# US 2025/0168185 A1 · Service plane optimizations with learning-enabled flow identification

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2025/0168185 A1
- Document ID: US20250168185A1
- Status: Published application
- Kind: A1
- Application number: 19/034,804
- Filed: 2025-01-23
- Published: 2025-05-22
- Earliest claimed: 2019-08-07
- Assignees: Cisco Technology, Inc.
- Applicants: Cisco Technology, Inc.
- Inventors: Ramanathan Subramanian; Jeslin Antony Puthenparambil
- Web page: https://patentgrep.com/patent/US20250168185A1

## CPC

- H04L 41/16: using machine learning or artificial intelligence
- H04L 63/1425: {Traffic logging, e.g. anomaly detection}
- H04L 43/062: related to network traffic
- H04L 43/0876: Network utilisation, e.g. volume of load or congestion level
- H04L 63/1408: {by monitoring network traffic (monitoring network traffic per se H04L43/00)}
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

The disclosed technology relates to a process for optimizing data flow within a computer network. The technology utilizes shared memory and machine learning logic to improve the efficiency of how computing resources are used during a transmission of data packets in the computer network. The shared memory is implemented during the transmission of data packets between the data plane and the service plane so that the copying of data packets after the data packets have been received and processed by an application is not necessary. The machine learning logic is implemented during the processing of the data packets in order to adjust a frequency or extent that the data packets (and corresponding source of the data packets) need to be evaluated to ensure that malicious content is not being transmitted across the computer network.

## Claims

### Claim 1

A method for optimizing data flow, the method comprising:
storing, in a shared memory, metadata that identifies a data packet to yield stored metadata, wherein the shared memory is accessible only between a data plane of a first computer network where the data packet originates and a service plane of a second computer network;
transmitting the data packet to an application on the second computer network;
receiving, from the application, a data package comprising a verdict associated with the data packet and the metadata that identifies the data packet;
matching, based on the metadata received from the application, the stored metadata with the data packet to generate a match; and
carrying out, based on the match, the verdict on data packet.

### Claim 2 (depends on 1)

The method of claim 1, wherein data package comprises data plane identifying information that is stored in a ring buffer associated with the shared memory.

### Claim 3 (depends on 1)

The method of claim 1, further comprising:
calculating a confidence level using machine learning logic, wherein the confidence level is calculated based on the data packet; and
assigning a source of the data packet with the confidence level, wherein the confidence level identifies a level of trustworthiness of the source and its corresponding data packets.

### Claim 4 (depends on 3)

The method of claim 3, wherein the confidence level defines a frequency of performing subsequent evaluations on other data packets associated with the source, wherein the frequency identifies a period of time until a next evaluation.

### Claim 5 (depends on 4)

The method of claim 4, wherein the frequency also identifies a number of data packets that are transmitted without evaluations until the next evaluation.

### Claim 6 (depends on 3)

The method of claim 3, wherein the confidence level defines an extent of evaluating other data packets associated with the source, wherein the extent identifies a number of data packets that are evaluated during a next evaluation.

### Claim 7

A non-transitory computer-readable medium comprising instructions that optimize data flow in a computer network, the instructions, when executed by a computing system, cause the computing system to be configured to:
store, in a shared memory, metadata that identifies a data packet to yield stored metadata, wherein the shared memory is accessible only between a data plane of a first computer network where the data packet originates and a service plane of a second computer network;
transmit the data packet to an application on the second computer network;
receive, from the application, a data package comprising a verdict associated with the data packet and the metadata that identifies the data packet;
match, based on the metadata received from the application, the stored metadata with the data packet to generate a match; and
carry out, based on the match, the verdict on data packet.

### Claim 8 (depends on 7)

The non-transitory computer-readable medium of claim 7, wherein the non-transitory computer-readable medium stores additional instructions which, when executed by the computing system, cause the computing system to be configured to:
store data plane identifying information in the data package in a ring buffer associated with the shared memory.

### Claim 9 (depends on 7)

The non-transitory computer-readable medium of claim 7, wherein the non-transitory computer-readable medium stores additional instructions which, when executed by the computing system, cause the computing system to be configured to:
calculate a confidence level using machine learning logic, wherein the confidence level is calculated based on the data packet; and
assign a source of the data packet with the confidence level, wherein the confidence level identifies a level of trustworthiness of the source and its corresponding data packets.

### Claim 10 (depends on 9)

The non-transitory computer-readable medium of claim 9, wherein the confidence level defines a frequency of performing subsequent evaluations on other data packets associated with the source, wherein the frequency identifies a period of time until a next evaluation.

### Claim 11 (depends on 10)

The non-transitory computer-readable medium of claim 10, wherein the frequency also identifies a number of data packets that are transmitted without evaluations until the next evaluation.

### Claim 12 (depends on 9)

The non-transitory computer-readable medium of claim 9, wherein the confidence level defines an extent of evaluating other data packets associated with the source, wherein the extent identifies a number of data packets that are evaluated during a next evaluation.

### Claim 13

A system optimizing data flow in a computer network, the system comprising:
a processor; and
a memory storing instructions that, when executed by the system, cause the system to perform operations comprising:
store, in a shared memory, metadata that identifies a data packet to yield stored metadata, wherein the shared memory is accessible only between a data plane of a first computer network where the data packet originates and a service plane of a second computer network;
transmit the data packet to an application on the second computer network;
receive, from the application, a data package comprising a verdict associated with the data packet and the metadata that identifies the data packet;
match, based on the metadata received from the application, the stored metadata with the data packet to generate a match; and
carry out, based on the match, the verdict on data packet.

### Claim 14 (depends on 13)

The system of claim 13, wherein the memory stores additional instructions that, when executed by the system, cause the system to be configured to:
store, from the data package, data plane identifying information in a ring buffer associated with the shared memory.

### Claim 15 (depends on 13)

The system of claim 13, wherein the memory stores further instructions that, when executed by the system, cause the system to be configured to:
calculate a confidence level using machine learning logic, wherein the confidence level is calculated based on the data packet; and
assign a source of the data packet with the confidence level, wherein the confidence level identifies a level of trustworthiness of the source and its corresponding data packets.

### Claim 16 (depends on 15)

The system of claim 15, wherein the confidence level defines a frequency of performing subsequent evaluations on other data packets associated with the source, wherein the frequency identifies a period of time until a next evaluation.

### Claim 17 (depends on 16)

The system of claim 16, wherein the frequency also identifies a number of data packets that are transmitted without evaluations until the next evaluation.

### Claim 18 (depends on 15)

The system of claim 15, wherein the confidence level defines an extent of evaluating other data packets associated with the source, wherein the extent identifies a number of data packets that are evaluated during a next evaluation.

### Claim 19 (depends on 15)

The system of claim 15, wherein the memory stores additional instructions that, when executed by the system, cause the system to be configured to:
evaluate the data packet at the application by determining whether the data packet includes malicious content.

### Claim 20 (depends on 15)

The system of claim 15, wherein the memory stores further instructions that, when executed by the system, cause the system to be configured to:
generate a verdict based on an evaluation of the data packet, wherein the verdict instructs the computer network whether the data packet is to be forwarded or dropped.
