Chapter 2 · Reading 1 of 9
The Crown Jewels: Data
Security+ SY0-801 · Objectives 3.3
A financial aid application starts with a student typing into a form. By the time a decision reaches that student, several people and systems may have handled the same information. A counselor reviews it, a payment service receives a reference, and a reporting job counts awards. Which of those people actually needs to see a family’s income? That question gives classification a purpose before anyone draws a network diagram.
Security professionals spend a lot of time protecting sensitive information. We act as caretakers, making sure the data our organizations create and the data our customers trust us with are safe. We focus on keeping this data private, accurate, and accessible (think CIA triad).
Data can be classified in three different states:
- Data at rest: This is data stored on hard drives, tapes, in the cloud, or other storage places. It can be stolen by people inside the organization or by outside attackers who gain access to these storage systems.
- Data in transit: This is data moving over a network. When data travels over an untrusted network, it can be intercepted by anyone who has access to that network.
- Data in use: This is data currently being used by a computer system, including data stored in memory while processing is happening. An attacker who controls the system might be able to read this data and steal sensitive information.
We use different security methods to protect data in each of these states. By doing this, we build strong defenses to safeguard our organization's important information.
An In-Depth Look Into Data
Data exists in three states: at rest, in transit, and in use, each requiring different protection strategies. Data at rest is stored on media such as hard drives and cloud storage, vulnerable to theft and unauthorized access, which can be mitigated through encryption, access controls, and physical security. Data in transit, moving across networks, faces threats like eavesdropping and man-in-the-middle attacks, protected by encryption protocols (e.g., TLS), VPNs, and secure communication methods (e.g., HTTPS). Data in use, actively processed by systems, can be compromised through memory scraping and malware, safeguarded by secure coding practices, memory encryption, and stringent access controls. A comprehensive data protection strategy involves risk assessment, data classification, policy development, employee training, and regular audits to ensure robust defenses against breaches, maintaining data confidentiality, integrity, and availability.
Encryption uses complex mathematical algorithms to protect information from being accessed by unauthorized users, at any state of data. Encrypted data looks like random, unreadable text to anyone who doesn’t have the correct decryption key. This makes it safe to store and send encrypted data even over networks that aren't secure.
Data Loss Prevention
Data Loss Prevention (DLP) systems are critical tools that help organizations enforce information handling policies and prevent data loss and theft. These systems are designed to search for unsecured sensitive information within systems and monitor network traffic to detect and block potential data breaches, alerting administrators to any attempted violations before damage occurs.
DLP systems operate in two primary environments
- Agent-based DLP:
- Software Agents: These are installed directly on individual systems within the organization. The agents continuously search for sensitive information, such as Social Security numbers, credit card numbers, and other personal or confidential data that may be stored in various places, often unexpectedly.
- User and Configuration Monitoring: Agent-based DLP can also oversee system configurations and user activities. For example, it can prevent users from accessing USB-based removable media devices, which could be used to transfer sensitive information out of the organization's secure environment.
- Agentless (network-based) DLP:
- Dedicated Network Devices: These devices are placed on the network to monitor outbound traffic. They watch for any unencrypted sensitive information being transmitted.
- Blocking and Encryption: When these systems detect unencrypted sensitive data, they can block the transmission to prevent data loss. In some cases, they can automatically apply encryption to the content, which is especially useful for email communications, ensuring compliance with the organization's security policies.
DLP systems utilize two primary mechanisms to identify and protect sensitive information:
- Pattern Matching:
- Recognition of Sensitive Formats: DLP systems look for patterns that indicate sensitive information. For instance, they can recognize the format of credit card numbers or Social Security
numbers.
- Sensitive Terms Database: They also contain databases of sensitive terms, such as "Top Secret," "Confidential," or specific project names, and can trigger alerts when these terms appear in outbound transmissions.
- Watermarking: Administrators can apply electronic tags or watermarks to sensitive documents. DLP systems then monitor for the presence of these tags in unencrypted content being transmitted over the network.
- Digital Rights Management (DRM): Watermarking is also a key component of DRM solutions, which enforce copyright and data ownership restrictions, ensuring that sensitive and proprietary information is not shared without authorization.
Segmentation involves dividing a network into smaller, distinct segments or subnetworks. Each segment operates as an independent unit with its own security controls. The main purpose of segmentation is to contain and manage the flow of traffic within and between segments, enhancing security and performance.
Key Aspects of Segmentation
- Network Security: By segmenting a network, you can implement security policies tailored to each segment’s needs. For example, you might have stricter controls for segments containing sensitive information, like financial records or personal data.
- Access Control: Segmentation allows for more granular control over who has access to different parts of the network. Users and devices are only given access to the segments necessary for their roles, reducing the risk of unauthorized access.
- Performance Optimization: By reducing the amount of traffic within each segment, you can improve overall network performance and reduce congestion.
- Containment of Breaches: If a breach occurs, segmentation helps contain the attack to a specific segment, preventing it from spreading throughout the entire network. This containment limits the damage and simplifies incident response and recovery.
