International Business Machines Corporation

Lenovo (singapore) Pte. Ltd.

Globalfoundries Inc.

Avaya Inc.

Kyndryl, Inc.

Rockstar Consortium US LP

Keshav G Kamble

Vijoy A Pandey

Dar-Ren Leu

Casimer M DeCusatis

Dayavanti G Kamath

Mircea R Gusat

Sushma Anantharam

Nandakumar Peethambaram

Renato John Recio

Cyriel Johan Minkenberg

Abhijit Prabhakar Kumbhare

Bhalachandra G Banavalikar

Alexander Philip Campbell

Chandarani J Mendon

Nilanjan Mukherjee

Nirapada Ghosh

Jayakrishna Kidambi

Robert Birke

Daniel Crisan

Atul A Tambe

Claude Basso

Vinit Jain

David Iles

Stephan Benny

Uday Shankar Nagaraj

Deepak Kumar

Sisir Chowdhury

Simarbir Singh

Kenneth Michael Valk

Fredy Daniel Neeser

Nikolaos Chrysos

Harshad S Padhye

Dan-Alexandru Calavrezo

Mohanraj Krishnaraj

Chidambaram Bhagavathiperumal

Santosh Rajagopalan

Vaishali Pandya

James A Maples

Pramodh Mallipatna

Talha J Ilyas

Selvaraj Rajan

Amitabha Biswas

Roger Lapuh

Zenon Kuc

Vipin K Garg

Hesham Elbakoury

Immanuel Rahardja

Jagdish S Patel

Gregory Allen Foster

Sreenath Bandari

Ion A Iavorschi

Virgiliu Pop

Andreea-Simona Anghel

Sorin A Pistirica

Kumbhare P Abhijit

Dayavanti G Karnath

Nilanjan Mukharjee

Mihail-Liviu Manolachi

Andrei S Pistirica

Phil Campbell

In one embodiment, a method includes initiating cluster parameters that govern how a non-volatile memory (NVM) cluster functions and operates. Submission and completion queues of shared NVM on other nodes in the NVM cluster are mapped based on details of the shared NVM on the other nodes in the NVM cluster. The submission queue is configured to store commands to access the shared NVM according to a first-in-first-out (FIFO) scheme. The completion queue is configured to store completed commands after being processed through the submission queue. In another embodiment, a host-based data storage system includes NVM configured to store data. The host-based data storage system further includes a processor and logic integrated with and/or executable by the processor to perform the foregoing method.

Host based non-volatile memory clustering mechanism using network mapped storage

In one embodiment, a method includes creating on a per-tunnel basis, by a hardware processor, statistics about overlay-encapsulated packets which are received by or sent by the hardware processor across an overlay network, including counting events associated with de-encapsulation of one or more inner packets from an overlay-encapsulated packet; recording, by the hardware processor, the statistics in association with a virtual network identifier and a tunnel identifier that identifies a tunnel on which the overlay-encapsulated packets are received or sent; and maintaining, by the hardware processor, a table indexed for the virtual network identifier and the tunnel identifier, the table including the virtual network identifier; the tunnel identifier; and statistic bucket identifiers pointing to locations where statistics associated with the virtual network identifier and the tunnel identifier are stored.

Switch monitoring statistics gathering at servers and gateways for overlay networks

Reassembly of member cells into a packet comprises receiving an incoming member cell of a packet from a switching fabric wherein each member cell comprises a segment of the packet and a header, generating a reassembly key using selected information from the incoming member cell header wherein the selected information is the same for all member cells of the packet, checking a reassembly table in a content addressable memory to find an entry that includes a logic key matching the reassembly key, and using a content index in the found entry and a sequence number of the incoming member cell within the packet, to determine a location offset in a reassembly buffer area for storing the incoming member cell at said location offset in the reassembly buffer area for the packet for reassembly.

Segmentation and reassembly of network packets for switched fabric networks

In one embodiment, an apparatus includes a buffer memory, at least one ingress port, at least one egress port, at least one processor, and logic integrated with and/or executable by the at least one processor, the logic being configured to communicate with a software-defined network (SDN) controller, store one or more look-up tables in a first portion of the buffer memory, receive a packet using an ingress port, and determine an egress port for the packet. In another embodiment, a method for switching packets in a SDN includes storing one or more look-up tables in a first portion of a buffer memory of a SDN-capable switching device, receiving a packet using an ingress port of the switching device, and determining an egress port for the packet.

Dual purpose on-chip buffer memory for low latency switching

In one embodiment, a method includes receiving a plurality of flows, each flow comprising packets of data and assigning a service credit to each of the plurality of flows. In addition, the method includes assigning a weight parameter to each of the plurality of flows, and selecting a flow from a head of a first control queue unless the first control queue is empty or there is indication that the first control queue should be avoided. A flow is selected from a head of a second control queue in response to a determination that the first control queue is empty or there is indication that the first control queue should be avoided. Additionally, the method includes providing a number of units of service to the selected flow. Moreover, the method includes decreasing the selected flow's service credit by an amount corresponding to the number of units of service provided thereto.

