Moving to these new markets will break them. Moreover, consumer and SOHO users typically do not have an IT department to do initial set-up, configuration, and reconfiguration, let alone handle remediation. What is needed is an automated technology base that CSPs can use to handle these functions effectively and efficiently.

The only effective way to provide
crucial automation

That brings us to distributed dynamic orchestration, which is the only effective way to provide this automation. The easiest way to think about orchestration is as a third layer. The first layer is the data or operational layer. This is where things get done. The second layer is the management layer, where the work of managing the operation of the underlying subsystems goes on. The third—or orchestration—layer is where the other two layers are configured and observed to make sure that they are working well together. In this observation, orchestration watches for failures and acts (remediates) to fix them. These failures can be the result of naturally occurring problems or from cyberattacks.

The orchestration layer must be dynamic, because the underlying data and management layers are constantly changing (as a result of software updates, new products and technologies, and more). In addition, the nature of cyberattacks is also constantly changing. Therefore, orchestration must be able to respond effectively to things that have not, and cannot, be fully anticipated or scripted in advance.

Finally, orchestration has to be distributed because of the volume of behavioral data. Without the behavioral data, it is very hard to find the attacks that have penetrated the outer defenses. With today’s attack volume, even a small percentage that gets through the outer defenses represents a very large number. So, it is critical to identify them very quickly (in some cases, in less than 19 minutes). Trying to capture, hold, and analyze all the behavioral data in consumer and SOHO markets in central site systems is impossible. It takes too long to get the data there. And once there, it takes too long to find “the needle in the haystack.” 

The only feasible way is to geographically distribute both the data and the behavioral analysis. This way, local systems only have to deal with relatively small amounts of local data. And, when necessary, the local systems can work cooperatively. Distributing also has very positive impacts on privacy, among other issues.

With automated distributed dynamic orchestration, plus the existing access to portals and distributed resources, CSPs would be able to quickly capture a new high-profit revenue stream. Pricing would most likely be tiered, starting with low-end consumer and moving up into SOHO. This would help with overall CSP profitability, which is especially important in the face of regulatory and antitrust pressure on existing product pricing. Maybe even more importantly, such an offering would position CSPs to show that they are meeting their responsibility to society. The social responsibility aspect would have a positive impact on employee morale. Thus, deploying such an automated dynamic orchestration service would deliver CSP benefits with investors, customers, regulators, and employees.