NETWORK HEALTH DATA AGGREGATION SERVICE

• US 20180091413A1
• Filed: 09/28/2016
• Published: 03/29/2018
• Est. Priority Date: 09/28/2016
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Richards, Kenneth Grey, Thompson, Schuyler David, Siefker, Adam, MILLER, KEVIN CHRISTOPHER, Rameshkumar, Meenakshi
• Granted Patent: US 10,917,324 B2
• Time in Patent Office: Days
• CPC Class Codes: H04L 12/4641 Virtual LANs, VLANs, e.g. virtual private networks [VPN] LAN interconnection over a bridge based backbone H04L12/462; encapsulation techniques H04L12/4633; routing of packets H04L45/00; packet switches H04L49/00; virtual private networks for security H04L63/0272
• CPC Class Codes: H04L 43/08 Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
• CPC Class Codes: H04L 43/0811 by checking connectivity
• CPC Class Codes: H04L 43/0829 Packet loss
• CPC Class Codes: H04L 43/0852 Delays
• CPC Class Codes: H04L 43/14 using software, i.e. software packages network security related monitoring H04L63/1408
• CPC Class Codes: H04L 43/20 the monitoring system or the monitored elements being virtualised, abstracted or software-defined entities, e.g. SDN or NFV

Abstract

Respective network metrics sets corresponding to one or more data sources are examined at a network health manager. Network health states corresponding to one or more endpoint pair categories are determined based on the analysis of the network metric sets. An indication of the network health state of a particular endpoint pair category is stored.

Claims

1. A system, comprising:

one or more computing devices of a network health manager associated with a virtualized computing service of a provider network;

wherein the one or more computing devices are configured to;

identify one or more endpoint pair categories corresponding to resources associated with a customer of the virtualized computing service, wherein a path between endpoints of an endpoint pair category of the one or more endpoint pair categories comprises a virtual representation of one or more physical network links, and wherein a particular endpoint pair category of the one or more endpoint pair categories comprises an endpoint within an isolated virtual network;

obtain respective sets of network metrics from one or more sources ;

determine, based at least in part on the respective sets of network metrics, a respective network health state corresponding to the one or more endpoint pair categories, wherein a particular set of network metrics of the respective sets of network metrics is obtained from a physical device used for network traffic associated with a corresponding endpoint pair category; and

transmit an indication of at least the network health state corresponding to the particular endpoint pair category to a selected destination.

2. The system as recited in claim 1, wherein the one or more computing devices are configured to:

in response to detecting, based at least in part on a particular set of network metrics from a first source of the one or more sources, a network health impairment with respect to the particular endpoint pair category,

verify, prior to transmitting an indication of impaired network health state of the particular endpoint pair category, that a different set of network metrics confirms the network health impairment.

3. The system as recited in claim 1, wherein the indication of network health state corresponding to the particular endpoint pair category indicates a degree of impairment of network traffic of the particular endpoint pair category.

4. The system as recited in claim 1, wherein a particular set of network metrics of the plurality of sets of network metrics includes one or more of:

(a) a request-response success rate, (b) a message latency, (c) a packet loss rate or (d) a metric indicating a variation in message latency over time.

5. The system as recited in claim 1, wherein the one or more computing devices are configured to:

assign respective weights to individual ones of the respective sets of network metrics, wherein a particular weight assigned to a particular set of the network metrics is based at least in part on one or more of;

(a) a physical location of a resource to which the particular set corresponds, (b) a layer of a networking stack to which the particular set corresponds, or (c) a collection time of at least one metric of the particular set; and

utilize at least one weight of the respective weights to determine the network health state corresponding to the particular endpoint pair category.

6. The system as recited in claim 1, wherein the one or more sources comprise (a) a connectivity verifier and (b) an infrastructure monitor, wherein the connectivity verifier comprises a node instantiated at a particular virtual machine of the virtualized computing service, and wherein the infrastructure monitor is configured to collect a particular network metric of a particular network metric set from a private networking device of the provider network.

7. A method, comprising:

performing, by one or more computing devices;

obtaining respective sets of network metrics corresponding to one or more data sources;

determining, based at least in part on the respective sets of network metrics, a respective network health state corresponding to one or more endpoint pair categories; and

storing an indication of at least the network health state corresponding to a particular endpoint pair category of the one or more endpoint pair categories.

