Abstract: One embodiment provides a system that mitigates vibrations caused by cooling fans in a computer system. More specifically, the system includes a cooling fan mechanically coupled to the chassis of the computer system, wherein vibrations generated by the cooling fan are coupled to the chassis. The system also includes an actuation mechanism that creates a relative displacement between the cooling fan and the chassis when a control signal is applied to the actuation mechanism. The system additionally includes a detection mechanism which detects the relative displacement and generates a feedback signal which represents the relative displacement. The system further includes a control signal generation mechanism which converts the feedback signal into the control signal, which is subsequently applied to the actuation mechanism. When the control signal is applied to the actuation mechanism, the relative displacement between the cooling fan and the chassis vibrationally decouples the cooling fan from the chassis.

Abstract: One embodiment of the present invention provides a system that facilitates reducing the power needed for proximity communication. This system includes an integrated circuit with an array of transmission pads that transmit signals using proximity communication. This array is comprised of a set of macropads, where each given macropad is comprised of a set of micropads that can be configured to transmit a signal. A steering fabric routes signals to and within macropads, such that a subset of the micropads in the array can be configured to transmit the signal to a receiving component. Each macropad receives a limited number of input signals, with the steering fabric routing input signals to the micropads of the macropads. By limiting the number of input signals that are routed to the micropads of the macropads, the steering fabric eliminates redundant steering configurations for the array and reduces the power needed to transmit the signal.

Abstract: Systems and methods of testing integrated circuits are disclosed. The systems include a test module configured to operate between automated testing equipment and an integrated circuit to be tested. The testing interface is configured to test time sensitive parameters of the integrated circuit. The testing interface includes components for generating addresses, commands, and test data to be conveyed to the integrated circuit as well as a clock adjustment component. By adjusting the clock synchronization controlling the test signals to be conveyed to the integrated circuit, set-up time and hold time can be tested. The systems are configured to test set-up time and hold time of individual data channels, for example, an individual address line of the integrated circuit.

Abstract: A base mechanism for use in a multi-chip module (MCM) is described. This base mechanism includes a substrate having top and bottom surfaces. The bottom surface includes first electrical connectors that convey power, and through-substrate vias (TSVs) between the top and bottom surfaces are electrically coupled to these electrical connectors. Furthermore, a bridge chip is rigidly mechanically coupled to the top surface. This bridge chip includes proximity communication connectors that communicate information via proximity communication with one or more island chips in the MCM. Additionally, spacers are rigidly mechanically coupled to the top surface of the substrate. In conjunction with the bridge chip, the spacers define cavities on the top surface, which include second electrical connectors. These second electrical connectors are electrically coupled to the TSVs, and communicate additional information with and convey power to the one or more island chips.

Abstract: One embodiment of the present invention provides a system for facilitating re-transmitting packets in a communication network. During operation, the system transmits one or more acknowledgments to a neighbor in a designated time slot. Next, the system receives one or more acknowledgments from the neighbor, where a respective acknowledgment confirms that the data packets in a given past time slot have been processed by the neighbor. The system also receives a processed-through-counter (PTC) from the neighbor. Next, the system identifies a time slot up to the time slot indicated in the PTC and which has not been acknowledged by the neighbor. After identifying the time slot, the system re-transmits previously transmitted packets associated with the identified time slot to the neighbor without using a re-transmit timer.

Abstract: A method for providing control signals to a fan in a computer system is described. During the method, an electronic device receives temperature measurements and a fan-speed measurement performed in the computer system. Using a pattern-recognition model, the electronic device validates the measurements, and excludes any inaccurate measurements, such as those associated with drifting or failed sensors. Next, the electronic device determines control signals for a fan in the computer system using a model of coolant flow in the computer system and/or a slope of a phase-frequency curve of a cross power spectral density function corresponding to a pair of temperature profiles measured, as a function of time, by a pair of thermal sensors. Then, the determined control signals are provided to the fan.

Abstract: Methods and apparatus are provided for adaptively triggering garbage collection. During relatively steady or decreasing rates of allocation of free memory, a threshold for triggering garbage collection is dynamically and adaptively determined on the basis of memory drops (i.e., decreases in free memory) during garbage collection. If a significant increase in the rate of allocation of memory is observed (e.g., two consecutive measurements that exceed a mean rate plus two standard deviations), the threshold is modified based on a memory drop previously observed in conjunction with the current memory allocation rate, or a memory drop estimated to be possible for the current allocation rate.

