In a previous article, we covered the importance of cyber resilient risk management, and disaster recovery as the costs of data breaches and the number of cyberattacks grow.
As promised in this second part, we look at a staged cyber resilience lifecycle approach and its different components.
When setting up a cybersecurity strategy with the necessary applications, backup, and recovery possibilities and defenses to protect your business from attacks and data breaches, you no doubt have already gone through planning and protection stages.
Yet, cyber resilience is broader, and as mentioned in our previous blog, it starts from the prioritization of what’s crucial for the business (data, systems, processes,…) and includes testing, continuous updating/learning, and post-incident recovery.
So, it also requires a more holistic and strategic approach that involves various stakeholders, thus also creating internal awareness since you follow a business- and risk-driven approach. That’s one reason why such a cyber resilience lifecycle framework comes in handy.
Secondly, for most organizations, it’s impossible to do everything alone and automating several tasks that are necessary to strengthen robustness and guarantee business continuity is inevitable. And last but not least, a comprehensive framework enables you to spot internal weaknesses and overcome existing limitations in your current approach.
A cyber resilience lifecycle framework and the place of DRaaS
In an interview on ransomware and Disaster Recovery as a Service (DRaas), IRIS Professional Solutions COO Bart Donné gives a few examples such as a lack of fixes, a single-line defense approach, and issues with user access control. These and other challenges can be identified and tackled using the model we describe below.
IBM (IRIS Professional Solutions is a Platinum IBM Business Partner) has developed a cyber resilience lifecycle framework that is based on the cybersecurity framework of NIST. It consists of five phases. Below is a description of each one with the various outcomes per stage, which include avoiding the weaknesses Bart mentioned and overcoming other hurdles.
The fact that most organizations can’t do everything themselves and this is such an all-encompassing area with many elements to take into account and potentially severe consequences if not done correctly is precisely why many organizations go for a DRaaS approach. For smaller businesses, it offers the benefits of an end-to-end approach at affordable costs, and larger organizations can quickly scale it as the number of servers, applications, and databases they have grows.
Moreover, DRaaS, by default, includes the capabilities you need in the context of cyber resilience as it differs from traditional on-site disaster recovery, enabling rapid post-incident recovery from within a separate environment. An overview of the five phases of IBM’s cyber resilience lifecycle to make it more tangible.
Stage 1 : Identify – a matter of understanding and priorities
In the identification stage, we primarily identify the critical assets, looking at what we have and how we need to protect it.
The ‘identify’ phase is built around the premise that you will be compromised, which is what cyber resilience is about to begin with. In other words: we identify from the perspective of how you will recover from a breach.
This phase includes the automated classification of your systems and data, with prioritization of what data needs to be restored first. We also define where what data is backed up, which depends on the Recovery Time Objective (RTO) for specific types of data classes. Further, the recovery process plan gets defined (for instance, how often is a test needed), and we want to understand the ‘normal’ data profile, or how data is ‘normally’ used. This is essential since it enables you to know when a ransomware attack, which typically is a gradual process, is being prepared as we can see suspicious behavior and usage patterns. Once the ‘identify’ phase is finalized and documented, the resulting plans should be tested to make sure that data can be covered and nothing has been left out.
Stage 2 : Protect – beyond the backup
The protect phase is about detecting vulnerabilities before they are exploited and having the protection in place in your infrastructure with the use of air-gapped data protection and immutable storage technologies.
‘Protect’ isn’t just about a backup whereby you protect your data by copying it somewhere else, encrypting it, and providing an air-gap copy. It’s also about the protection of the overall environment. This includes safeguards such as identity management and role-based access controls, setting up a regular maintenance cycle (including, for instance, patching, as mentioned today often a weakness), and defining proper information protection processes.
A plan regarding awareness and training and decisions on the leverage of protective technology also are typical outcomes in this phase, which essentially concerns developing and implementing the right safeguards to ensure delivery of critical services.
Stage 3 : Detect – monitoring to identify cybersecurity events
In the detect phase, we aim to discover cybersecurity events in a timely way. This is done by using automated testing and verification of backed-up data with advances analytics, enabling to stop attacks and minimize downtime rapidly.
Detecting attacks and breaches is an ongoing task and certainly not an easy one. Monitoring data performance anomalies and setting up continuous detection processes is possible on the primary infrastructure, the backup infrastructure, and in a logical way, with different solutions that IBM and IRIS Professional Solutions offer, with the right mix depending on your scenario and needs. They help identify day to day activity in the primary and copy data areas. In the detect stage, we monitor the performance of revenue-producing data and the copying of revenue-producing data. We also understand the changes in data reduction results.
Stage 4 : Respond – plan, communicate, isolate, shutdown and prepare recovery
Effectively responding to cybersecurity incidents, in particular cyber outbreaks, is our primary goal in the response phase.
The development and implementation of the right activities and actions to take when a cybersecurity incident is detected include multiple aspects such as response planning, communications, and mitigation of the event to minimize disruption. In case a breach effectively occurs, one of the main actions to take is disconnecting systems from networks to avoid that breaches impact more data systems and isolate and shutdown what requires doing so.
This also means that you need to have an air gap copy available or can disconnect your backup servers from your live network and start the recovery of your core business applications in another, ideally new, environment.
Stage 5 : Recover – optimizing for the most efficient recovery time objective when cyberattacks occur
And that brings us to the final phase. The ‘recover’ phase concerns the development and implementation of appropriate activities to maintain plans for resilience and to restore any capabilities or services that were impaired due to a cybersecurity incident, per the definition of NIST.
In practice, it means that we create and test procedural plans to recovery for when a cyberattack or breach occurs, and data on primary revenue-producing systems is compromised. In other words: recovering access to the critical data and applications the business needs. The recovery phase focuses on recovery planning but also on improvements and, again, on communications, essential in times of adverse cyber events.
The previous phases in our cyber resilience lifecycle now all matter. While we identified the critical data to recover first, now we want to do so in practice, keeping in mind the most efficient RTO, the recovery time objective.
Cyber resilience solutions for each scenario and phase
As you can see, cyber resilience can be quite a daunting challenge indeed, with several considerations, plans, and actions to take.
Fortunately, a DRaaS solution helps you automate many functions and performs several of the mentioned tasks, also offering post-incident recovery thanks to the replication of physical or virtual servers to allow switchover.
However, as said, no organization is the same. That’s why, with IBM, we offer a full portfolio of solutions that are typically important across different of the mentioned phases, enable an orchestrated approach, allow for the use of various (multi-cloud and on-premises environments) and support different storage means. Moreover, most solutions enable additional functionalities. Describing them all in a single blog would take us too far, so get in touch to talk about the cyber resilience solutions that fit your organization and plans best.
You can also read more about two of the solutions, IBM Spectrum Virtualize and IBM Spectrum Control, in our article on total storage. IBM Spectrum Virtualize is one of the key components of the overall offering and enables to virtualize the storage layer and give it additional capabilities such as compression, encryption, and performance monitoring, which can be passed on to other solutions.
Get in touch to learn about all the solutions and capabilities for your needs.