Method and traffic processing unit for handling traffic in a communication network

Patent metadata

Field	Value
Publication number	US-11290385-B2
Application number	US-201716954138-A
Country	US
Kind code	B2
Filing date	Dec 15, 2017
Priority date	Dec 15, 2017
Publication date	Mar 29, 2022
Grant date	Mar 29, 2022

Abstract

A method and a traffic processing unit (200) for handling traffic in a communication network when the traffic is distributed across a set of traffic processing units. When receiving a packet of a traffic flow distributed to said traffic processing unit, the traffic processing unit (200) assigns a packet class to the received packet, which class can be active or inactive in the traffic processing unit. The traffic processing unit obtains state information of the assigned packet class. If the packet class is detected as active the state information is retrieved from a local storage (200C) in the traffic processing unit, and if the packet class is detected as inactive the state information is fetched from a central storage (204). The traffic processing unit then performs stateful packet processing of the received packet based on the obtained state information.

First claim

The invention claimed is:

1. A method performed by a traffic processing unit (TPU) for handling traffic in a communication network, the method comprising:
for each packet class identifier included in a set of packet class identifiers, the TPU storing an associated storage indicator that indicates whether or not state information for the packet class identified by the packet class identifier is stored in a local storage;
the TPU receiving a packet of a traffic flow;
based on information derived from the received packet, the TPU assigning one of said packet class identifiers to the received packet;
the TPU using the assigned packet class identifier to retrieve the storage indicator associated with the assigned packet class identifier;
the TPU retrieving the state information for the packet class identified by the assigned packet class identifier based on the retrieved storage indicator; and
performing stateful packet processing of the received packet based on the obtained state information, wherein
retrieving the state information for the packet class identified by the assigned packet class identifier based on the retrieved storage indicator comprises:
determining whether or not the retrieved storage indicator indicates that the state information is stored in the local storage; and either
i) retrieving the state information from the local storage after and as a result of determining that the retrieved storage indicator indicates that the state information is stored in the local storage, or
ii) retrieving the state information from a non-local storage after and as a result of determining that the retrieved storage indicator indicates that the state information is not stored in the local storage.

2. The method of claim 1 , wherein retrieving the state information for the packet class identified by the assigned packet class identifier based on the retrieved storage indicator comprises retrieving the state information from the non-local storage after and as a result of determining that the retrieved storage indicator indicates that the state information is not stored in the local storage.

3. The method of claim 2 , further comprising, after retrieving the state information from the non-local storage, storing the retrieved state information in the local storage and changing the storage indicator associated with the assigned packet class identifier to indicate that the state information is stored in the local storage.

4. The method of claim 1 , wherein the storage indicator associated with the assigned packet class identifier is maintained in a packet class record with storage indicators of different packet classes for traffic flows that have been handled by the traffic processing unit.

5. The method of claim 1 , wherein said local storage comprises multiple local memory volumes containing state information of different packet classes.

6. The method of claim 5 , wherein each local memory volume contains state information of one or more packet classes.

7. The method of claim 1 , wherein said information derived from the received packet comprises at least one of a source address or a destination address.

8. The method of claim 1 , further comprising:
after retrieving the state information, determining that the state information has not been used by the traffic processing unit for at least a certain time-out period; and
deleting the state information from the local storage as a result of determining that the state information has not been used by the TPU for at least the certain time-out period.

9. The method of claim 1 , wherein the traffic is received when distributed by a stateless distribution device such as a layer-3 Ethernet switch.

10. The method of claim 1 , wherein the traffic processing unit is implemented as a Virtual Machine.

11. A traffic processing unit (TPU) arranged to handle traffic in a communication network, wherein the TPU comprises:
a receiver for receiving a packet of a traffic flow; and
a processing circuitry configured to:
for each packet class identifier included in a set of packet class identifiers, store an associated storage indicator that indicates whether or not state information for the packet class identified by the packet class identifier is stored in a local storage;
based on information derived from the received packet, assign one of said packet class identifiers to the received packet;
use the assigned packet class identifier to retrieve the storage indicator associated with the assigned packet class identifier;
retrieve the state information for the packet class identified by the assigned packet class identifier based on the retrieved storage indicator; and
perform stateful packet processing of the received packet based on the obtained state information, wherein
the TPU is configured to retrieve the state information for the packet class identified by the assigned packet class identifier based on the retrieved storage indicator by:
determining whether or not the retrieved storage indicator indicates that the state information is stored in the local storage; and either
i) retrieving the state information from the local storage after and as a result of determining that the retrieved storage indicator indicates that the state information is stored in the local storage, or
ii) retrieving the state information from a non-local storage after and as a result of determining that the retrieved storage indicator indicates that the state information is not stored in the local storage.

12. The traffic processing unit of claim 11 , wherein the TPU is configured such that, after retrieving the state information from the non-local storage, the TPU stores the state information in the local storage.

13. The traffic processing unit of claim 12 , wherein the TPU is configured such that, upon storing the state information in the local storage, the TPU changes the storage indicator associated with the assigned packet class identifier to indicate that the state information is stored in the local storage.

14. The traffic processing unit of claim 11 , wherein the traffic processing unit is configured to maintain the storage indicator associated with the assigned packet class identifier in a packet class record with storage indicators of different packet classes for traffic flows that have been handled by the traffic processing unit.

15. The traffic processing unit of claim 11 , wherein said local storage comprises multiple local memory volumes adapted to contain state information of different packet classes.

16. The traffic processing unit of claim 15 , wherein each local memory volume is adapted to contain state information of one or more packet classes.

17. The traffic processing unit of claim 11 , wherein the traffic processing unit is configured to derive from the received packet said information comprising at least one of a source address or a destination address.

18. The traffic processing unit of claim 11 , wherein if the state information for said traffic flow is changed, the local memory volume and the central storage are updated accordingly.

19. The traffic processing unit of claim 11 , wherein the traffic is received when distributed by a stateless distribution device such as a layer-3 Ethernet switch.

20. A computer program product comprising a non-transitory computer readable medium storing a computer program comprising instructions which, when executed on at least one processor, cause the at least one processor to carry out the method of claim 1 .

Assignees

• Ericsson Telefon Ab L M

Inventors

• Hammam Tarik
• Franzen Anders
• Kövér János

Classifications

• H04L41/0897

  by horizontal or vertical scaling of resources, or by migrating entities, e.g. virtual resources or entities · CPC title

• H04L41/0895

  Configuration of virtualised networks or elements, e.g. virtualised network function or OpenFlow elements · CPC title

• H04L47/2441 Primary

  relying on flow classification, e.g. using integrated services [IntServ] · CPC title

• H04L47/24 Primary

  Traffic characterised by specific attributes, e.g. priority or QoS · CPC title

• H04L47/2483

  involving identification of individual flows · CPC title