Hot service flow hardware offloads based on service priority and resource usage

Patent metadata

Field	Value
Publication number	US-9596184-B1
Application number	US-201514925111-A
Country	US
Kind code	B1
Filing date	Oct 28, 2015
Priority date	Jul 23, 2012
Publication date	Mar 14, 2017
Grant date	Mar 14, 2017

Abstract

Embodiments are directed towards improving the performance of network traffic management devices by optimizing the management of hot connection flows. A packet traffic management device (“PTMD”) employs a data flow segment (“DFS”) and control segment (“CS”). The CS performs high-level control functions and per-flow policy enforcement for connection flows maintained at the DFS, while the DFS performs statistics gathering, and per-packet policy enforcement, on connection flows maintained at the DFS. The DFS may include high-speed flow caches and other high-speed components. Making efficient use of the high speed flow cache capacity may be improved by maximizing the number of hot connection flows based on unique service and network traffic characteristics through adaptive feedback pattern learning together with administrator configurable service preferences that may have flow control data for most bandwidth hungry and desired hot services offloaded to the high-speed flow cache, at appropriate time.

First claim

What is claimed is:

1. A method for managing communication over a network with a traffic management device (TMD) that includes a plurality of components that employ one or more processors to perform actions, comprising:
employing one or more control segment (CS) components to perform actions, including:
determining one or more directional-bound patterns and one or more service patterns for received network packets, wherein the one or more directional-bound patterns and the one or more service patterns are employed to identify one or more hot connection flows that are handled by one or more data flow segment (DFS) components at the network packet level; and
employing the directional bound patterns and service patterns to determine when one or more hot connection flows allocated for handling by the one or more DFS components is instead handled by the one or more CS components.

2. The method of claim 1 , wherein the one or more DFS components handle one or more of a plurality of connections flows, and wherein the one or more CS components manage the plurality of connection flows and handle a remainder portion of the plurality of connection flows that is non-handled by the one or more DFS components.

3. The method of claim 1 , further comprising generating the one or more directional-bound patterns that indicate one or more directional traffic distributions within one or more service types based on one or more received network packets for one or more of a plurality of connection flows.

4. The method of claim 1 , further comprising generating the one or more service patterns that indicate a traffic distribution among a plurality of services based on one or more received network packets for one or more of a plurality of connection flows.

5. The method of claim 1 , further comprising modifying the one or more hot connection flows based on one or more of a service priority, a service resource usage limit, or a prediction of the one or more service resource patterns.

6. The method of claim 1 , further comprising employing one or more of a metric or a rule to either move a hot connection flow from the one or more DFS components to the one or more CS components or move the hot connection flow from the one or more CS components to the one or more DFS components.

7. The method of claim 1 , further comprising when a capacity of a DFS component is reached, performing actions, including:
ranking each connection flow for a predetermined period of time based on an amount of network packets that are handled by the DFS component; and
identifying the one or more hot connection flows based on the ranking.

8. A network device, comprising:
a memory that is operative to store instructions; and
one or more processors that is operative to execute instructions that perform actions, including:
employing one or more control segment (CS) components to perform actions, including:
determining one or more directional-bound patterns and one or more service patterns for received network packets, wherein the one or more directional-bound patterns and the one or more service patterns are employed to identify one or more hot connection flows that are handled by one or more data flow segment (DFS) components at the network packet level; and
employing the directional bound patterns and service patterns to determine when one or more hot connection flows allocated for handling by the one or more DFS components is instead handled by the one or more CS components.

9. The network device of claim 8 , wherein the one or more DFS components handle one or more of a plurality of connections flows, and wherein the one or more CS components manage the plurality of connection flows and handle a remainder portion of the plurality of connection flows that is non-handled by the one or more DFS components.

10. The network device of claim 8 , further comprising generating the one or more directional-bound patterns that indicate one or more directional traffic distributions within one or more service types based on one or more received network packets for one or more of a plurality of connection flows.

11. The network device of claim 8 , further comprising generating the one or more service patterns that indicate a traffic distribution among a plurality of services based on one or more received network packets for one or more of a plurality of connection flows.

12. The network device of claim 8 , further comprising modifying the one or more hot connection flows based on one or more of a service priority, a service resource usage limit, or a prediction of the one or more service resource patterns.

13. The network device of claim 8 , further comprising employing one or more of a metric or a rule to either move a hot connection flow from the one or more DFS components to the one or more CS components or move the hot connection flow from the one or more CS components to the one or more DFS components.

14. The network device of claim 8 , further comprising when a capacity of a DFS component is reached, performing actions, including:
ranking each connection flow for a predetermined period of time based on an amount of network packets that are handled by the DFS component; and
identifying the one or more hot connection flows based on the ranking.

15. A non-transitory computer readable medium, having computer-executable instructions stored thereon, that in response to execution by one or more processors, causes a network device to perform operations, comprising:
employing one or more control segment (CS) components to perform actions, including:
determining one or more directional-bound patterns and one or more service patterns for received network packets, wherein the one or more directional-bound patterns and the one or more service patterns are employed to identify one or more hot connection flows that are handled by one or more data flow segment (DFS) components at the network packet level; and
employing the directional bound patterns and service patterns to determine when one or more hot connection flows allocated for handling by the one or more DFS components is instead handled by the one or more CS components.

16. The medium of claim 15 , further comprising modifying the one or more hot connection flows based on one or more of a service priority, a service resource usage limit, or a prediction of the one or more service resource patterns.

17. The medium of claim 15 , further comprising employing one or more of a metric or a rule to either move a hot connection flow from the one or more DFS components to the one or more CS components or move the hot connection flow from the one or more CS components to the one or more DFS components.

18. The medium of claim 15 , further comprising when a capacity of a DFS component is reached, performing actions, including:
ranking each connection flow for a predetermined period of time based on an amount of network packets that are handled by the DFS component; and
identifying the one or more hot connection flows based on the ranking.

19. The medium of claim 15 , further comprising generating the one or more directional-bound patterns that indicate one or more directional traffic distributions within one or more service types based on one or more received network packets for one or more of a plurality of connection flows.

20. The medium of claim 15 , further comprising generating the one or more service patterns that indicate a traffic distribution among a plurality of services based on one or more received network packets for o

Assignees

• F5 Networks Inc

Inventors

• Cai Hao
• Barthelow Michael Thomas
• Szabo Paul Imre
• Michels Timothy Scott

Classifications

• H04L47/10

  Flow control; Congestion control · CPC title

• H04L47/2441

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

• H04L47/125 Primary

  by balancing the load, e.g. traffic engineering · CPC title

• H04L45/38

  Flow based routing · CPC title

• H04L47/621

  Individual queue per connection or flow, e.g. per VC · CPC title