End-Host Based Network Management System

• US 20090290491A1
• Filed: 05/22/2008
• Published: 11/26/2009
• Est. Priority Date: 05/22/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Key, Peter, Gkantsidis, Christos, Karagiannis, Thomas, Athanasopoulos, Elias
• Granted Patent: US 8,059,541 B2
• Time in Patent Office: Days
• US Class Current: 370/235
• CPC Class Codes: H04L 41/14 Network analysis or design
• CPC Class Codes: H04L 43/026 using flow identification
• CPC Class Codes: H04L 43/0817 by checking functioning
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/125 by balancing the load, e.g. traffic engineering

Abstract

An end-host based network management system and methods are described. The methods are performed independently at each end-host within the network based on data on local flows which is shared between end-hosts. In an embodiment, an end-host shares data on constrained local flows with other end-hosts and receives such data from other end-hosts. Based on this data, the end-host determines which flows from other nodes are competing for a shared resource with a constrained local flow and allocates the capacity of the shared resource between all the competing flows. This allocation is then enforced for the local flow by the end-host. Other end-hosts with competing flows perform similar methods and through an iterative process the contention for the shared resource is resolved and the utilization of the shared resource is optimized.

Claims

1. One or more device-readable media with device-executable instructions for performing steps comprising:

collecting data on local network activity;

sharing data with other end-hosts within a network; and

controlling local network activity based on analysis of data on local network activity and data on network activity received from the other end-hosts within the network.

2. One or more device-readable media according to claim 1, wherein collecting data on local network activity comprises:

collecting data on local flows; and

monitoring performance of the local flows.

3. One or more device-readable media according to claim 1, wherein sharing data with other end-hosts within a network comprises:

detecting a performance change in the local network activity; and

sending data relating to the performance change to other end-hosts within the network.

4. One or more device-readable media according to claim 3, wherein detecting a performance change in the local network activity comprises:

monitoring a base performance of a local flow;

monitoring a current performance of the local flow; and

detecting a performance change when a difference between the base performance and the current performance exceeds a threshold.

5. One or more device-readable media according to claim 4, wherein said data relating to the performance change comprises an indication of a type of performance change, wherein said type comprises one of:

a new flow, an increase in performance and a decrease in performance.

6. One or more device-readable media according to claim 1, wherein controlling local network activity based on analysis of data on local network activity and data on network activity received from the other end-hosts within the network comprises:

identifying competing flows within the network; and

allocating resources to local competing flows based on the data on local network activity and data received from other end-hosts.

7. One or more device-readable media according to claim 6, wherein identifying competing flows within the network comprises:

on detection of a performance change in a local flow, determining if a reduction in performance occurred in a second flow within the network within a defined time period; and

performing a pairwise correlation between data for the local flow and data for the second flow.

8. One or more device-readable media according to claim 6, wherein allocating resources to local competing flows based on the data on local network activity and data received from other end-hosts comprises:

setting a rate limit for each local competing flow;

determining an actual rate for each local competing flow; and

adjusting the rate limits to achieve a target utilization of a shared resource.

9. One or more device-readable media according to claim 8, wherein the competing flows relate to a shared resource and wherein allocating resources to local competing flows based on the data on local network activity and data received from other end-hosts further comprises:

estimating a capacity of the shared resource.

10. One or more device-readable media according to claim 8, wherein the rate limit for each local competing flow is set based on a weight for the flow and wherein the weight is set based on at least one of:

user input and flow type.

11. One or more device-readable media according to claim 6, wherein controlling local network activity based on analysis of data on local network activity and data on network activity received from the other end-hosts within the network further comprises:

enforcing an allocation of resources on local competing flows.

12. A network comprising a plurality of end-hosts, wherein each end-host is arranged to independently perform network management of local flows based on information shared between the end-hosts.

13. A network according to claim 12, wherein the network further comprises a shared resource and wherein each end-host is arranged to independently operate a virtual queue for the shared resource.

14. A network according to claim 12, wherein each end-host comprises:

a processor;

a network interface; and

a memory arranged to store executable instructions to cause the processor to;

collect data on local data flows transmitted and received via the network interface;

transmit data relating to at least one of the local data flows to other end-hosts within the network;

receive data on remote data flows in the network from other end-hosts within the network; and

control at least one of the local data flows based on said transmitted and received data.

15. A network according to claim 14, wherein the executable instructions to cause the processor to transmit data relating to at least one of the local data flows to other end-hosts within the network comprise executable instructions to cause the processor to:

detect a change in performance in a local data flow; and

transmit data on said local data flow to other end-hosts within the network.

16. A network according to claim 15, wherein the executable instructions to cause the processor to detect a change in performance in a local data flow comprise executable instructions to cause the processor to:

monitor a long term average performance of a local data flow;

monitor a current performance of the local data flow; and

detect a change in performance in the local data flow when a difference between the current performance and the long term average performance exceeds a threshold.

17. A network according to claim 14, wherein the executable instructions to cause the processor to control at least one of the local data flows based on said transmitted and received data comprise executable instructions to cause the processor to:

identify data flows within the network which are competing for access to a shared resource in the network;

estimate a capacity of the shared resource based on data received from other end-hosts; and

allocate a portion of the capacity to at least one the identified local data flows.

18. A network according to claim 17, wherein the executable instructions to cause the processor to control at least one of the local data flows based on said transmitted and received data further comprise executable instructions to cause the processor to:

enforce a capacity allocation on at least one of the local data flows.

19. A network management method comprising:

collecting data on local connections at each of a plurality of end-hosts;

sharing data on constrained connections between the plurality of end-hosts; and

performing independent network control of the local connections at each of the plurality of end-hosts.

20. A network management method according to claim 19, wherein performing independent network control of the local connections at each of the plurality of end-hosts comprises, at each of the plurality of end-hosts:

identifying competing connections from the constrained connections;

allocating capacity between said competing connections; and

enforcing the capacity allocation on local connections.