# US 9,438,489 B2 · Computing a performance characteristic of a network device

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

- Number: US 9,438,489 B2
- Document ID: US9438489B2
- Status: Granted
- Kind: B2
- Application number: 14/003,883
- Filed: 2011-04-19
- Published: 2013-12-26
- Granted: 2016-09-06
- Earliest claimed: 2011-04-19
- Assignees: Hewlett Packard Enterprise Development LP
- Applicants: Yordanos Beyene
- Inventors: Yordanos Beyene
- Web page: https://patentgrep.com/patent/US9438489B2

## CPC

- H04L 41/145: {involving simulating, designing, planning or modelling of a network}
- H04L 41/147: for predicting network behaviour
- H04L 43/04: Processing captured monitoring data, e.g. for logfile generation
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

A traffic profile or description of a mix of different types of traffic is provided to compute, using a model, a performance characteristic of a network device. The model relates a performance characteristic of the network device to parameters associated with different types of traffic.

## Claims

### Claim 1

A method of characterizing a network device, comprising:
receiving, by a system having a processor, traffic information, wherein the traffic information describes a mix of traffic according to different application-level protocols, wherein the traffic information comprises information relating to a plurality of flows, each respective flow of the plurality of flows corresponding to a respective application-level protocol of the different application-level protocols;
converting, by the system, the traffic information to a traffic profile that describes traffic according to a transport protocol;
inputting, by the system, the traffic profile corresponding to the traffic information into a model stored in a non-transitory storage medium, wherein the model relates a performance characteristic of the network device to parameters associated with different types of traffic; and
computing, from the model in response to the input traffic profile, the performance characteristic of the network device.

### Claim 2 (depends on 1)

The method of claim 1, wherein inputting the traffic profile into the model comprises inputting the traffic profile into the model that relates the performance characteristic of the network device to cost parameters associated with traffic according to different transport protocols.

### Claim 3 (depends on 1)

The method of claim 1, wherein receiving the traffic information comprises receiving a trace of actual traffic in a network in which the network device is to be deployed.

### Claim 4 (depends on 1)

The method of claim 1, wherein computing the performance characteristic of the network device comprises computing a performance characteristic relating to an amount of data that can be processed by the network device.

### Claim 5 (depends on 1)

The method of claim 1, wherein computing the performance characteristic of the network device comprises computing performance characteristics relating to amounts of respective different types of data that can be processed by the network device.

### Claim 6

A system comprising:
a non-transitory storage medium to store a model that relates a performance characteristic of a network device to parameters associated with different types of traffic;
at least one processor to:
receive traffic information that describes a mix of traffic according to different application-level protocols, wherein the traffic information comprises information relating to a plurality of flows, each respective flow of the plurality of flows corresponding to a respective application-level protocol of the different application-level protocols;
convert the traffic information to a traffic profile that describes traffic according to a transport protocol;
input the traffic profile into the model; and
compute a performance characteristic of the network device using the model in response to the input traffic profile, wherein the model outputs a value for the performance characteristic in response to parameters associated with the different types of traffic.

### Claim 7 (depends on 6)

The system of claim 6, wherein the traffic profile describes traffic according to different transport protocols, the different transport protocols comprise a Transmission Control Protocol and a User Datagram Protocol, and wherein the parameters comprise a first parameter associated with traffic of the Transmission Control Protocol, and a second parameter associated with traffic of the User Datagram Protocol.

### Claim 8

An article comprising at least one non-transitory machine-readable storage medium storing instructions that upon execution cause a system having a processor to:
receive traffic information that describes a mix of traffic according to different application-level protocols, wherein the traffic information comprises information relating to a plurality of flows, each respective flow of the plurality of flows corresponding to a respective application-level protocol of the different application-level protocols;
convert the traffic information to a traffic profile that describes traffic according to a transport protocol;
input the traffic profile corresponding to the traffic information into a model stored in a non-transitory storage medium, wherein the model relates a performance characteristic of the network device to cost parameters associated with different types of traffic; and
compute, from the model in response to the input traffic profile, the performance characteristic of the network device.

### Claim 9 (depends on 8)

The article of claim 8, wherein the traffic profile describes a mixture of different traffic types according to different transport protocols.

### Claim 10 (depends on 8)

The article of claim 8, wherein the performance characteristic is at least one selected from among a maximum number of packets that can be processed by the network device in a given time interval, and maximum bytes that can be processed by the network device in the given time interval.

### Claim 11 (depends on 2)

The method of claim 2, wherein the cost parameters related by the model to the performance characteristic of the network device comprise a first central processing unit (CPU) utilization for a first transport protocol of the different transport protocols, and a second CPU utilization for a second transport protocol of the different transport protocols.

### Claim 12 (depends on 11)

The method of claim 11, wherein the traffic profile specifies respective amounts of traffic for the different transport protocols, wherein the traffic profile specifies a first amount of traffic for the first transport protocol, and a second amount of traffic for the second transport protocol, and wherein the model outputs the performance characteristic of the network device based on the first and second CPU utilizations and the first and second amounts.

### Claim 13 (depends on 1)

The method of claim 1,
wherein converting the traffic information to the traffic profile that describes traffic according to a transport protocol comprises using first indicators and second indicators associated with the plurality of flows, each first indicator of the first indicators to indicate whether or not a respective flow of the plurality of flows has traffic according to a first transport protocol, and each second indicator of the second indicators to indicate whether or not a respective flow of the plurality of flows has traffic according to a second transport protocol.

### Claim 14 (depends on 6)

The system of claim 6, wherein the parameters related by the model to the performance characteristic of the network device comprise a first central processing unit (CPU) utilization for a first transport protocol, and a second CPU utilization for a second transport protocol.

### Claim 15 (depends on 14)

The system of claim 14, wherein the traffic profile specifies a first amount of traffic for the first transport protocol, and a second amount of traffic for the second transport protocol, and wherein the model outputs the performance characteristic of the network device based on the first and second CPU utilizations and the first and second amounts.
