Oracle International Corporation

Oracle America, Inc.

Sun Microsystems, Inc.

Graphcore Limited

Other

Bjorn Dag Johnsen

Line Holen

Bartosz Bogdanski

Dag Georg Moxnes

Ernst Gunnar Gran

Arvind Srinivasan

Feroz Zahid

Bartosz Bogdański

Ola Tørudbakken

Harald Høeg

Evangelos Tasoulas

David Brean

Shimon Muller

Martin Paul Mayhead

Rabin Sugumar

Lars Paul Huse

Ajoy Siddabathuni

Predrag Hodoba

Brian Manula

Richard P Mousseau

Ola Torudbakken

Vadim Makhervaks

Wei Lin Guay

Prabhunandan Narasimhamurthy

Avneesh Pant

Ankita Bhandary

Christopher J Jackson

Daniel John Pelham Wilkinson

Andreas V Bechtolsheim

Sumanta Kumar Chatterjee

Kant C Patel

Feroz Alam Khan

Bhaskar Mathur

Hans Rygh

Sudeep Vatsanath Reguna

Amarnath Jolad

Tor Skeie

Kurt Tjemsland

Morten Schanke

Yatin Gajjar

Jean De Lavarene

Raghunath Shenbagam

Daniel Delfatti

Inge Lars Birkeli

Pramod Venkatesh

Roy Arntsen

William M Ortega

Ben Sum

Julia D Harper

Steinar Forsmo

Martin G Vickers

Srinivas Madhur

Robert W Wittosch

Knut Tvete

Stephen McKinty

Prabhunandan B Narasimhamurthy

Ola Torudbaken

Sven-Arne Reinemo

Hans Ludwig Opheim

Hans Ludvig Opheim

The same test data frame is dispatched from a network interface device a plurality of times so as to test a network. Since the same test data frame is used, it may be unnecessary for a new test data frame to be provided and protocol processed each time one is required to be sent. The protocol processing resources of the network interface device are then available for sending further traffic in parallel with the dispatch of the test data frames. On the receive side, the network interface device collects statistics regarding the reliable receipt of test frames, without requiring the test frames to be further processed and provided to a driver of the network interface device. In this way, the processing and buffering capacity in the network interface device is available for handling further traffic in parallel with the test traffic.

Network path testing via independent test traffic

One or more bits of the destination MAC address indicate a number of times a reset event has occurred. These bits may be referred to as a generation number. The generation number in a destination MAC address is updated when a reset event occurs. In this way, frames issued by the sender prior to the reset may be distinguished from frames issued after the reset, since the destination MAC addresses in those frames will not match. In this way, the recipient device is protected from stale packets.

Generation number for handling resets

Aspects of the present disclosure include implementing fabric availability and synchronization (FAS) agents within a fabric network. In one example, a first FAS agent executing on a first network device may receive, from a second network device, a command to modify a configuration of a second network device. The first FAS may upgrade the configuration of the first network device based on the command from a current configuration to a new configuration. The first FAS agent increment a state identifier associated with the configuration of the first network device to a new state identifier associated with the new configuration. The first FAS agent may then transmit a control packet that includes the new state identifier. A second FAS agent executing on the second network device may receive the control packet and execute the command to update the configuration of the second network device to the new configuration.

Fabric availability and synchronization

Each of the nodes stores a number, referred to herein as a generation number, which is updated whenever the respective node undergoes a reset and restart from checkpoint. Since the nodes of the system participate in the same reset event, at most times, each generation number held by a node will be the same across the system. However, in some cases, when one node resets before another node, the generation numbers between those two nodes will differ. The data frames sent between the nodes each comprise a generation number of the sending node, which is checked by the recipient and only accepted if the generation number in the frames matches the generation number of the recipient node.

Reset of a multi-node system

Each of the processing devices stores an event vector, which is updated when certain events (e.g. memory errors, overtemperature events) occur on the device. Different elements of the vector correspond to different types of events. When an event of a given type occurs on one device, the update to the event vector on that device is propagated to other devices in the system. Those other devices, in response, update the corresponding element in their own event vector to indicate that an event of that given type has occurred in the system. In this way, events are aggregated between the different devices using the event vector. The event vector is considered to be a global event vector, since its elements indicate whether certain events have occurred across the entire system, and the vector is consistent across the system.

Global event aggregation

System and method for correlating fabric-level group membership with subnet-level partition membership in a high performance computing environment. An exemplary embodiment can provide an admin partition in a subnet of a network environment, associated with a resource domain at the fabric level of the network environment. The admin partition can be defined by a P_Key that is stored in a fabric-level database. Each fabric resource is identified by an identifier that is stored in the fabric-level database. When fabric resources are added as members to the resource domain, a relationship is created between the P_Key and the identifier of the fabric resource, where the relationship defines the resource as a member of the admin partition. Access rights can be granted based on the relationship.