Isolation is the practice of separating systems, applications, or data to ensure they operate independently from one another. Isolation is used to prevent unauthorized access and to protect sensitive information by creating secure environments.
Key Aspects of Isolation
- Virtual Machines (VMs): VMs allow for the isolation of different applications and operating systems on a single physical machine. Each VM operates in its own environment, separate from others, providing a secure and controlled space.
- Sandboxing: This technique isolates applications or processes to prevent them from affecting the rest of the system. Sandboxing is commonly used to test untrusted software or to run potentially harmful code in a controlled environment.
- Quarantine Zones: In network security, isolation can involve placing suspicious devices or users into quarantine zones where their activity is closely monitored and restricted, preventing potential threats from spreading.
- Dedicated Environments: Sensitive data or critical applications can be isolated in dedicated environments with strict access controls and enhanced security measures, ensuring that only authorized users can access them.
Implementing a comprehensive DLP strategy involves several key steps:
- Risk Assessment: Regularly evaluating the potential risks and vulnerabilities associated with data in the organization.
- Data Classification: Identifying and categorizing data based on its sensitivity and criticality to the organization.
- Policy Development: Establishing clear policies and procedures for data protection, including guidelines for encryption, access control, and incident response.
- Employee Training: Educating employees on the importance of data protection and the organization's DLP policies, ensuring they understand how to handle sensitive information properly.
- Regular Audits and Updates: Continuously reviewing and updating security measures to adapt to new threats and changes in the organization’s data landscape.
Data Minimization
Data minimization techniques aim to reduce risk by limiting the amount of sensitive information maintained. The most effective way to achieve data minimization is to destroy data when it is no longer needed for its original business purpose. When complete removal is not feasible, data can be transformed to deidentify the sensitive information, making it difficult to link back to an individual and reducing its sensitivity.
Deidentification: This process removes personal identifiers from data, so it cannot be traced back to an individual. It’s a key method for reducing the sensitivity of data while retaining its usefulness for analysis.
Data Obfuscation: When deidentification is not possible, data can be transformed into a format where the original information cannot be retrieved. This can be achieved through several methods:
- Hashing: A hash function transforms data into a hash value. For example, applying a strong hash function to a data element and replacing the original value with the hashed value. While this makes direct retrieval impossible, it is susceptible to rainbow table attacks, where attackers use precomputed hash values to identify the original data.
- Tokenization: Sensitive values are replaced with unique identifiers using a lookup table. For instance, a student ID might be replaced with a randomly generated number. The original ID can only be retrieved using a secure lookup table.
- Masking: This technique involves partially redacting sensitive information by replacing certain parts with blank characters. For example, masking a credit card number by showing only the last four digits and replacing the rest with X's or *'s.
A comprehensive data minimization strategy involves
- Data Destruction: Regularly destroying data that is no longer needed.
- Data Transformation: Using deidentification, obfuscation, hashing, tokenization, and masking to protect data that cannot be destroyed.
- Security Measures: Implementing additional security measures, such as salting for hashed data, to protect against attacks.
- Policy Development: Establishing clear policies for data retention, destruction, and transformation.
- Employee Training: Educating staff on the importance of data minimization and the proper handling of sensitive information.
By effectively minimizing the amount of sensitive data stored and transforming data to protect its confidentiality, organizations can significantly reduce the risk of data breaches and enhance overall data security.
Rainbow Table Attacks
A rainbow table attack is a method used by attackers to crack hashed passwords by using precomputed tables of hash values. Hashing can be vulnerable if an attacker has a list of potential values. For example, if student IDs are hashed in a file listing students who failed courses, an attacker could use a list of all students to compute their hash values and match them to the list. This makes it crucial to use additional measures, such as salting (adding random data to the input of a hash function) to strengthen the security of hashed values.
Following secondary copies
Classification follows the information into spreadsheets, backups and diagnostic logs. A data owner assigns sensitivity and permitted use. A custodian implements access, encryption and retention settings. Trace a sample export through its destination and ask whether the original restrictions still apply to everyone who can open it.
Following a record past the front door
Consider a counselor who downloads a spreadsheet to finish a review. Access controls on the original form still work, yet the copy now has its own readers, storage location and lifetime. Put both copies on your data-flow drawing. Beside each, name who decides its use, who implements that decision and who can verify it. Once you know where a disclosure could matter, you can begin asking which threats deserve the most attention.
Inspect the three records, then use the evidence to answer the question.
1. At rest
A report sits in a storage volume.
Encrypted drive stores the student roster
2. In transit
A connection carries the report.
Encrypted channel transfers a copy
3. In use
An authorized user reads it on screen.
Reporting process holds readable rows in memory
Apply the reading
Trace one grade report from storage to a browser. Where can transport encryption help, and where does another control take over?
Compare your reasoning
Encryption protects the connection. Once displayed, access controls, endpoint safeguards, and appropriate handling still matter.