Abstract: A system includes a physical NIC associated with the computer and including receive rings for storing network traffic addressed to the VNIC, where the receive rings include a local receive ring for storing local traffic and a forwarding receive ring for storing forwarding traffic. The physical NIC further includes a hardware classifier configured to register a local Internet Protocol (IP) address associated with the VNIC in a lookup table, receive a packet including a destination IP address, classify the packet as local traffic or forwarding traffic using the lookup table, and store the packet in the local receive ring or the forwarding receive ring based on the type. The VNIC is configured to retrieve the packet from the receive ring, and process the packet in a type-specific manner based on the receive ring where the packet was stored, where the type-specific manner is different for local traffic and forwarding traffic.

Abstract: A computer readable medium including instructions executable by a processor to perform a method, the method including obtaining a packet by a load balancer, obtaining queued packet information for a plurality of target hosts operatively connected to the load balancer, selecting the one of the plurality of target hosts using the queued packet information, and sending the packet to the selected target host using a first communication channel between the load balancer and the selected target host.

Abstract: Methods and systems thereof for controlling access to resources are described. When a user attempts to access a resource via a remote interface such as a Web server, the request is initially evaluated by a source of policy definitions such as a policy server. This source returns a policy decision to the remote interface. The policy decision is stored in memory by the remote interface. The remote interface can then evaluate subsequent requests from the user for the resource using the stored policy decision instead of having to communicate again with the source for the policy decision. Enhancements to this approach are also described. Accordingly, policy definitions and decisions are more efficiently implemented.

Abstract: Incoming network data is processed according to a current hardware classification “engine” configuration. As data is propagated from a network interface to a host system, an activity of one or more components of the host system is monitored. If it is determined that a desired/optimal resource utilization of the host system and/or a desired/optimal network performance is not being achieved, the hardware classification “engine” configuration is dynamically modified.

Abstract: Adaptive modifications of spinning and blocking behavior in spin-then-block mutual exclusion include limiting spinning time to no more than the duration of a context switch. Also, the frequency of spinning versus blocking is limited to a desired amount based on the success rate of recent spin attempts. As an alternative, spinning is bypassed if spinning is unlikely to be successful because the owner is not progressing toward releasing the shared resource, as might occur if the owner is blocked or spinning itself. In another aspect, the duration of spinning is generally limited, but longer spinning is permitted if no other threads are ready to utilize the processor. In another aspect, if the owner of a shared resource is ready to be executed, a thread attempting to acquire ownership performs a “directed yield” of the remainder of its processing quantum to the other thread, and execution of the acquiring thread is suspended.

Abstract: One embodiment of the present invention provides a store queue that applies the stores to a memory subsystem in program order. This store queue includes a content-addressable memory (CAM), which holds pending stores and facilitates looking up stores based on addresses for the stores, wherein the CAM does not keep track of program order between stores to different addresses. The store queue also includes a program-order queue which keeps track of program order between the stores in the CAM and thereby facilitates applying the stores to the memory subsystem in program order. In a variation on this embodiment, the CAM is a priority CAM which holds separate copies of multiple stores with identical addresses, and when a lookup based on an address matches multiple stores, returns the youngest matching store.

Abstract: One embodiment of the present invention provides a system that facilitates precise inter-chip alignment for proximity communication and power delivery. The system includes a first integrated circuit chip, whose surface has etch pit wells. The system also includes a second integrated circuit chip, whose surface has corresponding etch pit wells configured to align with the etch pit wells of the first integrated circuit chip. A shaped structure is placed in an etch pit well of the first integrated circuit chip such that when the corresponding etch pit well of the second integrated circuit chip is substantially aligned with the etch pit well of the first integrated circuit chip, the shaped structure mates with both the etch pit well of the first integrated circuit chip and with the corresponding etch pit well of the second integrated circuit chip, thereby precisely aligning the first integrated circuit chip with the second integrated circuit chip.

Abstract: Rack mountable display apparatus is provided for an electronic equipment rack. The apparatus includes a support for the display whereby the display is moveable between a storage position within the rack and a deployed position external to the rack. The display can be in the form of sheets of material supported on a mount, or could be in the form of an electronic display in combination with a processor. The processor could be located within the rack, or could be deployed with the display, for example where the display is part of a portable computer. The display is used to provide maintenance information to a maintenance engineer to facilitate maintenance of the rack mounted equipment.

Abstract: Data buffering allocation in a microprocessor complex for a request of memory allocation is supported through a remote buffer batch allocation protocol. The separation of control and data placement allows simultaneous maximization of microprocessor complex load sharing, and minimization of inter-processor signaling/metadata migration. Separating processing control from data placement allows the location of data buffering to be chosen so as to maximize bus bandwidth utilization and achieve non-blocking switch behavior. This separation reduces the need for inter-processor communication and associated interrupts thus improving computation efficiency and performance.

