|By Raja Patel||
|August 28, 2014 01:15 PM EDT||
The cloud plays an integral role in enabling the agility required to take advantage of new business models and to do so in a very convenient and cost-effective way. However, this also means that more personal information and business data will exist in the cloud and be passed back and forth. Maintaining data integrity is paramount.
Today's approach to security in the cloud may not be sufficient; it doesn't focus on putting controls close to data, which is now more fluid, and it doesn't discriminate one set of data from another. All data is not created equal and should not be treated in the same manner; a one-size fits all model doesn't work.
In this always-connected world, protection measures in the cloud need to focus on what really matters - the type of data, how it is used, and where it goes.
In order to adequately protect data in the cloud, organizations need to start considering how to classify data. One approach is to use a three-tier data protection model to cater to data of different sensitivities and relevance across industries. This model would include:
Tier 1, Regulated: Data subject to regulation, or data that carries with it proprietary, ethical, or privacy considerations such as personally identifiable information (PII). Unauthorized disclosure of regulated data may have serious adverse effects on an organization's reputation, resources, services, or individuals and requires the most stringent level of control.
Tier 2, Commercial: Industry-related, ecommerce or transactional and intellectual property data whose unauthorized disclosure may have moderately adverse effects on an organization's reputation, resources, services, or individuals. Commercial data requires a moderate level of security.
Tier 3, Collaborative: Collaborative and DevOps-type data that typically is publicly accessible, requires minimal security controls and poses little or no risk to the consuming organization's reputation, resources, or services.
Using this model, security teams can strategically partner with business users to understand requirements and determine the right approach for their organization. Small to mid-sized organizations, enterprises, and service providers can apply this model to begin classifying their data based on contextual attributes such as how the data will be accessed, stored, and transmitted. Once the data is classified, they can then apply appropriate data protection measures focused on protecting work streams and transactions that continue to evolve to enable business agility. Given that most of today's data breaches are a result of user-access issues, security considerations such as Identity and Access Management, Authorization, and Authentication are critical.
The Data Integrity Challenge
Understanding and classifying data is just a first step, albeit an important one. Organizations also need to determine how to ensure data integrity when the perimeter is amorphous and control of the endpoints and the data is diminished mobility and cloud services.
Business departments are increasingly encouraged to find efficient and innovative ways to generate new business. This requires identifying new applications and ways to support the business anywhere and anytime. Business users often make the decision to use the cloud before involving IT since they can get up and running in a fraction of the time and cost it would take to provision in house.
With this unprecedented change in operations and infrastructure comes an unprecedented need for ensuring data integrity - ultimately working through the life cycle of data that can, at any point, be within the confines of a company, out to a network of partners and suppliers, or floating in a cloud. The challenge in this fractured landscape is that the perimeter is amorphous, but legacy security solutions are not; designed for a time when there was a more well-defined perimeter. The result is that attackers now use various techniques to bypass traditional perimeter-based defenses and compromise data - be it through tampering, stealing, or leaking data. Point-in-time defenses are no longer sufficient.
To effectively protect data wherever it may be, defenses must go beyond simply blocking and detection to including capabilities such as data correlation, continuous data analysis, and retrospective action when data has been found to have been corrupted, tampered with, or exfiltrated.
A New Approach to Applying Controls
In order to protect the classes of data described earlier - regulated, commercial, and collaborative - security teams need a mix of policy, process, and technology controls. These controls should be applied based on user and location context and according to a security model that is open, integrated, continuous, and pervasive:
- Open to provide access to global intelligence and context to detect and remediate breaches and to support new standards for data protection.
- Integrated solutions that enable policy to be automated and minimize manual processes can close gaps in security and support centralized management and control according to data classifications.
- Both point-in-time solutions as well as continuous capabilities are needed to identify new threats to data.
- Pervasive security delivers protection across the full attack continuum - before, during, and after an attack.
Let's take a closer look at the advantages of applying controls to protect data based on this model.
- The opportunity to participate in an open community of users and standards bodies to ensure consistent data classification and standards of policy and process.
- Easy integration with other layers of security defenses to continue to uphold data protection best practices as IT environments and business requirements change.
- The ability to access to global intelligence with the right context to identify new threats and take immediate action.
- Technology controls that map to data tiers and also track data through different usage contexts and locations to support the fundamental first step of data classification.
- Identity and access controls, authorization, and authentication that work in unison to map data protection to data classifications.
- Encryption controls applied based on deemed data sensitivity to further strengthen protection, including strong encryption key standards (minimum AES256) and encryption keys retained by data owners.
- Security solutions and technologies that seamlessly work together to protect data across its entire lifecycle.
- Centralized policy management, monitoring, and distributed policy enforcement to ensure compliance with regulatory and corporate policies.
- Technologies and services to constantly aggregate and correlate data from across the connected environment with historical patterns and global attack intelligence to maintain real-time contextual information, track data movement, and detect data exfiltration.
- The ability to leverage insights into emerging new threats, take action (automatically or manually) to stop these threats, and use that intelligence to protect against future data breaches.
Pervasive translates into:
- Defenses (including technologies and best practices) that address the full attack continuum - before, during, and after an attack. Before an attack, total, actionable visibility is required to see who is accessing what data from where and how, and to correlate that information against emerging threat vectors. During an attack, continuous visibility and control to analyze and take action in real time to protect data is necessary. After an attack, the key is to mitigate the damage, remediate, quickly recover, and prevent similar, future data breaches, data tampering, or data corruption activities.
- The ability to address all attack vectors - including network, endpoints, virtual, the cloud, email and Web - to mitigate risk associated with various communications channels that could be used by an attacker to compromise data.
Today's cloud-driven, always-connected world is enabling organizations to be very agile but it is also putting data integrity at risk. IT teams need to quickly adapt to this new way of doing business despite having less control of the endpoints and the data. Traditional data protection models fail due to their inability to discriminate one set of data from another. By putting in place protection measures based on the type of data, how it is used, and where it goes, and backed by a security model that is open, integrated, continuous, and pervasive, organizations can take advantage of new business opportunities the cloud affords without sacrificing data integrity.
Feb. 7, 2016 06:45 PM EST Reads: 136
Feb. 7, 2016 05:15 PM EST Reads: 312
Feb. 7, 2016 04:45 PM EST Reads: 184
Feb. 7, 2016 04:15 PM EST Reads: 257
Feb. 7, 2016 03:30 PM EST Reads: 370
Feb. 7, 2016 02:45 PM EST Reads: 138
Feb. 7, 2016 02:00 PM EST Reads: 553
Feb. 7, 2016 02:00 PM EST
Feb. 7, 2016 01:45 PM EST Reads: 360
Feb. 7, 2016 01:30 PM EST
Feb. 7, 2016 01:15 PM EST Reads: 181
Feb. 7, 2016 01:00 PM EST Reads: 147
Feb. 7, 2016 01:00 PM EST Reads: 331
Feb. 7, 2016 12:00 PM EST Reads: 341
Feb. 7, 2016 11:45 AM EST Reads: 125