Are you Following Micro-Segmentation Best Practices?

With IT infrastructures increasingly becoming more virtualized and software-defined, micro-segmentation is fast becoming a priority for IT teams to enhance security measures and reduce the attack surface of their data center and cloud environments. With its fine-grained approach to segmentation policy, micro-segmentation enables more granular control of communication flows between critical application components that goes a step further than traditional network segmentation methods in support of moving to a Zero Trust security model.

Finding the Right Segmentation Balance

If not approached in the right way, micro-segmentation can be a complex process to plan, implement, and manage. For example, overzealous organizations may run too fast to implement these fine-grained policies across their environment, leading to over-segmentation, which could have a negative impact on the availability of IT applications and services, increase security complexity and overhead, and actually increase risk. At the same time, businesses need to be aware of the risks of under-segmentation, leaving the attack surface dangerously large in the case of a breach.

With a well-thought-out approach to micro-segmentation, organizations can see fast time to value for high-priority, short term use cases, while also putting in place the right structure for a broader implementation of micro-segmentation across future data center architectures. To achieve your micro-segmentation goals without adding unnecessary complexity, a business should consider these micro-segmentation security best practices.

Start with Granular Visibility Into Your Environment

It’s simple when you think about it – how can you secure what you can’t see? Whether you’re using application segmentation to reduce the risk between individual or groups of applications, or tier segmentation to define the rules for communication within the same application cluster, you need visibility into workloads and flows, at a process level. Process-level visibility allows security administrators to identify servers with similar roles and shared responsibilities so they can be easily grouped for the purpose of establishing security policies.

At first blush, this may seem to be a daunting task and is likely the first impediment to effective micro-segmentation. However, with the aid of graphic visualization tools that enable administrators to automatically discover and accurately map their data center applications and the communication processes between them, the complexity of implementing an effective micro-segmentation strategy can be greatly simplified.

Once administrators have gained this depth of visibility, they can begin to filter and organize applications into groups for the purpose of setting common security policies – for example, all applications related to a particular workflow or business function. The micro-segmentation best practices are to then create policies that can be tested and refined as needed for each defined group.

Micro-Segmentation Best Practices for Choosing the Right Model

There are two basic models for the implementation of micro-segmentation in a data center or cloud environment: network-centric, which typically leverages hypervisor-based virtual firewalls or security groups in cloud environments, and application-centric, which typically are agent-based distributed firewalls. Both have some pros and cons.

In a network-centric model, traffic control is managed by network choke points, third-party controls, or by trying to enforce rules onto each workload’s existing network enforcement.

In contrast, an application-centric model deploys agents onto the workload itself. This has a number of benefits. Visibility is incomparable, available down to Layer 7, and without the constraints or encryption that proprietary applications may enforce. An agent-based solution is also suitable across varied infrastructures, as well as any operational environment. This gives one consistent method across technologies, even when you consider new investments in containers and other microservices-based application development and delivery models.

Additionally, as there are no choke points to consider, the policy is entirely scalable, and can follow the workload even as it moves between environments, from on-premises to public cloud and back. Also, an application-centric approach allows you to define more granular policies, which reduces the attack surface beyond what can be accomplished with a network-centric model. Tools built for a specific environment are simply not good enough for hybrid multi-cloud data center needs, which explains why agent-based solutions have become micro-segmentation security best practices in recent years.

Also with agent-based approaches, one can more easily align with the DevOps models most enterprises use today. Business can leverage automation and autoscaling to streamline provisioning and management of workloads. Micro-segmentation policies are able to be easily and dynamically incorporated. There is no need for manual moves, adds and changes like you would have in the network-centric model.

Define “Early Win” Use Cases

Organizations that are successful with micro-segmentation typically start by focusing on projects that are tangible, fairly easy to complete, and in which the benefits will be readily apparent. These typically include something as basic as environment segmentation, such as separating servers and workloads in development or quality assurance from those in production.

Another common starting point is the isolation of applications for compliance purposes, known to be one of micro-segmentation security best practices. Regulatory regimes such as SWIFT, PCI, or HIPAA typically spell out the types of data and processes that must be protected from everyday network traffic. Micro-segmentation allows for the quick isolation of these applications and data, even if the application workloads are distributed across different environments, such as on-premises data centers and public clouds.

Organizations may also undertake projects to restrict access to data center assets or services from outside users or Internet of Things devices. In health care, hospitals will use micro-segmentation to isolate medical devices from the general network. Businesses might use micro-segmentation as a means of traditional ring-fencing to isolate their most critical applications.

The common thread running through these examples is that they represent business needs and challenges for which micro-segmentation is ideally suited. They are easily defined projects with clear business objectives while at the same providing a proving ground for micro-segmentation.

Think Long Term and Consider the Cloud

Organizations that have successfully implemented micro-segmentation typically take a phased approach, piloting on a few priority projects, getting comfortable with the tools and the process, and gradually expanding. A pragmatic approach to micro-segmentation is to align your requirements with both your current and future-state data center architectures.

A key component of this is consideration of “coverage” in your micro-segmentation tool stack. Look for tools that cover not only a single environment, but provide support for workloads in both your current and future data center architectures. This typically includes workloads running on legacy systems, bare metal servers, virtualized environments, containers and public cloud.

In addition, don’t assume that native security controls offered by IaaS or public cloud services will be adequate enough to fully protect your cloud workloads. Cloud service providers operate on a shared security model, in which the provider takes responsibility for securing the cloud infrastructure while customers are responsible for their own operating systems, applications and data. A cloud provider’s controls are only effective in that provider’s environment. Enterprises would have to manage multiple security platforms and make manual adjustments as applications move among different cloud environments. Furthermore, most native security controls are directed at the port level (Layer 4) and not at the process level (Layer 7) where vulnerable applications reside. That means they will not reduce the attack surface sufficiently to be effective.

Integrate with Complementary Controls

When evaluating solutions, another of micro-segmentation best practices is to look for those where there are value-added and integrated complementary controls. This helps reduce security management complexity, as you can find solutions that give you more than just micro-segmentation out of the box.

Single-platform micro-segmentation solutions might be effective at segmenting your applications and workloads to reduce risk. Micro-segmentation security best practices, however, are to look for a choice that takes you to the next level. Threat detection and response is a perfect example of a valuable complementary control. It allows you to do more than simply protect processes and check compliance off your to-do list. Of course, both breach detection and incident response are must-haves for any complex IT infrastructure.

The difference with an all-in-one solution is the reduction in administrative overhead of attempting to make disparate solutions work in tandem. As micro-segmentation tackles risk reduction in both data centers and clouds – threat detection and incident response can take the obvious next step in quickly detecting and mitigating active breaches, which can help your dramatically reduce dwell time and reduce the cost and impact of a breach.

A Summary of Micro-Segmentation Security Best Practices

From choosing an application-centric model that deploys agents onto the workload itself and comes with valuable complementary controls, to ensuring visibility from the start and looking for the ‘quick wins’ that provide early value, following these micro-segmentation security best practices will give your business the best chance of successful implementation.

For more information on micro-segmentation, visit our Micro-Segmentation Hub

0 comments

Leave a Comment

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *