Distributed traffic management system and techniques

Patent metadata

Field	Value
Publication number	US-2016088020-A1
Application number	US-201414495631-A
Country	US
Kind code	A1
Filing date	Sep 24, 2014
Priority date	Sep 24, 2014
Publication date	Mar 24, 2016
Grant date	—

Abstract

Approaches, techniques, and mechanisms are disclosed for implementing a distributed firewall. In an embodiment, many different computer assets police incoming messages based on local policy data. This local policy data is synchronized with global policy data. The global policy data is generated by one or more separate analyzers. Each analyzer has access to message logs, or information derived therefrom, for groups of computer assets, and is thus able to generate policies based on intelligence from an entire group as opposed to an isolated asset. Among other effects, some of the approaches, techniques, and mechanisms may be effective even in computing environments with limited supervision over the attack surface, and/or computing environments in which assets may need to make independent decisions with respect to how incoming messages should be handled, on account of latency and/or unreliability in connections to other system components.

First claim

What is claimed is:

1 . A computer system comprising:
a data repository storing global policy data that describes policies; a plurality of computer assets, implemented at least partially by first computer hardware, each asset of the plurality of computer assets configured to: receive messages from client devices; store local policy data describing the policies at a computing device that implements the asset; determine which of the policies apply to which of the messages; identify policy-based actions to perform with respect to the messages based on which of the policies apply to which of the messages; send message information logged from the messages to an analyzer component; and update the local policy data to reflect updates to the global policy data; and an analyzer component, implemented at least partially by second computer hardware, configured to: receive the message information from each of the plurality of computer assets; collectively analyze the message information from each of the plurality of computer assets; generate new policies based on collectively analyzing the message information; and update the global policy data to describe the new policies.

2 . The computer system of claim 1 , wherein each asset of the plurality of computer assets is deployed at an edge of a first network, and wherein the client devices are deployed within a second network.

3 . The computer system of claim 1 , wherein the plurality of computer assets and the analyzer are deployed in a first region of the computer system, the computer system further comprising one or more additional regions, each region comprising a separate plurality of computer assets and a separate analyzer, the data repository being shared between the first region and the one or more additional regions.

4 . The computer system of claim 1 , wherein each policy of the policies is a data structure that indicates logic for determining whether the policy applies to a given message, as well as one or more instructions indicating one or more particular policy-based actions to perform with respect to the given message if the policy applies to the given message.

5 . The computer system of claim 1 , wherein the analyzer is further configured to: identify one or more of the new policies based on determining, from the collectively analyzed message information, that a condition described by a system-level policy exists; generate one or more asset-level policies comprising logic for identifying messages to block or redirect in view of the described condition; and update the global policy data to include the generated one or more asset-level policies.

6 . The computer system of claim 1 , wherein the analyzer is further configured to: identify a distributed attack on the computer system based on the collectively analyzed message information; generate a first policy comprising logic for identifying messages involved in the distributed attack; and update the global policy data to describe the first policy.

7 . The computer system of claim 1 ,
wherein the analyzer is further configured to, based on the collectively analyzed message information: identify an attack that is occurring at a first asset of the plurality of computer assets; generate a first policy comprising logic for identifying messages involved in the attack; and update the global policy data to describe the first policy; wherein a second asset of the plurality of computer assets is configured to: based on the updated global policy data, update the local policy data of the second asset to describe the first policy, the second asset having not yet received messages involved in the attack at a time the local policy data of the second asset is updated; and, based on the first policy, block or redirect a message involved in the attack.

8 . The computer system of claim 1 , wherein the messages indicate designated actions for the plurality of assets to perform, wherein each asset is configured to perform the applicable policy-based actions instead of or in addition to the indicated designated actions for messages to which the policies apply.

9 . The computer system of claim 1 , wherein a policy-based action indicated by a given policy of the policies is one of: blocking any message to which the given policy applies, redirecting any message to which the given policy applies, or allowing an asset to respond normally to any message to which the given policy applies.

10 . The computer system of claim 1 , wherein an asset of the plurality of computer assets is configured to apply policies, from the local policy data of the asset, to messages, even when the analyzer component and the data repository are inaccessible to the asset.

11 . A data processing method comprising:
storing, at a computer asset, local policy data describing policies; receiving, at the computer asset, messages from client devices; determining, at the computer asset, which of the policies apply to which of the messages; identifying, at the computer asset, policy-based actions to perform with respect to the messages based on which of the policies apply to which of the messages; sending, from the computer asset, message information logged from the messages to an analyzer component; and updating, by the computer asset, the local policy data to reflect updates to global policy data generated by the analyzer component; wherein the computer asset is implemented by one or more computing devices.

12 . The method of claim 11 , wherein each policy of the policies is a data structure that indicates logic for determining whether the policy applies to a given message, as well as one or more instructions indicating one or more particular policy-based actions to perform with respect to the given message if the policy applies.

13 . The method of claim 11 ,
wherein the messages indicate designated actions for the plurality of computer assets to perform; the method further comprising performing applicable policy-based actions instead of or in addition to the designated actions indicated for messages to which the policies apply.

14 . The method of claim 11 , further comprising applying policies from the local policy data to messages even when the analyzer component and the global policy data are inaccessible to the asset.

15 . The method of claim 11 , further comprising:
receiving, at the analyzer component, message information logged at each asset of a plurality of computer assets, including the computer asset; collectively analyzing, by the analyzer component, the message information logged at each of the plurality of computer assets; generating, by the analyzer component, new policies based on collectively analyzing the message information; updating, by the analyzer component, the global policy data to describe the new policies.

16 . A data processing method comprising:
receiving message information from each asset of a plurality of computer assets, the message information describing messages received by the asset; collectively analyzing the message information from each of the plurality of computer assets; generating policies based on collectively analyzing the message information, the policies describing logic for identifying messages to which the policies apply, as well as policy-based actions to perform with respect to the messages to which the policies apply; updating a global policy data repository to describe the policies; sending policy data to the plurality of computer assets, the policy data at least describing the policies in the global policy data repository; wherein the method is performed

Assignees

• Netflix Inc

Inventors

• Chan Jason
• Udupi Poornaprajna
• Madappa Shashi

Classifications

• H04L67/10

  in which an application is distributed across nodes in the network (software deployment G06F8/60; multiprogramming arrangements G06F9/46) · CPC title

• G06F16/22

  Indexing; Data structures therefor; Storage structures · CPC title

• H04L63/1408 Primary

  by monitoring network traffic (monitoring network traffic per se H04L43/00) · CPC title

• H04L63/0218

  Distributed architectures, e.g. distributed firewalls · CPC title

• H04L63/0227

  Filtering policies (mail message filtering H04L51/212) · CPC title