Abstract: Systems and methods for supporting redundant independent networks in a high performance computing environment. A method can provide, at a computer comprising one or more microprocessors, one or more switches, one or more racks, each of the one or more racks comprising a set of the one or more switches, each set of the one or more switches comprising at least a leaf switch, a plurality of host channel adapters, at least one of the plurality of host channel adapters comprising a firmware and a processor, and a plurality of hosts. The method can provision two or more rails, the two or more rails providing redundant connectivity between the plurality of hosts. The method can isolate data traffic between the plurality of hosts to a rail of the two or more rails.

Abstract: Systems and methods are provided for supporting efficient reconfiguration of an interconnection network having a pre-existing routing. An exemplary method can provide a plurality of switches, a plurality of end nodes, and one or more subnet managers, including a master subnet manager. The method can calculate, via the master subnet manager, a first set of one or more leaf-switch to leaf-switch multipaths. The method can store this first set of one or more leaf-switch to leaf-switch multipaths at a metabase. The method can detect a reconfiguration triggering event, and call a new routing for the interconnection network. Finally, the method can reconfigure the network according to the new routing for the interconnection network.

Abstract: Systems and methods for dynamically assigning membership in a data partition to an end-port of a requesting host channel adapter. An exemplary embodiment can provide a subnet manager configured to operate within a subnet of a network fabric. The subnet can include a plurality of nodes, and the plurality of nodes can include at least one switch and a plurality of end-nodes, where the subnet manager executes on one of the plurality of nodes. A host channel adapter of a node in the subnet can request membership for an end-port of the host channel adapter. In response to the request, the subnet manager can request data from the data store to confirm that the end-port is a member of an admin partition and that the admin partition is associated with the data partition in which membership was requested.

Abstract: Systems and methods for supporting target groups for congestion control in a private fabric in a high performance computing environment. An exemplary method can provide, at one or more microprocessors, a first subnet, the first subnet comprising a plurality of switches, a plurality of host channel adapters, and a plurality of end nodes, including a plurality of virtual machines. The method can define a target group on one of an inter-switch link or at a port of a switch of the plurality of switches, wherein the target group defines a bandwidth limit on the at least one of an inter-switch link between two switches of the plurality of switches or at a port of a switch of the plurality of switches. The method can provide a target group repository stored in a memory of the host channel adapter where the defined target group in the target group repository is recorded.

Abstract: Systems and methods for providing bandwidth congestion control in a private fabric in a high performance computing environment. An exemplary method can provide, at one or more microprocessors, a first subnet, the first subnet comprising a plurality of switches, and a plurality of host channel adapters, wherein each of the host channel adapters comprise at least one host channel adapter port, and wherein the plurality of host channel adapters are interconnected via the plurality of switches, and a plurality of end nodes. The method can provide, at a host channel adapter, an end node ingress bandwidth quota associated with an end node attached to the host channel adapter. The method can receive, at the end node of the host channel adapter, ingress bandwidth, the ingress bandwidth exceeding the ingress bandwidth quota of the end node.

Abstract: Systems and methods for providing RDMA (remote direct memory access) read requests as a restricted feature in a high performance computing environment. An exemplary method can provide, at one or more microprocessors, a first subnet, the first subnet comprising a plurality of switches, a plurality of host channel adapters, wherein each of the host channel adapters comprise at least one host channel adapter port, and wherein the plurality of host channel adapters are interconnected via the plurality of switches, and a plurality of end nodes, including a plurality of virtual machine. The method can associate a host channel adapter with a selective RDMA restriction. The method can host a virtual machine of the plurality of virtual machines at the host channel adapter that comprises a selective RDMA restriction.

Abstract: 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.

Abstract: Systems and methods for defining virtual machine fabric profiles of virtual machines. An exemplary embodiment can provide a virtual machine identifier, a virtual host channel adapter instance ID, and a virtual globally unique identifier. The virtual machine identifier, virtual host channel adapter instance ID, and virtual globally unique identifier can be mapped to each other so that the virtual host channel adapter instance ID, and virtual globally unique identifier can be retrieved through access to the virtual machine identifier. Further, a relationship can be created between a P_Key that defines an admin partition and the virtual globally unique identifier, where the relationship between the P_Key and the virtual globally unique identifier defines the virtual globally unique identifier as a member of the admin partition defined by the P_Key.

Abstract: Systems and methods for supporting unique multicast forwarding across multiple connected subnets in a high performance computing environment. In accordance with an embodiment, by enforcing that incoming (i.e., incoming on a router port of a subnet) multicast packets have SGIDs (source global identifiers) that correspond to a restricted set of source subnet numbers when entering the ingress router ports to a local subnet, it is possible to ensure that multicast packets sent from one subnet are never returned to the same subnet through a different set of connected router ports (i.e., avoid looping multicast packets).

