To be effective, a multicloud security strategy needs to provide end-to-end protection that gives agencies visibility across hosts and services.
As federal agencies continue to move applications and workloads to multicloud domains, security will have to keep up with the demands of these sprawling environments, especially with the increase in cyberattacks that are more sophisticated, targeted and widespread.
Multicloud environments add more complexity to federal networks that now extend beyond their boundaries to accommodate the needs of a mobile and remote workforce, as well as connect to a host of mobile devices and sensors. As data moves through more access points, the attack surface that hackers can exploit expands.
To be effective, a multicloud security strategy needs to be holistic, providing end-to-end protection that gives agencies visibility across hosts and services.
The benefits of a multicloud strategy are undeniable. Multiple deployment modes in any combination—including public clouds, private clouds and colocation facilities—allow agencies to put the right data and workload at the right place and at the right time.
By using different public cloud providers such as Amazon Web Services, Google Cloud Platform or Microsoft Azure, agencies can match the workloads with the right cloud platform based on their unique requirements for performance, data location, scalability and compliance. Multicloud offers customizable flexibility, allowing an agency to choose the best of each cloud type for their needs at a given time. However, those benefits come with increased complexity, including the need to secure and monitor disparate systems running workloads and hosting data.
True Multicloud? Not Yet …
Agencies are using all types of cloud environments—infrastructure-as-a-service, software-as-a-service, platform-as-a-service—and there is a need for management capabilities across all those ecosystems.
From a marketing perspective this is viewed by some as the use of multicloud domains. But are they truly operating as a multicloud environment, or as separate independent cloud environments without common features and services? Are they taking a holistic approach to security across all environments, particularly with respect to monitoring and availability?
The reality is most agencies do not have true multicloud environments, at least not from an enterprise perspective.
There may be lines of businesses using different cloud environments—whether on-prem or public cloud—in some form or fashion for specific use cases. But they are limited to those use cases that deliver some sort of service or application to specific users.
From an enterprise perspective, an agency IT department might want every component or unit within the agency to use one cloud provider. Therefore, they build a unified environment around that cloud platform to attract all business units. It might work to a certain extent. Typically, agencies have an on-premises cloud ecosystem built around VMware and a public cloud provider, such as AWS or Microsoft. They are able to build security controls and policies using a combination of cloud native services, custom tooling and monitoring capabilities.
However, that could change.
… But It Is Coming
Inevitably agencies will introduce additional public cloud providers into the mix and new complexities and challenges will arise to consistently secure, monitor and enforce policies across multiple providers.
The Central Intelligence Agency last year awarded its long-awaited multibillion-dollar commercial cloud enterprise (C2E) contract to five major cloud titans, including Amazon, Microsoft, Google, Oracle and IBM. The deal will deliver cloud computing capabilities such as data storage, computing, data analytics and other services such as machine learning to the CIA and the other 16 intelligence agencies for the next 15 years.
The Defense Department recently canceled its Joint Enterprise Defense Infrastructure contract with Microsoft for an enterprise cloud platform designed to bring decisive data to the battlefield. Instead, the Pentagon is looking to award the new Joint Warfighter Cloud Capability contract to AWS and Microsoft and is researching the capabilities of other U.S. cloud providers, as well.
The use of multiple different types of cloud platforms by agencies in some sort of capacity is real. The need to manage security homogeneously across those environments is also real.
How then should agencies secure existing multiple cloud environments and prepare to secure true multicloud environments?
Agency managers must ensure that the basic security blocking, tackling and hygiene information security professionals have been championing for years are implemented across the primary cloud ecosystem they have deployed today as well as proactively examining ways to implement them in a multicloud architecture. They should also:
- Conduct a risk assessment across their agency’s cloud footprint to strategically discover all assets and identify risky configurations.
- Assess their resources against compliance standards (i.e., FISMA, FedRAMP, HIPAA, NIST, etc.).
- Implement appropriate technology and services across all environments to continuously correlate siloed data sources, monitor privileged users, uncover critical hosts; and discover cloud resource changes.
The major cloud service providers have built in robust security controls and tools. Still, it is the responsibility of agencies to implement the requisite protocols and solutions to secure data especially when it sits in an on-premises environment and travels to and from the cloud. Additionally, an agency might end up managing a complex set of policies and controls in multiple environments making it harder to implement and enforce policies uniformly.
To that end, a multicloud security strategy must include authentication policies that ensure users access only the cloud-based resources they need for their jobs. The mobile devices the workforce rely on to connect to cloud services must also be protected. Multicloud security should be flexible enough to meet each agency’s unique needs. Visibility into the entire infrastructure is a key requirement for meeting all the security challenges.
Vimesh Patel serves as chief technology advisor to public sector at World Wide Technology.
Maafu Shabaz serves as principal consultant for strategy and risk at World Wide Technology.