Abstract: A device for the transfer of heat away from a heat source comprising a base having first and second surfaces a plurality of fins extending adjacent to the second surface of the base. The base further including a chamber disposed between the first surface and the second surface of the base. The chamber further including a divider disposed therein adjacent the first surface. A pump is also disposed within the chamber to circulate fluid within the chamber.

Abstract: An electrically powered cooling unit (130), such as a fan unit, is mountable on a support structure (110) within a housing (140), for example a computer housing. The support structure (110) can mount one or more electrically powered cooling units (130). Each cooling unit (130) includes an electrical connector (400) fixed relative to the cooling unit and cooperable with a connector of the support structure, so that offering the cooling unit up to its mounting position completes the electrical connections between the cooling unit and the support structure. In alternative embodiments, a single structure is used both to physically support and electrically connect the cooling unit (130) to the support structure (110).

Abstract: Latent faults are detected by a testing system in a redundant DC power supply system having at least two power supplies connected to a common load via respective isolation devices. The testing system selects as a subject for a test procedure one isolation device and its corresponding power supply. The power supplies are controlled during the test procedure to marginally vary the output voltage of at least one power supply such that a differential voltage is applied between the output of the selected power supply and the outputs of the remaining power supplies for selectively changing the conductive state of at least one isolation device to a non conductive state by reducing its forward voltage to less than its respective forward bias voltage value.

Abstract: A processing unit, for example a computer server, that is connectable to a data communications network, has a device reader for reading a supplied network identity from a portable storage device such as a smart card or the like. The processing unit then uses the supplied network identity from the portable storage device for communicating via the data communications network. The processing unit monitors the continued presence of the portable storage device. In the event that the processing unit detects that the portable storage device has been removed from the device reader, it signals a fault state. The processing unit can be arranged to power itself down where a portable storage device having same network identity is not returned in the device reader within a predetermined time.

Abstract: A method and apparatus of improving prediction accuracy of a branch instruction scheme includes reading an individual instruction in a current set of instructions, fetching the individual instruction when an instruction fetch unit determines that the individual instruction is valid, and allowing the instruction fetch unit to use an index address for the fetched individual instruction. A method and apparatus of improving branch prediction accuracy includes receiving a set of instructions having an assigned address, making a prediction for a branch instruction in the set of instructions using the assigned address, and retaining the assigned address for the branch instruction in the set of instructions.

Abstract: A method for bonding diamond heat distribution structures to integrated circuit packages using optical contacting. In one embodiment, a heat spreader comprising diamond slab has a flat contact surface which is polished to a high degree of smoothness. An integrated circuit's package also has a flat contact surface which is polished to a high degree of smoothness. The contact surfaces of the diamond slab and the package are thoroughly cleaned and are then placed in contact with each other, establishing an optical contact bond between them. In one embodiment, the contact surfaces of the diamond and package which are to be bonded together are first polished, then a layer of an intermediate material such as silicon carbide is deposited on the polished surfaces. The silicon carbide layers on the contact surfaces are cleaned and placed in contact with each other to establish an optical contact bond.

Abstract: Provided is a method for performing loop-unrolling optimization during program execution. In one example, a method for loop optimization within a dynamic compiler system is disclosed. A computer program having a loop structure is executed, wherein the loop structure includes a loop exit test to be performed during each loop iteration. The loop structure is compiled during the execution of the computer program, and an unrolled loop structure is created during the compiling operation. The unrolled loop structure includes plurality of loop bodies based on the original loop structure. Further, the unrolled loop structure can include the loop exit test, which can be performed once for each iteration of the plurality of loop bodies.

Abstract: A graphics system and method for displaying lines on a display device. The system may comprise a sample buffer, a rendering unit and a sample-to-pixel calculation unit. The rendering unit may (a) generate a plurality of sample positions in a two-dimensional space, (b) determine a sample normal distance for each of the sample positions with respect to a line defined by the line-draw command, (c) assign sample values to the sample positions based on the sample normal distance of each of the sample positions, and (d) store the sample values in the sample buffer. The sample-to-pixel calculation unit may read sample values from the sample buffer, filter them to determine a pixel value, and transmit the pixel value to the display device. The rendering unit may render the line sample values with a narrower width to pre-compensate for the line-expanding effect of the filtering performed by the sample-to-pixel calculation unit.

Abstract: One embodiment of the present invention provides a system that uses a hot spot cache to alleviate the performance problems caused by hot spots in cache memories, wherein the hot spot cache stores lines that are evicted from hot spots in the cache. Upon receiving a memory operation at the cache, the system performs a lookup for the memory operation in both the cache and the hot spot cache in parallel. If the memory operation is a read operation that causes a miss in the cache and a hit in the hot spot cache, the system reads a data line for the read operation from the hot spot cache, writes the data line to the cache, performs the read operation on the data line in the cache, and then evicts the data line from the hot spot cache.

