Chelsio Communications Assigned Two Patents

Kernel bypass for ISCSI and NVME/TCP applications
Chelsio Communications, Inc., Sunnyvale, CA, has been assigned a patent (12524366) developed by M; Venkata Suman Kumar, and Prakash; Varun, Bengaluru, India, for a “kernel bypass for ISCSI and NVME/TCP applications.“

The abstract of the patent published by the U.S. Patent and Trademark Office states: “Techniques for host devices to offload iSCSI and NVMe/TCP data plane processing for data plane traffic to a NIC, and for the NIC to perform the data plane traffic processing in hardware. Traditionally, network protocol stacks have been implemented within the kernel of an operating system of a computing device. In light of this, iSCSI and NVMe/TCP user space applications running on host devices interact with a kernel of an operating system using system calls in order to send network traffic. However, the system calls, TCP/IP processing, and data copying required when communicating via the kernel increases CPU utilization as well as I/O latency. Techniques described herein include configuring the host device to enable kernel bypass for data path traffic for iSCSI and NVMe/TCP user space applications, and a NIC may include hardware configured to perform the iSCSI and NVMe/TCP processing for iSCSI and NVMe/TCP connections.”

The patent application was filed on 2024-09-30 (18/902741).

Programmable processing of network protocol packets
Chelsio Communications, Inc., Sunnyvale, CA, has been assigned a patent (12495005) developed by Hanel; John, and Krishna; Balekudru, Sunnyvale, CA, for a “programmable processing of network protocol packets.“

The abstract of the patent published by the U.S. Patent and Trademark Office states: “System and methods for processing state-based network packets are described herein. Processing of network protocol packets can include classifying events associated with the network protocol packet based on a current, internal protocol state and event type. The techniques may map these events to actions. For instance, the techniques may map these incoming events to specific indices within a TCAM. The indices may be used to lookup “actions” encoded within an SRAM. The system may utilize these actions to generate one or more outgoing messages as well as determine possible “next” state transitions. Accordingly, by utilizing contents of a TCAM and SRAM, the techniques may facilitate dynamic and flexible programmability of processing network protocol packets.”

The patent application was filed on 2024-03-22 (18/614311).