Arbitration of multiple-thousands of flows for convergence enhanced ethernet

A method includes associating overlay network attributes (ONAs) with overlay virtual networks. The ONAs are managed as portable entities. A movement operation is performed on components of at least one overlay virtual network of the overlay virtual networks to servers based on management of the ONAs. An ONA is modified to identify attributes associated with the movement operation of the components of the at least one overlay virtual network.

Overlay network movement operations

A distributed device architecture includes a master device and one or more member devices. A simple network management protocol (SNMP) agent of a master device receives an SNMP request from a managing device. Where the SNMP request pertains to a given member device, and where the SNMP request requires involvement of the given member device to fulfill the SNMP request, the master device generates a non-SNMP request corresponding to the SNMP request and transmits the non-SNMP request to the given member device. A non-SNMP agent of the given member device processes the non-SNMP request and transmits processing results back to the master device. The master device generates an SNMP response corresponding to the processing results, and the SNMP agent of the master device transmits the SNMP response back to the managing device.

SNMP request processing within distributed device architecture

In one embodiment, a system includes a plurality of network ports including multiple Peripheral Component Interconnect express (PCIe) ports, a network interface card (NIC) driver configured to interface with and support an accelerated NIC, a processor, and logic integrated with the processor, executable by the processor, or integrated with and executable by the processor. The logic is configured to cause the processor to provide a virtual switch to host one or more virtual machines (VMs). Also, the logic is configured to cause the processor to provide a hypervisor which processes at least some outbound packets received from the one or more VMs and processes at least some inbound packets sent to the one or more VMs. Moreover, the logic is configured to cause the processor to divert network traffic of the one or more VMs that has overlay functionality provided by the accelerated NIC to bypass the hypervisor.

Processing of overlay networks using an accelerated network interface card

A method includes associating overlay network attributes (ONAs) with virtual networks. A movement operation is performed that includes re-assigning the virtual networks to servers of overlay networks based on management of the ONAs.

In one embodiment, a system includes a hardware processor and logic integrated with and/or executable by the hardware processor. The logic is configured to create statistics about overlay-encapsulated packets which are received by or sent by the hardware processor across an overlay network. The logic is also configured to record the statistics with a virtual network identifier and a tunnel identifier associated with at least one overlay-encapsulated packet for which the statistics are created. Moreover, the logic is configured to maintain a table indexed for the virtual network identifier and the tunnel identifier. The table includes the virtual network identifier, the tunnel identifier, and statistic bucket identifiers pointing to locations where statistics associated with the virtual network identifier and the tunnel identifier are stored. Other systems, methods, and computer program products are disclosed according to more embodiments.

In one embodiment, a system includes a hardware processor and logic implemented with and/or executable by the processor to determine that expiration of a timer has occurred, and in response to expiration of the timer, import entries from a media access control (MAC) address table stored by a switch controller into a MAC address table stored by a switch until the MAC address table stored by the switch is full, where the logic to import entries includes logic to import entries in the MAC address table stored by the switch controller determined to be more often used into the MAC address table stored by the switch before entries determined to be less often used.

Avoiding unknown unicast floods resulting from MAC address table overflows

Certain embodiments of the present disclosure provide a method for communicating data flows. The method generally includes receiving control flow information corresponding to a data flow for communication from a first device to a second device, and determining one or more characteristics of the data flow based on the control flow information. For example, the one or more characteristics may include whether the data flow is a mice flow or an elephant flow. In certain aspects, the method further includes communicating the data flow based on the determined characteristics.

Data flow configuration in hybrid system of silicon and micro-electro-mechanical-switch (MEMS) elements

In one embodiment, an apparatus includes a buffer memory, ingress ports, egress ports, at least one processor, and logic integrated with and/or executable by the at least one processor. The logic is configured to communicate with a software-defined network (SDN) controller, store a look-up table in a first portion of the buffer memory, receive a packet using an ingress port of the apparatus, start an egress timer upon receipt of the packet, process the packet in order to finish processing prior to the egress timer expiring, determine an egress port for the packet, determine a packet size from information in a header of the packet when packet size information is available in the header, begin to route the packet via the egress port once the egress port is determined, and send the packet to the egress port upon expiration of the egress timer without further processing.

Congestion management for data traffic in a virtual domain identifies a congestion source and sends a message to the source to adjust data traffic rates. The source may be a virtual machine hosted by a physical server with one or more virtual servers incorporated. A congestion manager may identify the source and send the message to the source without affecting other data sources hosted by the physical server or the virtual servers. In some embodiments, information about the congestion source may be encapsulated in a packet payload readable only by the congestion source so only the congestion source receives the instruction to adjust the transmission rate.

Virtual quantized congestion notification

Growing community of inventors

Fremont, CA, United States of America

Keshav G Kamble