» Global Velocity Pattern Matching
A patented and patent-pending approach that offers incredible accuracy.
The demand to more intelligently classify and manage data on the wire is at a critical point in the Internet’s evolution. Bandwidth is growing rapidly and threats are becoming more sophisticated. New Internet‐based applications such as Voice over Internet Protocol (VoIP), video on demand (VoD), instant messaging (IM) and peer‐to‐peer (P2P) applications are launched every day. This forces all network operators, from enterprises to government to managed service providers, to continue to strive for deeper awareness of the traffic on their network and for better control. Traditional technical approaches aren’t scaling fast enough to meet these escalating needs.
Global Velocity has developed an advanced approach to identifying, classifying and controlling wire traffic at extremely high speeds through the use of reprogrammable hardware. Field Programmable Gate Array (FPGA) technology and advanced algorithms, including the patented use of Bloom filters and other advanced techniques in hardware, provide the foundation for the Global Velocity Pattern Recognition Engine (PRE). Global Velocity’s PRE performs Deep Packet Inspection (DPI) performs policy-based control of all OSI Layer 4 through 7 traffic at true wire speed.
The Global Velocity PRE implementation in reprogrammable hardware provides the ability to scan for tens of thousands of items in a flexible data set, such as patterns or signatures, with an extremely low (and tunable) false positive rate and zero false negatives.
- » Reprogrammable Hardware
- » Accurate Pattern Matching
- » Deep Packet Inspection
- » TCP Flow Control
- » MicroTCA™
» White Papers
» Join the Revolution
"“When we spoke with the folks at Global Velocity - the company's corporate headquarters is in Clayton, Mo., with an office in Palo Alto, Calif. - we were impressed with what we termed “DLP on steroids”. This is one hot box … We liked this product and the company for their creative approach to a real problem: deep analysis at very high speeds and reduction of the number of boxes needed to put on our perimeters.”"
Read The Article