Silver Peak Systems, Inc.

Hewlett Packard Enterprise Development LP

Zhigang Yin

David Anthony Hughes

John Burns

Damon John Ennis

Theodore Ernest Tedijanto

Pawan Kumar Singh

Srivathsa Sarangapani

Manoj Gopal Kulkarni

Manoj Kumar Dutta

Patrick Lee Mahan

Bowen Yan

Kennedy Tjong Ki Ngo

Ganesh Lakshmanan

Ravikumar Govindarao

Provided are methods, apparatus, and system for policy based wide area network. A network of network appliances is configured with a policy configuration. Each network appliance is configured to validate each wide area network packet against the policy configuration. The validation can include verifying that the packets meet the SD-WAN network segment requirements and security rules including verifying that the source and destination address of the packet meet the firewall zone requirements. Each wide area network packet contains a policy header that is checked by the sending and receiving network appliance against the policy configuration.

Method device and system for policy based packet processing

A system for matching data using flow based packet data storage includes a communications interface and a processor. A communications interface receives a packet between a source and a destination. The processor identifies a flow between the source and the destination based on the packet. The processor determines whether some of packet data of the packet indicates a potential match to data in storage using hashes. The processor then stores the data from the most likely data match and second most likely data match without a packet header in a block of memory in the storage based on the flow.

Data matching using flow based packet data storage

Systems and methods for identification of data stored in memory are provided. A data packet is received and a first packet byte within the data packet is compared to a first memory byte within a memory. A mismatch is determined between the first packet byte and the first memory byte. A memory location is accessed that contains a second memory byte that is non-consecutive with the first memory byte. A packet location accessed that contains a second packet byte that is non-consecutive with the first packet byte. The second packet byte is compared to the second memory byte. A retrieval instruction is generated based at least in part on a result of the comparison between the second packet byte and the second memory byte. Various embodiments may be included in a network memory architecture to allow for faster data matching and instruction generation in a central appliance.

Identification of non-sequential data stored in memory

A system for processing packets includes a communications interface and a processor. A communications interface receives a packet between a source and a destination. The processor identifies a flow between the source and the destination based on the packet. The processor determines whether some of packet data of the packet matches to storage data in storage using hashes. The processor then stores the packet data in a block of memory in the storage based on the flow and if the packet data does not match the storage data.

Systems and methods for identification of data stored in memory are provided. A data packet is received and a first packet byte within the data packet is compared to a first memory byte within a memory. A mismatch is determined between the first packet byte and the first memory byte. A memory location is jumped to that contains a second memory byte that is non-consecutive with the first memory byte. A packet location is jumped to that contains a second packet byte that is non-consecutive with the first packet byte. The second packet byte is compared to the second memory byte. A retrieval instruction is generated based at least in part on a result of the comparison between the second packet byte and the second memory byte. Various embodiments may be included in a network memory architecture to allow for faster data matching and instruction generation in a central appliance.

Identification of data stored in memory

A method for deferred storage of data is disclosed. A packet comprising instructions and literal data for rebuilding packet data from a network device is received. The packet data is rebuilt according to the instructions and the literal data. The rebuilt packet data is rewritten to a temporary page in memory. A page utilization metric is calculated based on the contents of the temporary page, the contents comprising the rebuilt data packet. Whether the contents of the temporary page sufficiently match data in data storage is determined. The contents of the temporary page are discarded or stored based on the determination.

Deferred data storage

A system for managing network memory comprises a communication interface and a processor. The communication interface receives a status message from another appliance. The status message indicates an activity level of a faster memory and a slower memory associated with the other appliance. The communication interface also receives a data packet. The processor processes the status message to determine the activity level of the faster memory and the slower memory. The processor also processes the data packet to identify any matching data in the other appliance and estimate whether the matching data is stored in the faster memory based on the activity level. Based on the estimate, the processor determines whether to generate an instruction to retrieve the matching data.

Dynamic load management of network memory

A first appliance for ensuring data integrity in network memory is disclosed. The first appliance includes a processor and a communications interface that communicates over a communication network with a second appliance of the network memory. The processor determines whether reconciliation between a first peer state information in the first appliance and a second peer state information in the second appliance is needed. The first peer state information and the second peer state information track the status of data in the network memory. The processor determines an action to reconcile the first peer state information with the second peer state information based on a positive determination that the reconciliation is needed. The processor then performs the action to reconcile the first peer state information with the second peer state information.

Growing community of inventors

Mountain View, CA, United States of America

Zhigang Yin