SYSTEMS AND METHODS FOR MANAGING DYNAMIC PROXIMITY IN MULTI-CORE GSLB APPLIANCE

• US 20110153723A1
• Filed: 12/23/2009
• Published: 06/23/2011
• Est. Priority Date: 12/23/2009
• Status: Active Grant

• Current Assignee: Citrix Systems, Inc. (Cloud Software Group)
• Original Assignee: Citrix Systems, Inc. (Cloud Software Group)
• Inventors: Kamath, Sandeep, Mutnuru, Rishi, Narayana, Raghav Somanahalli
• Granted Patent: US 8,230,054 B2
• Time in Patent Office: Days
• US Class Current: 709/203
• CPC Class Codes: H04L 61/4511 using domain name system [DNS]
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 69/16 Implementation or adaptation of Internet protocol [IP], of transmission control protocol [TCP] or of user datagram protocol [UDP]

Abstract

The present invention is directed towards systems and methods for providing dynamic proximity load balancing via a multi-core intermediary device. An intermediary device providing global server load balancing (GSLB) identifies a local domain name service (LDNS) entries database and assigns each LDNS entry in the LDNS entries database to one of the plurality of packet processing engine base on a source internet protocol (IP) address of each LDNS entry. The first packet processing engine on the appliance receives a LDNS request for an IP address, determines that the LDNS entry for the IP address is assigned to a second packet processing engine of the plurality of packet processing engines, transmits a request to the second packet processing engine for the LDNS entry for the IP address, and determines a response to the LDNS request based on the LDNS entry for the IP address received from the second packet processing engine.

Claims

1. A method of providing dynamic proximity load balancing via a multi-core intermediary device, the method comprising:

(a) identifying, by an intermediary device providing global server load balancing (GSLB), a local domain name service (LDNS) entries database, the intermediary device comprising a plurality of packet processing engines executing on a corresponding core of a plurality of cores of the intermediary device;

(b) assigning, by the intermediary device, each LDNS entry in the LDNS entries database to one of the plurality of packet processing engine based on a source internet protocol (IP) address of each LDNS entry;

(c) receiving, by a first packet processing engine of a plurality of packet processing engines, an LDNS request for an IP address;

(d) determining, by the first packet processing engine, that the LDNS entry for the IP address is assigned to a second packet processing engine of the plurality of packet processing engines;

(e) transmitting, by the first packet processing engine, a request to the second packet processing engine for the LDNS entry for the IP address; and

(f) determining, by the first packet processing engine, a response to the LDNS request based on the LDNS entry for the IP address received from the second packet processing engine.

2. The method of claim 1, wherein step (b) further comprises storing, by the first packet processing engine of the plurality of packet processing engines operating on a first core of the plurality of cores, in memory for the first packet processing engine a set of LDNS entries assigned to the first packet processing engine.

3. The method of claim 2, wherein step (b) further comprises storing, by the first packet processing engine, in a cache memory for the first packet processing engine a set of LDNS entries assigned to one or more of the other packet processing engines.

4. The method of claim 1, wherein step (c) further comprises determining, by the first packet processing, that the LDNS entry for the IP address is not stored in a memory cache of the first packet processing engine.

5. The method of claim 1, wherein step (e) further comprises transmitting the request to the second packet processing engine via first core to second core messaging.

6. The method of claim 1, wherein step (f) further comprises determining a domain name for a geographically distributed server farm based on one or more round trip times for the LDNS entry of the IP address.

7. The method of claim 1, further comprising establishing a monitor for each LDNS entry, the monitor determining a round trip time.

8. The method of claim 7, further comprising determining a source port for monitor communications that result in a response being received by a core that is assigned the LDNS entry.

9. The method of claim 7, further comprising determining, by a flow distributor of the intermediary device via packet steering, to forward a response received for the monitor to a core that is assigned the LDNS entry.

10. The method of claim 1, further comprising determining, by the first packet processing engine, that a round trip time for one or more LDNS entries assigned to the first packet processing engine has changed and transmitting the one or more LDNS entries to one or more of the other packet processing engines.

11. A system of providing dynamic proximity load balancing via a multi-core intermediary device, the system comprising:

an intermediary device comprising a plurality of packet processing engines executing on a corresponding core of a plurality of cores and providing global server load balancing, wherein the intermediary device identifies a local domain name service (LDNS) entries database and assigns each LDNS entry in the LDNS entries database to one of the plurality of packet processing engine based on a source internet protocol (IP) address of each LDNS entry;

a first packet processing engine of the plurality of packet processing engines that receives a LDNS request for an IP address, determines that the LDNS entry for the IP address is assigned to a second packet processing engine, transmits a request to the second packet processing engine for the LDNS entry for the IP address, and determines a response to the LDNS request based on the LDNS entry for the IP address received from the second packet processing engine.

12. The system of claim 11, wherein the first packet processing engine of the plurality of packet processing engines operating on a first core of the plurality of cores stores in memory for the first packet processing engine a set of LDNS entries assigned to the first packet processing engine.

13. The system of claim 12, wherein the first packet processing engine stores in a memory cache for the first packet processing engine a set of LDNS entries assigned to one or more of the other packet processing engines.

14. The system of claim 11, wherein the first packet processing determines that the LDNS entry for the IP address is not stored in a memory cache of the first packet processing engine.

15. The system of claim 11, wherein the first packet processing engine transmits the request to the second packet processing engine via a first core to second core messaging.

17. The system of claim 11, wherein the intermediary device establishes a monitor for each LDNS entry, the monitor determining a round trip time.

18. The system of claim 17, wherein the intermediary device determines a source port for monitor communications that result in a response being received by a core that is assigned the LDNS entry.

19. The system of claim 17, wherein a flow distributor of the intermediary device determines via packet steering to forward a response received for the monitor to a core that is assigned the LDNS entry.

20. The system of claim 11, wherein the first packet processing engine determines that a round trip time for one or more LDNS entries assigned to the first packet processing engine has changed and transmits the one or more LDNS entries to one or more of the other packet processing engines.

16. The system of claim 111, wherein the first packet processing engine determines a domain name for a geographically distributed server farm based on one or more round trip times for the LDNS entry of the IP address.