Abstract: 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.

Abstract: Systems and methods for using InfiniBand routing algorithms for Ethernet fabrics in a high performance computing environment. The method can provide, at a computer comprising one or more microprocessors, a plurality of switches, a plurality of hosts, a topology provider (TP) module, a routing engine (RE) module, and a switch initializer (SI) module. The method can perform a discovery sweep, by the TP, of the plurality of hosts and the plurality of switches and assigns an address to each of the plurality of hosts and the plurality of switches. The method can calculate, by the routing engine, a routing map, based upon a routing scheme, for the plurality of hosts and the plurality of switches, the routing map comprising a plurality of forwarding tables. The method can configure, each of the plurality of switches with a forwarding table of the plurality of forwarding tables calculated by the routing engine.

Abstract: Systems and methods of reserving a specific queue pair number for proprietary management traffic in a subnet of a high performance computing environment. An exemplary embodiment can provide a host channel adapter (HCA), including a control circuit with a processor and a memory, and at least one port operative to connect the HCA to a network fabric. Additionally, the HCA can include a subnet management agent (SMA) that executes on the HCA, and a configurable attribute stored in the memory of the HCA. A communication including a number value, can be received by the subnet management agent from a subnet manager of the subnet. The value of the configurable attribute can be set equal to the value of the received number. A QP number that corresponds to the set number value of the configurable attribute can then be reserved for sending and receiving proprietary network management traffic.

Abstract: 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.

Abstract: Systems and methods for supporting SMA level abstractions at router ports for enablement of data traffic in a high performance computing environment. In accordance with an embodiment, a subnet manager in a local subnet is responsible for enabling data traffic between subnets in a high performance computing environment. The SM can configure and set a data attribute at a switch port configured as a router port such that incoming data packets can be checked against the attribute to determine whether the data packet's destination is allowed or disallowed to receive inter-subnet data traffic.

Abstract: Systems and methods are provided for supporting efficient reconfiguration of an interconnection network having a pre-existing routing. An exemplary method can provide a plurality of switches, a plurality of end nodes, and one or more subnet managers, including a master subnet manager. The method can calculate, via the master subnet manager, a first set of one or more leaf-switch to leaf-switch multipaths. The method can store this first set of one or more leaf-switch to leaf-switch multipaths at a metabase. The method can detect a reconfiguration triggering event, and call a new routing for the interconnection network. Finally, the method can reconfigure the network according to the new routing for the interconnection network.

Abstract: Systems and methods for supporting SMP connectivity checks across virtual router in a high performance computing environment. In accordance with an embodiment, SMA model enhancements allow for the possibility to send a packet (i.e., SMP) that is addressed to a local router port. The SMA where the packet is addressed can receive the packet, and then apply a new attribute that defines that the requested information is on a remote node (e.g., connected by a physical link across subnets). In accordance with an embodiment, the SMA can operate as a proxy (receives a SMP and sends another request), or the SMA can modify the original packet and send it on as an inter-subnet packet.

Abstract: Systems and methods for path record handling in a fabric without host stack cooperation in a high performance computing environment. In a case where the subnet manager has determined “homogenous subnet/fabric” or “semi-homogenous subnet/fabric” status for the current topology, but is still receiving path queries, the subnet manager can use the relevant status to avoid any route evaluation and generate path record either only based on the configuration status of the requesting port in the homogenous case, or by comparing the configuration status of both ports in the semi-homogenous case.

Abstract: Systems and methods can provide for fabric level verification of host defined port GUIDs in a high performance computing network. A lightweight scheme for fabric level verification of host defined port GUIDs is provided wherein a virtual machine is assigned a set of GUIDs that travel with the virtual machine, even on migration or re-start.

Abstract: Systems and methods for supporting efficient virtualization in a lossless interconnection network. An exemplary method can provide, one or more switches, including at least a leaf switch, a plurality of host channel adapters, wherein each of the host channel adapters comprise at least one virtual function, at least one virtual switch, and at least one physical function, a plurality of hypervisors, and a plurality of virtual machines, wherein each of the plurality of virtual machines are associated with at least one virtual function. The method can arrange the plurality of host channel adapters with one or more of a virtual switch with prepopulated local identifiers (LIDs) architecture or a virtual switch with dynamic LID assignment architecture. The method can assign each virtual switch with a LID. The method can calculate one or more linear forwarding tables based at least upon the LIDs assigned to each of the virtual switches.

Abstract: 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.