Abstract: An apparatus and method for implementing multicast in system area network channel adapters are provided. With the apparatus and method, a multicast packet is received in a channel adapter of an end node. The channel adapter determines which local queue pairs are party of the multicast group identified by a destination local identifier in the multicast data packet. Based on this determination, the channel adapter replicates the data packet and delivers a copy of the data packet to each local queue pair that is part of the multicast group.

Abstract: A mechanism for managing multicast groups with send-without-receive (SWR) joiners without the use of traps on creation and deletion of groups is provided. Group information is maintained continuously while the SWR member exists. When an SWR join is attempted and the group does not already exist, the group information (MLID) is marked as used and the first switch to which the SWR packets are sent is routed to discard all packets sent to the group. When receiving members join the group, the routing is updated so that the SWR member begins sending to the receiving members. When the last receiving member leaves the group, the first switch is again routed to discard the packets.

Abstract: A method for managing the protocol of read/write messages in a PCI Express communication link is disclosed. The method comprises maintaining queues of write requests and read requests associated with each of a plurality of request identifications that are contained in a message header, wherein the read requests associated with a request identification are held in abeyance until such time that write requests associated with the request identification are completed.

Abstract: A method for managing the protocol of read/write messages in a PCI Express communication link is disclosed. The method comprises maintaining queues of write requests and read requests associated with each of a plurality of request identifications that are contained in a message header, wherein the read requests associated with a request identification are held in abeyance until such time that write requests associated with the request identification are completed.

Abstract: Systems and arrangements to adjust resource accessibility based upon usage modes are contemplated. Embodiments may include a supervisor instance such as an operating system that is adapted to select one or more supervisor keys from a set of supervisor keys available to the host node or platform upon which the supervisor instance is operating. The supervisor instance may select the supervisor key(s) based upon an association of the supervisor instance with a usage mode represented by a supervisor key name. In many embodiments, supervisor key names may be associated with resources via with one or more of the supervisor keys based upon the resources needed when operating in the usage modes by a system administrator. Once the supervisor instance is initialized, the supervisor instance may issue client keys that allow access to subsets of the resources available to the supervisor instance based upon usage modes of the clients.

Abstract: A method, system, and product in a data processing system are disclosed for providing centralized management of an INFINIBAND distributed system-area network that includes multiple end nodes. A manager application is established in one of the end nodes. An agent application is established in one or more end nodes. Each agent application is independent from the manager application. The manager application maintains a current list of active agent applications and uses the list to manage the agent applications in the end nodes.

Abstract: A method in a network computing system for managing configuration information for a set of components in a network computing system. The configuration information is stored for the set of components in the network computing system to form stored configuration information. In the depicted examples, the components may be nodes or devices within nodes. Current configuration is obtained in response to a power cycle. The current configuration information is compared with the stored configuration information to form a comparison. The stored configuration information is updated if a difference is present in the comparison. The stored configuration information is used to configure the components when a power cycle occurs.

Abstract: A method for managing multicast groups with send-without-receive (SWR) joiners without the use of traps on creation and deletion of groups is provided. Group information is maintained continuously while the SWR member exists. When an SWR join is attempted and the group does not already exist, the group information (MLID) is marked as used and the first switch to which the SWR packets are sent is routed to discard all packets sent to the group. When receiving members join the group, the routing is updated so that the SWR member begins sending to the receiving members. When the last receiving member leaves the group, the first switch is again routed to discard the packets.

Abstract: A mechanism for managing multicast groups with send-without-receive (SWR) joiners without the use of traps on creation and deletion of groups is provided. Group information is maintained continuously while the SWR member exists. When an SWR join is attempted and the group does not already exist, the group information (MLID) is marked as used and the first switch to which the SWR packets are sent is routed to discard all packets sent to the group. When receiving members join the group, the routing is updated so that the SWR member begins sending to the receiving members. When the last receiving member leaves the group, the first switch is again routed to discard the packets.

Abstract: A method, system, and product in a data processing system are disclosed for providing centralized management of an INFINIBAND distributed system-area network that includes multiple end nodes. A manager application is established in one of the end nodes. An agent application is established in one or more end nodes. Each agent application is independent from the manager application. The manager application maintains a current list of active agent applications and uses the list to manage the agent applications in the end nodes.

Abstract: A method, system, and product in a data processing system are disclosed for providing centralized management of an INFINIBAND distributed system-area network that includes multiple end nodes. A manager application is established in one of the end nodes. An agent application is established in one or more end nodes. Each agent application is independent from the manager application. The manager application maintains a current list of active agent applications and uses the list to manage the agent applications in the end nodes.

