation and compliance might be around a workload profile. Specific workloads often benefit from having several BIOS settings with unique configurations. For example, you may want the BIOS tuned differently for running a virtualized workload versus one that requires low latency. A policy can be created to "tune" the BIOS settings to be most beneficial for your workload.

When you attach the policy to a group, or multiple groups, you know that those servers will be configured in accordance with the policy. The examples above demonstrate active configuration policies, but there are other types as well, such as dealing with a hardware failure.

Dealing with hardware failure

When the hardware fails, you want and need to be notified immediately. Compute Ops Management will send you or your team an email when a system experiences a hardware failure.

Once policies are defined and applied to a server, you can also greatly benefit from knowing when a server does not conform to your stated policy. This is especially important as organizations get larger and grow the number of servers required to support their operations; policy-based automation is a terrific way to scale compute resources – without the need to add additional staff. It also allows for automation of policy enforcement, reporting when a server is out of compliance, and verifying what policy is applied to which servers in your environment.

HPE GreenLake for Compute Ops Management allows all these facets of policy-based management to ease the manual operational burden on administrators. It also helps ensure compliance across your entire fleet of compute devices – including HPE ProLiant Gen10, Gen10 Plus and Gen11 servers as well as new HPE Alletra 4000 storage systems.

See for yourself how easy it is to set up a firmware baseline policy. 

About our author

Blaine Southam is currently the Chief Technologist for HPE Compute Cloud Services. In this role he is for the architecture for HPE GreenLake for Compute Ops Management and associated products. He is also part of the cross-BU technologist team driving HPE to deliver HPE products as a service and is a member of the Edge to Cloud leadership team, as well. Blaine has delivered numerous management solutions, as well as cloud offerings for HP/HPE, for more than 20 years.

Compute Experts
Hewlett Packard Enterprise

twitter.com/hpe_compute
linkedin.com/showcase/hpe-servers-and-systems/
hpe.com/servers