Juniper Networks, Inc.

Abbas Y Sakarwala

Sriram Raghunathan

Surinder Singh

Yong Luo

Wai Leung Chan

Harmeet Sahni

Jian Cui

Gopi Krishna

Pankaj Gupta

Prabhakaran Ganesan

Thilak Raj Surendra Babu

Adithya Hemakumar

The disclosed method may include (1) identifying a data object that (A) was generated by a producer process running on a computing device and (B) is to be accessed by a consumer process running on the computing device, (2) determining at least one structural characteristic of the data object that indicates how data is stored within the data object, (3) storing the data object and the structural characteristic of the data object within a portion of shared memory on the computing device that is accessible to both the producer process and the consumer process, and then (4) disclosing, to the consumer process, locations of both the data object and the structural characteristic of the data object within the shared memory such that the consumer process is able to extract the data in the data object from the shared memory. Various other methods, systems, and apparatuses are also disclosed.

Apparatus, system, and method for efficiently sharing data between processes

A multi-router system is described in which hardware and software components of one or more standalone routers can be partitioned into multiple logical routers. The multiple logical routers are isolated from each other in terms of routing and forwarding functions yet allow network interfaces to be shared between the logical routers. Moreover, different logical routers can share network interfaces without impacting the ability of any of the logical routers to be independently scaled to meet the bandwidth demands of the customers serviced by the logical router.

Multi-router system having shared network interfaces

In one example, a serving gateway device includes one or more network interfaces configured to receive a packet fragment from a packet data network gateway (PGW) device, and a control unit configured to hash a source Internet protocol (IP) address, a destination IP address, and a fragment identifier value for the packet fragment to determine an entry of a hash table, wherein the entry of the hash table includes data defining a next expected offset, a next expected fragment identifier, and a reference to a fragment table comprising data for at least one previous packet fragment corresponding to the packet fragment, to compare a length of the packet fragment to the next expected offset and the fragment identifier value for the packet fragment to the next expected fragment identifier, and store the packet fragment using the fragment table based on the comparison.

In-line packet reassembly within a mobile gateway

Growing community of inventors

San Jose, CA, United States of America

Abbas Y Sakarwala