A 10 Networks, Incorporated

Feilong Xu

Rajkumar Ramniranjan Jalan

Ronald Wai Lun Szeto

Lalgudi Narayanan Kannan

Lee Chen

Rishi Sampat

Chih-Wei Chao

Rakjumar Jalan

Ronald Wei Lun Szeto

Provided are methods and systems for distributing application traffic. A method for distributing application traffic may commence with relaying a first service request for a first service session from a service gateway to a server. The first service request may be received from a host and may be associated with a service request time. The method may further include receiving, from the server, a service response. The service response may be associated with a service response time. The method may continue with calculating a service processing time based on the service request time and the service response time and comparing the service processing time with an expected service processing time. The method may further include receiving, from the host, a second service request for a second service session. The method may continue with selectively relaying the second server request to the server based on the service processing time.

Distributing application traffic to servers based on dynamic service response time

The processing of data packets sent over a communication session between a host and a server by a service gateway includes processing a data packet using a current hybrid-stateful or hybrid-stateless processing method. The processing then checks whether a hybrid-stateless or hybrid-stateful condition is satisfied. When one of the sets of conditions is satisfied, the process includes changing from a hybrid-stateful to a hybrid-stateless processing method, or vice versa, for a subsequently received data packet. If the conditions are not satisfied, the process continues as originally structured.

Combining stateless and stateful server load balancing

Provided are methods and systems for balancing servers based on a server load status. A method for balancing servers based on a server load status may commence with receiving, from a server of a plurality of servers, a service response to a service request. The service response may include a computing load of the server. The method may continue with receiving a next service request from a host. The method may further include determining, based on the computing load of the server, whether the server is available to process the next service request. The method may include selectively sending the next service request to the server based on the determination that the server is available to process the next service request.

System and method to balance servers based on server load status

A system and method for providing a network proxy layer are disclosed. The network proxy layer may receive a connection establishment event for a client connection of an application session and send the client connection event to an application proxy for the application session, the application proxy being associated with an application of a server. Upon establishment of the client connection, the network proxy layer may receive one or more data packets from the client connection. The network proxy layer may further receive a connection establishment event for a server connection of the application session of the server, and receive one or more data packets from the server connection.

Network proxy layer for policy-based application proxies

Provided are methods and systems for distributing application traffic. A method for distributing application traffic may commence with receiving, from a host, a first service request for a first service session. The first service request may be associated with a service request time. The method may continue with relaying the first service request from a service gateway to a server. The method may further include receiving, from the server, a service response. The service response may be associated with a service response time. The method may continue with calculating a service processing time for the first service request based on the service request time and the service response time. The method may further include receiving, from the host, a second service request for a second service session. The method may continue with selectively relaying the second server request to the server based on the service processing time.

In activating a service, a service gateway retrieves a service table entry using a service or server address of the service entry, where the service table entry has an association with another service entry. An association to the service entry is added and a marker value is set to indicate associations with two service entries. After a time duration, the association with the other service entry is removed, and the marker value is changed accordingly. In deactivating a service entry, the service gateway calculates a hash value for the service or server address of the service entry. After matching the hash value to a hash value of another service entry, an association with the other service entry is added. A marker value is set to indicate associations with two service entries. After a time duration, the association with the service entry is removed, and the marker value is changed accordingly.

Methods to manage services over a service gateway

A service gateway processes a service request received from a host based on a dynamic service response time of a server. In an exemplary embodiment, the service gateway relays a service request to a server over a service session between the service gateway and the server; receives a service response from the server; calculates a dynamic service processing time for the service request from a service request time and a service response time; compares the dynamic service processing time with an expected service processing time; updates a server busy indicator for the server in response to the comparing, where the server busy indicator is maintained at the service gateway; and processes future service requests in accordance with the server busy indicator at the service gateway.

A method, system, and computer program product for balancing servers based on server load status, include: receiving from a server a service response to a service request, the service response including a result from a processing of the service request and a server status indicating a computing load status of the server; obtaining the server status from the service response; receiving a next service request from a host, the next service request comprising a Uniform Resource Locator (URL); determining that the server is configured to process the URL; determining whether the server status indicates that the server is available to process the next service request; and in response to determining that the server status indicates that the server is available to process the next service request, sending the next service request to the server.

A service gateway processes a service request received from a host by: relaying the service request from the service gateway to a server over a service session between the service gateway and the server; determining a service request time for the service session; receiving by the service gateway a service response from the server; determining by the service gateway a service response time; calculating by the service gateway a service processing time for the service request from the service request time and the service response time; comparing the service processing time with an expected service processing time; and updating a server busy indicator for the server in response to the comparing. If the service processing time exceeds the expected service processing time, the server busy indicator is updated to indicate that the server is busy. Otherwise, the server busy indicator is updated to indicate that the server is not busy.

Exemplary embodiments for processing, by a service gateway, data packets sent over a communication session between a host and a server are disclosed. The service gateway processes data packets using a hybrid-stateful or hybrid-stateless processing method, depending on whether a hybrid-stateless or hybrid-stateful condition is satisfied. When the condition is satisfied, the service gateway allows for the changing from a hybrid-stateful to a hybrid-stateless processing method, or vice versa, for a subsequently received data packet. Otherwise, the service gateway continues processing the subsequently received data packets using the current hybrid processing method.

Growing community of inventors

San Jose, CA, United States of America

Feilong Xu