8. The method as recited in claim 7, further comprising performing, by the one or more computing devices:

assigning respective weights to individual ones of a plurality of sets of network metrics;

wherein said determining the respective network health state comprises utilizing the respective weights.

9. The method as recited in claim 8, wherein a particular weight assigned to a particular set of the plurality of sets of network metrics is based at least in part on one or more of:

(a) a physical location of a resource to which the particular set corresponds, (b) a layer of a networking stack to which the particular set corresponds, or (c) a collection time of at least one metric of the particular set.

10. The method as recited in claim 7, wherein the respective sets of network metrics comprise a first set of network metrics indicative of connectivity between a first resource and a second resource, wherein the first resource comprises a first virtual machine, and wherein the second resource comprises a second virtual machine.

11. The method as recited in claim 10, wherein the first virtual machine is configured within a first isolated virtual network established on behalf of a particular client.

12. The method as recited in claim 7, wherein the particular endpoint pair category is associated with a particular client account of a provider network, the method further comprising performing, by the one or more computing devices:

identifying one or more network-accessible services being utilized on behalf of the particular client account;

determining that a particular set of network metrics of the respective sets of network metrics pertains to a physical device being used for network traffic associated with a particular service of the one or more network-accessible services; and

utilizing the particular set of network metrics to determine the health state of the particular endpoint pair category.

13. The method as recited in claim 7, wherein the respective sets of network metrics comprise a first set of network metrics indicative of connectivity between an endpoint accessible via the public Internet and another endpoint.

14. The method as recited in claim 7, wherein the respective sets of network metrics comprise a first set of network metrics indicative of connectivity between a first resource of a first network-accessible service of a provider network and a second resource of a second network-accessible service of the provider network.

15. The method as recited in claim 7, wherein obtaining a particular set of network metrics of the respective sets of network metrics comprises utilizing one or more of:

(a) a tool configured to issue Domain Name Service (DNS) queries (b) a tool configured to run Virtual Private Network (VPN) tests or (c) a tool configured to run tests which utilize a dedicated physical link established between the provider network and a customer networking device.

16. The method as recited in claim 7, wherein the particular endpoint pair category comprises one of:

(a) a first endpoint within one availability container of the virtualized computing service and a second endpoint within a different availability container of the virtualized computing service, (b) a first endpoint within one geographical region and a second endpoint within a different geographical region, (c) a first endpoint within a virtualized computing service and a second endpoint at a customer data center, (d) a first endpoint within a virtualized computing service and a second endpoint within the public Internet (e) a first endpoint within a virtualized computing service and a second endpoint within a different network-accessible service, or (f) a first endpoint within a first isolated virtual network and a second endpoint outside the first isolated virtual network.

17. A non-transitory computer-accessible storage medium storing program instructions that when executed on one or more processors:

examine a respective sets of network metrics corresponding to one or more data sources;

determine, based at least in part on the examination of the first and second sets of network metrics, a respective network health state corresponding to one or more endpoint pair categories; and

store an indication of at least the network health state corresponding to a particular endpoint pair category of the one or more endpoint pair categories.

18. The non-transitory computer-accessible storage medium as recited in claim 17, wherein a particular set of network metrics of the respective sets of network metrics comprises a metric obtained from a non-public networking device, wherein the non-public networking device comprises one or more of:

(a) a router, (b) a gateway, (c) a switch, (d) a network interface card (NIC), (e) a physical link connecting a portion of the provider network to an external network or (f) an encapsulation protocol processing component.

19. The non-transitory computer-accessible storage medium as recited in claim 17, wherein the instructions when executed on the one or more processors:

analyze resource allocation metadata pertaining to a particular customer of a virtualized computing service to identify the one or more endpoint pair categories, wherein the resource allocation metadata indicates at least one isolated virtual network established on behalf of the particular customer.

20. The non-transitory computer-accessible storage medium as recited in claim 17, wherein to determine the respective network health state, the instructions when executed on the one or more processors:

utilize respective weights assigned to individual ones of a plurality of sets of network metrics.

21. The non-transitory computer-accessible storage medium as recited in claim 20, wherein a particular weight assigned to a particular set of the plurality of sets of network metrics is based at least in part on one or more of:

(a) a physical location of a resource to which the particular set corresponds, (b) a layer of a networking stack to which the particular set corresponds, or (c) a collection time of at least one metric of the particular set.