Heuristics for selecting nearest zone based on ICA RTT and network latency

Patent metadata

Field	Value
Publication number	US-10476946-B2
Application number	US-201715661296-A
Country	US
Kind code	B2
Filing date	Jul 27, 2017
Priority date	Jul 27, 2017
Publication date	Nov 12, 2019
Grant date	Nov 12, 2019

Abstract

Described embodiments provide systems and methods for zone selection for distributed services. A device records latency data measured for interactions between each of a plurality of clients and a service hosted by servers in two or more zones. The device directs network communications from each of the plurality of clients to respective servers hosting the service based on zones assigned to each of the plurality of clients. The device assigns clients to zones based on the recorded latency data. For example, the device identifies a grouping for a client, determines whether the recorded latency data indicates that latency for clients in the grouping is increasing faster than a threshold rate, and selects, responsive to the determination, a zone indicated by a selected set of recorded latency data as lowest in latency.

First claim

What is claimed is:

1. A method of zone selection for distributed services, the method comprising:
recording, by a device comprising a processor, latency data measured for interactions between each of a plurality of clients and a service hosted by servers in two or more zones of a network;
routing, by the device, network communications from each of the plurality of clients to respective servers hosting the service via the network based on the zones assigned to each of the plurality of clients;
assigning, by the device, a first client to a first zone of the network by:
identifying a first grouping of clients for the first client,
determining, for the first client, that the recorded latency data indicates that a rate of change in latency for the clients in the first grouping is not increasing faster than a threshold rate, and
selecting, responsive to the determining for the clients in the first grouping, the first zone indicated by a first set of recorded latency data as lowest in latency; and

assigning, by the device, a second client to a second zone of the network by:
identifying a second grouping of clients for the second client,
determining, for the second client, that the recorded latency data indicates that a rate of change in latency for the clients in the second grouping is increasing faster than the threshold rate, and
selecting, responsive to the determining for the clients in the second grouping, the second zone indicated by a second set of the recorded latency data as lowest in latency, the second set differing from the first set of the recorded latency data.

2. The method of claim 1 , wherein the latency data indicates latency measured in terms of one of: network round trip time and service response time.

3. The method of claim 1 , wherein the latency data includes source network address ranges, average latency for each of the network address ranges, and zone assignments corresponding to an average latency for each of the network address ranges.

4. The method of claim 1 , wherein the threshold rate corresponds to a regression coefficient of 0.5.

5. The method of claim 1 , comprising selecting the second set of the recorded latency data based on time values associated with the recorded latency data.

6. The method of claim 5 , comprising selecting the second set of the recorded latency data to include only latency records recorded within a pre-defined time span.

7. The method of claim 1 , comprising:
identifying the first grouping for the first client based on a first network address range that includes the first client; and
identifying the second grouping for the second client based on a second network address range that includes the second client.

8. The method of claim 1 , wherein the second zone includes a server with a network path to the first client that is shorter than any network path between the first client and any server in the first zone.

9. The method of claim 1 , wherein the first set of latency data is latency data associated with the first grouping and the second set of latency data is associated with the second grouping.

10. The method of claim 1 , wherein the second set of latency data is a sub-set of the first set of latency data.

11. A system for zone selection for distributed services, the system comprising a processor configured to:
record latency data measured for interactions between each of a plurality of clients and a service hosted by servers in two or more zones of a network;
route network communications from each of the plurality of clients to respective servers hosting the service via the network based on the zones assigned to each of the plurality of clients; and
assign a first client to a first zone of the network by:
identifying a first grouping of clients for the first client,
determining, for the first client, that the recorded latency data indicates that a rate of change in latency for the clients in the first grouping is not increasing faster than a threshold rate, and
selecting, responsive to the determining for the clients in the first grouping, the first zone indicated by a first set of recorded latency data as lowest in latency; and

assign a second client to a second zone of the network by:
identifying a second grouping of clients for the second client,
determining, for the second client, that the recorded latency data indicates that a rate of change in latency for the clients in the second grouping is increasing faster than the threshold rate, and
selecting, responsive to the determining for the clients in the second grouping, the second zone indicated by a second set of the recorded latency data as lowest in latency, the second set differing from the first set of the recorded latency data.

12. The system of claim 11 , wherein the latency data indicates latency measured in terms of one of: network round trip time and service response time.

13. The system of claim 11 , wherein the latency data includes source network address ranges, average latency for each of the network address ranges, and zone assignments corresponding to an average latency for each of the network address ranges.

14. The system of claim 11 , wherein the threshold rate corresponds to a regression coefficient of 0.5.

15. The system of claim 11 , the processor further configured to select the second set of the recorded latency data based on time values associated with the recorded latency data.

16. The system of claim 15 , the processor configured to select the second set of the recorded latency data to include only latency records recorded within a pre-defined time span.

17. The system of claim 11 , the processor further configured to:
identify the first grouping for the first client based on a first network address range that includes the first client; and
identify the second grouping for the second client based on a second network address range that includes the second client.

18. The system of claim 11 , wherein the second zone includes a server with a network path to the first client that is shorter than any network path between the first client and any server in the first zone.

19. The system of claim 11 , wherein the first set of latency data is latency data associated with the first grouping and the second set of latency data is associated with the second grouping.

20. The system of claim 11 , wherein the second set of latency data is a sub-set of the first set of latency data.

Assignees

• Citrix Systems Inc

Inventors

• Kumar Kiran

Classifications

• H04L43/16

  Threshold monitoring · CPC title

• H04L43/04

  Processing captured monitoring data, e.g. for logfile generation · CPC title

• H04L67/101 Primary

  based on network conditions · CPC title

• H04L41/5096

  wherein the managed service relates to distributed or central networked applications · CPC title

• H04L43/0864

  Round trip delays · CPC title