Abstract: A test system including a package with switchable paths. The package may have conductive paths that are selected by switches. The electrically switchable conductive paths may yield increased data without significantly increasing the required testing hardware.

Abstract: A test system including a package with interconnect paths. The package may have electrical paths that are electrically connected by the interconnect paths. The electrically connected electrical paths may yield increased data without significantly increasing the required testing hardware.

Abstract: Parallelize a computer program by scoping program variables at compile time and inserting code into the program. Identify as value predictable variables, variables that are: defined only once in a loop of the program; not defined in any inner loop of the loop; and used in the loop. Optionally also: identify a code block in the program that contains a variable assignment, and then traverse a path backwards from the block through a control flow graph of the program. Name in a set all blocks along the path until a loop header block. For each block in the set, determine program blocks that logically succeed the block and are not in the first set. Identify all paths between the block and the determined blocks as failure paths, and insert code into the failure paths. When executed at run time of the program, the inserted code fails the corresponding path.

Abstract: A method of translating data including obtaining a value of an implementation data structure from an instrumented program, accessing a translator associated with the instrumented program, wherein the translator comprises a plurality of transformations and translating the value of the implementation data structure using the translator to obtain translated data, wherein the translating comprises applying the plurality of transformations to convert a representation of the implementation data structure into an interface data structure.

Abstract: A computing system is configured to receive incoming packets at a first priority designated for use when receiving packets that are not associated with specific subscription levels. Incoming packets are examined to determine whether they are associated with a particular subscription level. Responsive to a determination that a given packet received over a particular connection is associated with a particular subscription level, the computing system is reconfigured to process packets subsequently received over the same connection at a different priority associated with the particular subscription level.

Abstract: A method, apparatus and computer program product for providing a device includes incorporating a session ID into the device name. Communication between an application and device is done by way of a pseudo device driver which identifies the appropriate daemon based on the session id, and forwards the data to the daemon where the daemon provides the data to the device. In operations, the system detects a requirement to access a device during a user session and obtains a session identification associated with the user session. A daemon creates a single device node having an identity that indicates the session identification associated with the user session and a pseudo device driver receives a request to access a device from an application associated with the user session.

Abstract: A method for serializer maintenance and coalescing in a distributed object store (DOS) including a first partition and a second partition, involving requesting an update of an object, wherein the object includes an active globally unique identifier (AGUID) object and at least one version globally unique identifier (VGUID) object, wherein the least one VGUID object includes a first generation number and a first serializer name, determining whether a first serializer is located in the first partition using the first serializer name, wherein the first serializer is associated with the first generation number, if the first serializer is not located in the first partition, constructing a second serializer using the first serializer name, assigning a second generation number to the second serializer, obtaining an order of the update to the object using the second serializer, and creating a new VGUID object.

Abstract: A device adapted to optimize air duct flow in an electronic system including a printed circuit board configured to support one or more electronic components having various heights. The device includes an air duct including a rigid cross-sectional member, the air duct being adapted to facilitate air flow from an inlet port to an outlet port. Further, the rigid cross-sectional member includes an opening having an interior perimeter, the interior perimeter of the opening being attached to a flexible sheet by a pressure sensitive adhesive. The rigid cross-sectional member attached to the flexible sheet is adapted to compress one or more electronic components to the printed circuit board when the rigid cross-sectional member is mounted on the printed circuit board, and is configured to form an airtight enclosure to optimize air duct flow.

Abstract: Concurrently branch predicting for multiple branch-type instructions demands of high performance environments. Concurrently branch predicting for multiple branch-type instructions provides the instruction flow for a high bandwidth pipeline utilized in advanced performance environments. Branch predictions are concurrently generated for multiple branch-type instructions. The concurrently generated branch predictions are then supplied for further processing of the corresponding branch-type instructions.

Abstract: A fan module, including a housing for receiving a fan, a connector configured to mate with a connector housing, and a vibration pad configured to isolate vibrations of the fan from transferring through the fan module, wherein the fan module is configured to receive an alignment attachment standoff.

Abstract: A network interface card in a networked client computer includes a network interface circuit that decodes and then compares incoming network packet addresses to known address bit patterns, the decoding and comparing circuitry being powered at all times. Receipt and recognition of certain addresses means the client computer must be powered-on, even if manually switched OFF. When such a server-transmitted address is recognized, a power-on signal is issued to a power control unit that causes full operating power to be coupled to the client computer. In this fashion, a server can broadcast power-on signals to a plurality of networked client computers or workstations.