System and method for correlating fabric-level group membership with subnet-level partition membership in a high-performance computing environment

Systems and methods for using multiple CE (congestion experienced) flags in both FECN (forward explicit congestion notification) and BECN (backward explicit congestion notification) in a high performance computing environment. An exemplary method can provide a first subnet comprising a plurality of switches, a plurality of host channel adapters, and a plurality of end nodes. The method can receive, at an end node attached to a host channel adapter, an ingress packet from a remote end node, wherein the ingress packet traversed at least a portion of the first subnet prior to being received at the end node. The method can, on receiving the ingress packet, send a response message from the end node attached to the host channel adapter to the remote end node, the response message indicating that the ingress packet experienced congestion during the traversal of the at least a portion of the first subnet.

System and method for supporting use of forward and backward congestion notifications in a private fabric in a high performance computing environment

Systems and methods for on-demand unicast forwarding in a high performance computing environment, in accordance with an embodiment. An exemplary method can provide, at a computer comprising one or more microprocessors, a plurality of switches, a plurality of host channel adapters, and a plurality of hosts. The method can provide, at two rails, redundant connectivity between the plurality of hosts, wherein each rail comprises a set of the plurality of switches. The method can detect a disruption i connectivity between two hosts of the plurality of hosts. The method can re-establish communication between the two hosts of the plurality of hosts via unicast packets on an inter-rail link, the inter-rail link being provided between the two rails.

System and method for on-demand unicast forwarding in a high performance computing environment

Methods and systems for supporting efficient load balancing among a plurality of switches and a plurality of end nodes arranged in a tree topology in a network environment. The methods and systems can sort the plurality of end nodes, wherein the plurality of end nodes are sorted in a decreasing order of a receive weight. The method and system may further route, in the decreasing order of receive weights, the plurality of end nodes, wherein the routing comprises selecting at least one down-going port and at least one up-going port. Further, the method and system can increase an accumulated downward weight on each selected down-going port by the receive weight of the routed end node, and increase an accumulated upward weight on each selected up-going port by the receive weight of the routed end node.

System and method for supporting efficient load-balancing in a high performance computing (HPC) environment

Systems and methods for providing explicit multicast local identifier assignment for per-partition default multicast local identifiers defined as subnet manager policy input in a high performance computing environment. In accordance with an embodiment, an explicit multicast local identifier (MLID) assignment policy can be provided (as, e.g., administrative input) that explicitly defines which MLIDs will be used for which partitions in a subnet. Further, an MLID assignment policy can also define which dedicated MLIDs will be associated with given multicast group identifiers (for example, partition independent MLIDs). By employing such an MLID assignment policy, a new or restarted master subnet manger can observe and verify the MLIDs used for existing partitions, instead of generating new MGID to MLID mappings. In this way, changes in MLID associations for any corresponding MGID can be avoided as a result of master SM restarts or failovers, or any subnet-merge operations.

System and method to provide explicit multicast local identifier assignment for per-partition default multicast local identifiers defined as subnet manager policy input in a high performance computing environment

Systems and methods for supporting inter subnet partitions in a high performance computing environment. In accordance with an embodiment, a fabric manager can define a range of P_Key values, among a plurality of P_Key values, as a inter subnet partition (ISP) P_Key range. The fabric manager can communicate this defined range of P_Key values to a number of subnets, via their subnet managers. The subnet managers in each subnet retain management over their subnets. As there is no central management that configures each side of inter subnet communication, subnet managers on within participating subnets can set up ISP membership, and then exchange information with the other subnet.

System and method for supporting inter subnet partitions in a high performance computing environment

System and method for supporting node role attributes in a high performance computing environment. In accordance with an embodiment, a node role attribute can comprise a vendor defined subnet management attribute. When a subnet manager attempts to discover a high performance computing environment, such as an InfiniBand subnet, or a switch topology, identifying a topology is quite complex when subnet manager can only observe connectivity, without context behind the connectivity (the roles of the different nodes in the connectivity). However, when a subnet has a node role attribute enabled, the subnet manager can map the interconnect more effectively as it can discover not only the connectivity during the initial sweep, but it can also discover the role of each node discovered, thus leading to a more efficient interconnect discovery.

System and method for supporting node role attributes in a high performance computing environment

Systems and methods for supporting a single logical IP subnet across multiple independent layer 2 subnets in a high performance computing environment. A method can provide, at a computer including one or more microprocessors, a logical device, the logical device being addressed by a layer 3 address, wherein the logical device comprises a plurality of network adapters, each of the network adapters comprising a physical port, and a plurality of switches. The method can arrange the plurality of switches into a plurality of discrete layer 2 subnets. The method can provide a mapping table at the logical device.

System and method for a single logical IP subnet across multiple independent layer 2 (L2) subnets in a high performance computing environment

A system and method for supporting network isolation in a multi-tenant cluster environment. An exemplary method can support one or more tenants, and can associate each of the one or more tenants with a partition of a plurality of partitions, and can also associate each of the plurality of partitions with one or more nodes of a plurality of nodes, each of the plurality of nodes being associated with a leaf switch of a plurality of switches. The method can mark each of the plurality of partitions with a policy parameter. The method can assign each node of the plurality of nodes a partitioning order based on the marked on the partition associated with each node. Finally, the method can, based at least upon the marking of the partition of the plurality of partitions, generate one or more linear forwarding tables for use in the multi-tenant cluster environment.

System and method for efficient network isolation and load balancing in a multi-tenant cluster environment

Growing community of inventors

Oslo, Norway

Bjorn Dag Johnsen