Abstract: A method and system for handling access to media content that is requested by a wireless device is provided. The method includes accessing, through a wireless device, a media provider over a wireless network. Then, the method includes selecting a first screen of the media provider. The first screen identifying a first storefront that defines a first type of media and associated selections of that first type of media. The method further includes downloading a first of the associated selections of that first type of media to a local buffer of the wireless device. The downloading is triggered automatically when the first screen is selected, and the first of the associated selections is displayed as available for sampling while a remainder of the associated selections continue to download to the local buffer. And, the remainder of the associated selections continue to download only while the first screen is selected.

Abstract: An integrated circuit package includes a lid with EMI containment features. The lid may include a plurality of projections adapted to couple a ground plane of a circuit board.

Abstract: An electromagnetic shielding structure is provided for a microprocessor or other electronic device that emits electromagnetic radiation. The structure includes a heat sink with an integrally formed depending skirt, and a conductive, compressible polymer is applied to a bottom surface of the skirt. The bottom surface mounts against a socket carried on a circuit board and is electrically coupled to a ground plane of the circuit board. The socket substantially surrounds the microprocessor in at least two dimensions (e.g. length and width). A shielding structure is formed at least partly by the heat sink, the socket and the ground plane.

Abstract: A system and method are disclosed to allow the tiling of sample jitter patterns to be independent of the tiling of clustered graphics accelerators. Each accelerator uses an x,y “bias” offset to shift the origin of the jitter pattern within the sample space region addressed by the accelerator. In this way, multiple accelerators may be programmed so that their jitter patterns integrate into one global pattern without discontinuities at the region boundaries.

Abstract: A server blade may comprise a processor and a network interface. The network interface can be configured be connected to a network remote the server blade. The server blade may be configured detect a property of an entity connected to the network. The server blade may be further configured as a field replaceable unit. Optionally, the server blade may further comprise an enclosure which encloses the processor.

Abstract: A method for utilizing postmortem object type identification to identify falsely shared memory objects is disclosed. In one method embodiment, the present invention accesses a postmortem object type identification containing type identifications for known nodes and inferred nodes. Next, an iteration for array nodes is performed on the postmortem object type identification. Then a list of the array nodes is generated, wherein the array nodes on the list are potential false sharing objects.

Abstract: A blockage aware zero skew clock routing method for calculating the distance, and therefore the delay, between two points takes into account any blockages along the path between the two points and therefore creates a more usable and realistic measure of delay and allows for minimization, or elimination, of clock skew in the system being designed using the method of the invention.

Abstract: A method for stabilizing an event queue during a graphical user interface test includes placing a test event on a tail of the event queue, generating a first empty event associated with the test event, placing the first empty event on the tail behind the test event, processing the test event off a head of the event queue, checking the event queue behind the first empty event for a later event, removing the first empty event if the later event is detected, processing the first empty event off the head of the event queue, and placing a second empty event associated with the test event on the tail of the event queue.

Abstract: One embodiment of the present invention provides a system that protects a server against denial-of-service attacks. During operation, the server receives a request for service from a client. Note that the client can be distinguished from other clients, for example, by its source IP address. In response to this request, the server sends a random number, y, and an identifier, id1, to the client, and allows the client to compute a preimage, x, such that y=h(x). Upon receiving an answer from the client including the preimage x and an identifier, id2, the server verifies that the identifier, id1, sent to the client matches the identifier, id2, received from the client. If the identifiers match, the server computes h(x), and compares h(x) against y. If h(x)=y, the server performs the requested service for the client. In this way, the server avoids computing h(x) until the server receives the answer with a matching identifier.

Abstract: A security architecture has been developed in which a single sign-on is provided for multiple information resources. Rather than specifying a single authentication scheme for all information resources, the security architecture associates trust-level requirements with information resources. Authentication schemes (e.g., those based on passwords, certificates, biometric techniques, smart cards, etc.) are employed depending on the trust-level requirement(s) of an information resource (or information resources) to be accessed. Once credentials have been obtained for an entity and the entity has been authenticated to a given trust level, access is granted, without the need for further credentials and authentication, to information resources for which the authenticated trust level is sufficient.

Abstract: The present invention provides a method and apparatus for controlling recursion at system startup. The method comprises creating one or more levels of device tree nodes branching from a root node, wherein the number of levels of device tree nodes is equal to a recursion depth. The apparatus comprises a storage unit adapted to store a recursion control process and a control unit adapted to execute the recursion control process, wherein the recursion control process causes the control unit to control recursion during system startup.