Network load balancing with session information

• US 7,613,822 B2
• Filed: 06/30/2003
• Issued: 11/03/2009
• Est. Priority Date: 06/30/2003
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Sivathanup, Karthic Nadarajapillai, Joy, Joseph M.
• Primary Examiner(s): Bates; Kevin
• Application Number:
  US10/610,321
• Publication Number:
  US 20040268357A1
• Time in Patent Office: 2,318 Days
• Field of Search: 709/227, 709/235, 709/238
• US Class Current: 709/235
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30
• CPC Class Codes: H04L 67/1029 using data related to the state of servers by a load balancer
• CPC Class Codes: H04L 67/14 Session management for real-time applications in data packet communications networks H04L65/1066
• CPC Class Codes: H04L 67/63 Routing a service request depending on the request content or context
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]
• CPC Class Codes: H04L 9/40 Network security protocols

Abstract

In a first exemplary implementation, one or more processor-accessible media include processor-executable instructions that, when executed, direct a system to perform actions including: receiving host status information from multiple hosts; and making load balancing decisions responsive to the received host status information. In a second exemplary implementation, a system includes: session tracking infrastructure that is adapted to acquire session information; and load balancing infrastructure that is adapted to utilize the session information when routing connection requests to multiple hosts. In a third exemplary implementation, one or more processor-accessible media include processor-executable instructions that, when executed, direct a system to perform actions including: acquiring session information for multiple session contexts from one or more applications that established the multiple session contexts; and routing network traffic to the one or more applications responsive to the acquired session information.

Claims

1. One or more computer-readable storage media comprising computer-executable instructions stored thereon, the computer-executable instructions, when executed, configuring a system to perform actions comprising:

acquiring session information for a plurality of session contexts from one or more applications that established the plurality of session contexts, wherein the acquired session information is received from a plurality of hosts;

caching the acquired session information in a distributed table, wherein;

the distributed table utilizes a session identifier for each of the established session contexts acquired in the session information, each session identifier representing an entry in the distributed table; and

each entry identifies an initial host where respective session information was first established;

routing network traffic to the one or more applications responsive to the acquired session information to perform load balancing;

in association with one of a plurality of replication levels, replicating session information for a particular session context of the plurality of session contexts at one or more additional devices based on a hash value derived from a session indicator for the particular session context; and

utilizing the one or more additional devices to preserve session affinity for the particular session context, such that when a client attempts to reestablish the particular session context with the initial host hosting a particular application, the client is mapped to the initial host via a particular session identifier.

2. The one or more computer-readable storage media as recited in claim 1, wherein the action of acquiring comprises an action of:

acquiring session information that indicates when the plurality of session contexts are established and terminated by the one or more applications.

3. The one or more computer-readable storage media as recited in claim 1, wherein the action of acquiring comprises an action of:

acquiring the session information for the plurality of session contexts from the one or more applications via notifications of an application programming interface (API).

4. The one or more computer-readable media as recited in claim 1, wherein the action of acquiring comprises an action of:

acquiring the session information for the plurality of session contexts from session tracking infrastructure via messages of a communications protocol.

5. The one or more computer-readable storage media as recited in claim 1, wherein the computer-executable instructions further configure the system to perform an action comprising:

replicating the session information across a plurality of load balancing units.

6. The one or more computer-readable storage media as recited in claim 1, wherein the computer-executable instructions further configure the system to perform actions comprising:

receiving a connection request that references a particular session context of the plurality of session contexts; and

consulting the cached session information for the connection request;

wherein the action of routing comprises an action of routing the network traffic to the one or more applications responsive to the consulting.

7. The one or more computer-readable storage media as recited in claim 6, wherein the action of caching comprises at least one action of:

calling an add session information function to add the received session information to the distributed table.

8. The one or more computer-readable storage media as recited in claim 1, wherein at least a portion of the computer-executable instructions comprise at least part of a load balancing software program.

9. The one or more computer-readable storage media as recited in claim 1, wherein at least a portion of the computer-executable instructions are adapted to run on the system wherein the system comprises a single device.

10. The one or more computer-readable storage media as recited in claim 1, wherein at least a portion of the computer-executable instructions are adapted to run on the system wherein the system comprises a plurality of devices.