Abstract: A method and system for efficiently merging subnets having individual subnet managers (SM) into a single network with one master SM. During discovery and configuration of a subnet, a subnet manager creates a Subnet Management Database (SMDB) representative of the subnet components being managed. Each subnet manager contains an independent SMDB. When two or more subnets are linked/connected together to form a single network, a single one of the subnet managers is selected as the master subnet manager and all the subnets' SMDBs must be merged. An SMDB record labeling mechanism is utilized to differentiate among components from the different subnets that may have the same parameter values, such as protection keys (P_keys) and re-assign individual parameters to create a single SMDB of uniquely defined components.

Abstract: Systems and arrangements to adjust resource accessibility based upon usage modes are contemplated. Embodiments may include a supervisor instance such as an operating system that is adapted to select one or more supervisor keys from a set of supervisor keys available to the host node or platform upon which the supervisor instance is operating. The supervisor instance may select the supervisor key(s) based upon an association of the supervisor instance with a usage mode represented by a supervisor key name. In many embodiments, supervisor key names may be associated with resources via with one or more of the supervisor keys based upon the resources needed when operating in the usage modes by a system administrator. Once the supervisor instance is initialized, the supervisor instance may issue client keys that allow access to subsets of the resources available to the supervisor instance based upon usage modes of the clients.

Abstract: A system and method for establishing multiple connections using a private data field of a communication management protocol is provided. With the present invention, a Service ID identifies a specific consumer and the private data field contains a list of connection attributes for each connection that is to be established. An active side requests a connection and the passive side replies to the connection request. The active side sends the passive side a connection establishment request. This connection establishment request includes a Service ID which identifies a passive side process associated with a service. This connection establishment request also includes communication attributes of one or more connected services and datagram services associated with the Service ID. The passive passes the connection request to a process associated with the service. If the passive side process does not wish to carry out the service, a negative reply message is returned to the active side.

Abstract: A method and system are provided for dynamically informing an operating system (OS) of a distributed computer system, when a (new) device is added on the network. An OS registers with the subnet administrator (SA) to be notified when a new component is added to the subnet and placed in a partition to which the OS has access. The subnet manager (SM) initializes and configures the new component that is added and, through the partition manager, associates the component to one or more partitions of the subnet. The SA then notifies the OSs with access to the partition(s) to which the component is associated and that have a right to access the new component that the new component is available to the OS. Following, the OS updates a database of available components to include the new component and establishes communication with the new component.

Abstract: A method for determining parameters needed to communicate with a remote node in a computer network is provided. The invention comprises determining the location of the remote node to which an InfiniBand (IB) node might desire to communicate. This resolution comprises determining the location of the remote node based on a desired application or service, and then determining the IB parameters needed to communicate with the remote node. The resolution might also involve determining the specific queue pairs that associated with the service dynamically. The general solution is applied to IP service resolution.

Abstract: A method in a node for managing authorized attempts to access the node. A packet is received from a source, wherein the packet includes a first key. A determination is made as to whether the first key matches a second key for the node. The packet is dropped without a response to the source if the first key does not match the second key. Information from the packet is stored in response to this absence of a match. The information is sent to a selected recipient in response to a selected event, which may be, for example, either immediately or in response to polling to see if the information is present.

Abstract: A method for routing System Area Network (SAN) packets to multiple partitions within a single end node is provided. A range of Local Identification addresses (LIDs) are assigned to a channel adapter port within the SAN. Lower order bits within the LID are then assigned to select the particular partition in the end node. The Local ID Mask Control (LMC) field is used to assign multiple LIDs to a single port, using those low order bits to then route the message to the appropriate partition in the end node.

Abstract: Apparatus and method for virtualizing a queue pair space to minimize time-wait impacts. Virtual queue pairs are allocated from a virtual queue pair pool of a node to connections between the node and other nodes. The connection is established between a physical queue pair of the node and physical queue pairs of other nodes. From the viewpoint of the other nodes, they are communicating with the present node using the virtual queue pair and not the physical queue pair for the present node. By using the virtual queue pairs, the same physical queue pair may accommodate multiple connections with other nodes simultaneously. Moreover, when a connection is torn down, the virtual queue pair is placed in a time-wait state rather than the physical queue pair. As a result, the physical queue pair may continue to function while the virtual queue pair is in the time-wait state.

Abstract: An apparatus and method for managing reliable datagram work queues, and associated completion queues, using head and tail pointers with end-to-end context error cache are provided. Reliable datagram (RD) queue head and tail pointers are maintained in the channel interface and the host channel adapter. The head and tail pointers in the host channel adapter include a RD queue page table index and a RD queue page index for identifying a position within the RD queue. For RD work queues, in the channel interface, the tail pointer is used to identify a next position where a work queue entry may be written and the head pointer is used only to determine whether the work queue is full. In the host channel adapter, the head pointer is used to identify a next work queue entry for processing and the tail pointer is used to determine if the queue is empty.