How Resilient are FedRAMP Clouds Anyway?

By Jodi Kohut

For the uninitiated, FedRAMP is the Federal Risk Authorizationand Management Program, a government-wide program that provides a standardized approach to security assessment, authorization, and continuous monitoring for cloud products and services. Implemented to support the Administration’s “Cloud First” policy, some have pointed to FedRAMP as a great model for commercial industry’s adoption of cloud as well. But when it comes to disaster recovery in the cloud, is that necessarily the case?

One of the questions I’ve been asked from the beginning of the Federal Cloud First initiative, is, “If my data is in   The answer is not as clear-cut as the question.  In theory, most cloud services offer extremely resilient platforms and a modicum of disaster recovery is built in. In fact, those cloud service provider (CSP) systems that have received an ATO through the FedRAMP program do have fairly sophisticated contingency plans in place, with Recovery Time Objectives (RTO) and Recovery Point Objectives (RPO) clearly articulated- and plenty of alternate processing sites, policies, and procedures in place in the event of a contingency.  So, it’s in there right?

Not so fast- it depends on what services you are acquiring and how you are deploying and managing them.  The baseline of this discussion is however rooted in availability and uptime. 

A CSP may be able to provide a more resilient infrastructure than an Agency can build internally.  For example, recent research from the International Working Group on Cloud Computing Resiliency (IWGCR) reported 2013 total downtime hours from major providers as follows:

The availability percentages of these providers range from 98.44-99.68%.  Even though the IWGCR believes this data may under report outages, the data may also overstate service downtime.  Let me explain.

The cloud providers mentioned here provide SLA’s for individual services.  Often these are subject to separate SLA’s rather than aggregated ones.   In practice, CSPs orchestrate these services in such a way that a customer can expect 100% availability at a fraction of a cost of building the same solution internally. Considering that only 8% of federal government agencies report confidence in being able to recover 100% of the data required by their governing SLA’s, FedRAMP authorized clouds seem to be perfect for addressing disaster recovery. These same agencies also report an inability to test their disaster recovery plans as often or as thoroughly as they would like. In addition, from an alternative processing site standpoint, Cloud Service Providers offer more, geographically distributed sites for a fraction of the cost of building equivalent solutions internally.  And contrary to the emotions of some, moving disaster recovery to the cloud does not mean relinquishing control of the process or data.  FedRAMP mandatory contractual clauses give the government absolute control of all of its data, all of the time.

So with this in mind, “Is FedRAMP a good model?” Compared to the current state of government IT affairs, the answer is an unequivocal YES! Budget cuts, rapidly increasing IT requirements and the rising threat of cyber-attack are also great arguments for rapid adoption of commercially available, FedRAMP authorized cloud baseddisaster response services. Commercial companies operating in government-regulated industries should leverage this process as well by making FedRAMP provisional approval a minimum requirement for their own cloud service providers.   The list of companies currently in process to receive provisional authorization status for FedRAMP shows industry commitment to security of systems “In the cloud”.   

Follow me at https://Twitter.com/Kevin_Jackson