11. The one or more computer-readable storage media as recited in claim 1, wherein the action of routing comprises an action of:

routing the network traffic to the one or more applications responsive to the acquired session information in order to enable a client to continue the particular session context of the plurality of session contexts at the initial host that includes the application of the one or more applications that established the particular session context of the plurality of session contexts.

12. A method comprising:

acquiring session information for a plurality of session contexts from one or more applications that established the plurality of session contexts, wherein the acquired session information is received from a plurality of hosts, each host being connected to a network;

caching the acquired session information in a first distributed table located at a first load balancing unit, wherein;

the first distributed table utilizes a session identifier for each of the established session contexts acquired in the session information, each session identifier representing an entry in the first distributed table; and

each entry identifies an original host where respective session information was first established;

replicating the session information for a particular session context of the plurality of session contexts at one or more additional load balancing units, each load balancing unit comprising a distributed table;

communicating one or more function calls, wherein the one or more function calls are communicated between the first load balancing unit and at least one of the one or more load balancing units in order to manipulate one or more entries stored in a particular distributed table in association with a replication level;

receiving a connection request that references the particular session context of the plurality of session contexts;

consulting the cached session information for the connection request, wherein the consulting attempts to identify an entry via the session identifier; and

routing the network traffic to the one or more applications responsive to the consulting, wherein the connection request is routed to the original host identified by the entry.

13. A method as recited in claim 12, wherein routing the connection request to the original host identified by the entry preserves session affinity.

14. A method as recited in claim 13, wherein the first load balancing unit assigns a higher priority to preserving the session affinity compared to load-based network load balancing decisions.

15. One or more computer-storage media comprising computer executable instructions that perform the method as recited in claim 12.

16. One or more computer-readable storage media comprising computer-executable instructions stored thereon, the computer-executable instructions, when executed, configuring a system to perform actions comprising:

acquiring session information for a plurality of session contexts from one or more applications that established each of the plurality of session contexts, wherein the acquired session information is received from a plurality of hosts, each host being connected to a network;

caching the acquired session information in a first distributed table located at a first load balancing unit, wherein;

the first distributed table utilizes a session identifier for each of the established session contexts acquired in the session information, each session identifier representing an entry in the first distributed table; and

each entry identifies an original host where a respective session context was first established;

replicating the session information for a particular session context of the plurality of session contexts in association with a replication level, wherein the replication level is selected from a plurality of replication levels, the plurality of replication levels comprising;

a first level, wherein the session information for the particular session context is only replicated at a second load balancing unit comprising a second distributed table with a plurality of entries, and is not replicated at a third load balancing unit comprising a third distributed table with a plurality of entries;

a second level, wherein the session information for the particular session context is only replicated at the second load balancing unit comprising the second distributed table with a plurality of entries and the third load balancing unit comprising the third distributed table with a plurality of entries, and is not replicated at a fourth load balancing unit comprising a fourth distributed table with a plurality of entries;

a third level, wherein the session information for the particular session context is replicated at three or more additional load balancing units, each load balancing unit comprising a distributed table with a plurality of entries; and

a fourth level, wherein the session information for the particular session context is replicated at all available load balancing units, each available load balancing unit comprising a distributed table with a plurality of entries;

communicating one or more function calls, wherein the one or more function calls are communicated between any two load balancing units to manipulate and identify one or more entries stored thereon, wherein the plurality of entries stored on each of the two load balancing units are not the same;

receiving, at the first load balancing unit, a connection request; and

consulting the first distributed table for the connection request, wherein the consulting attempts to identify an entry associated with the connection request via the session identifier;

in an event the consulting identifies the entry in the first distributed table located at the first load balancing unit, routing the network traffic to the one or more applications associated with the identified entry responsive to the consulting, wherein the connection request is routed to the original host identified by the entry; and

in an event the consulting does not identify an entry in the first distributed table located at the first load balancing unit, initiating a function call to one or more other load balancing units in order to identify an entry associated with the connection request.

17. One or more computer-readable storage media as recited in claim 16, wherein routing the connection request to the original host identified by the entry preserves session affinity.

18. One or more computer-readable storage media as recited in claim 17, wherein the first load balancing unit assigns a higher priority to preserving the session affinity compared to load-based network load balancing decisions.