Enterprises are faced with an incredibly large network of data that needs to be protected. NVIDIA Morpheus enables digital fingerprinting by monitoring all users, services, accounts, and machines across the enterprise data center to detect suspicious interactions. Combined with NVIDIA GPU and DPU accelerators and NVIDIA DOCA™ telemetry in NVIDIA certified servers, this brings a new level of security to
data centers.
Traditional methods for detecting leaked sensitive data rely on static, rule-based models that are limited by the quality of the training data. NVIDIA Morpheus, on the other hand, examines raw packet information as it is generated for potential data leakage. The DOCA telemetry agent of the
NVIDIA BlueField-2 DPU forwards the raw packets directly to Morpheus. A natural language processing (NLP) model determines if sensitive information - such as passwords and private keys - is leaked in the packet. Packets are immediately flagged and a recommended action is passed back to
DOCA. These real-time alerts are communicated to the operator so that countermeasures can be taken immediately after the data loss.
Fraud detection typically requires a huge amount of labeled data, as well as experts to process and label that data. NVIDIA Morpheus eliminates the need for a security analyst to provide these insights. It also allows for the detection of potential threats that humans might not detect. By leveraging graphical neural networks (GNNs), Morpheus enables detection that was previously impossible without large amounts of labeled data - now you can achieve highly accurate results with a fraction of the labeled data previously required... This allows organizations to increase fraud detection accuracy, potentially saving hundreds of millions.
Phishing emails remain one of the top three vectors for ransomware and malware attacks. This has been exacerbated by the trend towards remote working and teaching, as more attack surface is presented. NVIDIA's Morpheus cybersecurity AI framework uses an NLP model that analyzes emails to identify phishing attempts. Morpheus can analyze either the entire raw text of an email, just the URLs in the email, or both, with custom sequence classification built into cyber log accelerators - one of the building blocks of Morpheus.
Ransomware is among the most costly types of cybersecurity breaches. Incidents have increased globally as threat actors become more sophisticated in their attacks, and attacks could become more frequent if they are financially rewarding. NVIDIA Morpheus provides the optimal AI and machine learning-based cybersecurity pipeline for modern enterprise data centers, combined with NVIDIA BlueField as a high-speed, high-fidelity source and NVIDIA AppShield to identify and stop ransomware at the point of entry.
