System and method for predicting and handling short-term overflow

A system for early predicting of impending data overflow situations in data networks, comprising:

• a) one or more sensors being networked computers that do not provide services, for collecting monitored data regarding network traffic volume from content providers to an Internet Service Provider (ISP) entering or exiting the ISP via peering autonomous systems connected to the ISP via physical links;
• b) one or more processors executing one or more deep learning models, being adapted to:
• c) identify over said data network being handovers alternative peering links, which are not inherent Private Network Interconnects (PNI) between said content providers and said ISP;
• d) determine static or dynamic threshold for dataflow anomaly associated with overflow;
• e) for each alternative handover, predict impending overflow situation by applying ML algorithms to the collected data that corresponds to said each alternative handover;
• and
• f) provide in advance, an alert to each ISP regarding a situation of the data overflow and the alternative channels that should be operated in said situation.

A method for early predicting of impending data overflow situations in data networks, comprising:

• a) collecting monitored data regarding network traffic volume from content providers to an Internet Service Provider (ISP) entering or exiting the ISP via peering autonomous systems connected to the ISP via physical links;
• b) identifying alternative peering links being handovers, which are not inherent private network interconnects (PNIs) between said content providers and said ISP;
• c) determining static or dynamic threshold for dataflow anomaly associated with overflow;
• d) for each alternative handover, predicting impending overflow situation by applying ML algorithms to the collected data that corresponds to said each alternative handover;
• e) providing in advance, an alert to each ISP regarding a situation of the Data overflow and the alternative channels that should be operated in said situation;
• and
• f) allowing said ISP to select alternative handovers through which excess data will be routed during predicted overflow periods.

A system for early predicting of impending data overflow situations in data networks, comprising:

• a) one or more sensors being networked computers that do not provide services, for collecting monitored data regarding network traffic volume from content providers to an Internet Service Provider (ISP) entering or exiting the ISP via peering autonomous systems connected to the ISP via physical links;
• b) one or more processors executing one or more deep learning models, being adapted to:
• c) identify over said data network being handovers alternative peering links, which are not inherent Private Network Interconnects (PNI) between said content providers and said ISP;
• d) determine static or dynamic threshold for dataflow anomaly associated with overflow;
• e) for each alternative handover, predict impending overflow situation by applying ML algorithms to the collected data that corresponds to said each alternative handover;
• f) providing in advance, an alert to each ISP regarding a situation of the Data overflow and the alternative channels that should be operated in said situation;
• and
• g) allow said ISP to select alternative handovers through which excess data will be routed during predicted overflow periods.