Abstract: A system for facilitating voice calls over a content centric network (CCN) receives from a CCN user an expressed interest in receiving voice calls directed to the CCN user, wherein information in the CCN can be addressed, located, and disseminated by its content identifier and wherein data packets in the CCN are self-authenticating. Next, the system receives from a second network a packet for a voice call directed to the CCN user. The system then forwards to the CCN user the packet received from the second network with a CCN identifier corresponding to the interest expressed by the CCN user and forwards a packet from the CCN user for the voice call to the second network.

Abstract: One embodiment of the present invention provides a system that sizes a polygon in a layout. During operation, the system receives a polygon which is to be sized by a sizing amount. The system then selects one or more vertices of the polygon. Next, the system replaces each selected vertex with a set of replacement vertices, and subsequently assigns a projection path to each replacement vertex in the set of replacement vertices. The system next performs a sizing operation on the polygon according to the sizing amount. During the sizing operation, the system moves each replacement vertex in the set of replacement vertices along the assigned projection path, thereby creating a clipping on the angle associated with the selected vertex. Furthermore, this sizing operation is continuous: similar output polygons are obtained for similar sizing amounts.

Abstract: Embodiments of the present invention provide a system that allocates registers in a processor. The system starts by commencing a transaction, wherein commencing the transaction involves preserving a pre-transactional state of registers in a first register file. The system then allocates one or more registers for temporary use during the transaction. Upon finishing using each allocated register during the transaction, the system executes an instruction that restores the allocated register to the pre-transactional state.

Abstract: An encoder encodes data into a series of parallel codewords. Each codeword is expressed two sets of logic values (e.g., a set of logic 0s and a set of logic 1s) on output nodes. The encoder selects a current codeword from a group of codewords in a codespace which does not overlap the other group of codewords, i.e., codewords in a given group of codewords are not included in any other group of codewords in the codespace. This property allows a receiver of the codewords to be simplified. In particular, a mathematical operation performed on symbols in the current codeword uniquely specifies the corresponding group of codewords. This allows a decoder to decode the current codeword using comparisons of symbols received on a subset of all possible combinations of node pairs.

Abstract: One embodiment of the present invention provides a system that facilitates end-to-end quality of service (QoS) between a source and a destination in a network. During operation, the system allocates virtual channels on an output port of a switch to a number of quality of service (QoS) levels. The system further assigns a virtual channel to a traffic flow based on a QoS zone, wherein the QoS zone is identified by a host identifier, a target identifier, and a QoS level of the traffic flow. In addition, the system forwards data frames in the traffic flow via the assigned virtual channel.

Abstract: Some embodiments of the present invention provide a system that profiles a serial simulation of a circuit to estimate the performance of a parallel simulation of the circuit. During operation, the system profiles execution of module instances during a serial simulation of the circuit, wherein each module instance includes code which simulates signal propagation through a corresponding circuit module. Next, the system uses execution times for the module instances obtained from the serial simulation to estimate the performance of a parallel simulation of the circuit.

Abstract: An encoder encodes data into parallel codewords. Each codeword is expressed as a set of logic 0s and a set of logic 1s on two sets of output nodes. The encoder selects a current codeword which differs from the immediately preceding codeword by a fixed number of zero-to-one transitions on the first set of nodes and a fixed number of one-to-zero transitions on the second set of nodes. The current codeword is selected such that the first and second sets of nodes are different than additional nodes that contain transitions between the immediately preceding codeword and a bi-preceding codeword, and that logic values on additional nodes are unchanged between immediately preceding codeword and current codeword. A decoder decodes the codewords by comparing symbols on node pairs other than those for which transitions were expressed in the preceding code word, and decoding the results of those comparisons.

Abstract: One embodiment of the present invention provides a system that facilitates performing conditional compilation. The system operates by analyzing programming language source code prior to compilation to locate a preprocessing directive that specifies a condition to be evaluated, and also identifies source code to be conditionally included during compilation of the source code. Next, the system generates evaluation source code that evaluates the condition. The system then evaluates the condition by evaluating ordinary programming language and environmental variables within the evaluation source code. Note that the run-time variables are also available at run-time during execution of a compiled version of the original source code. Next, the system conditionally includes source code associated with a result of the evaluation and compiles the revised source code.

Abstract: One embodiment of the present invention provides a system that enables a computing device to save additional power by entering a “hibernation mode,” wherein the active state of the computing device is preserved in non-volatile storage while power to volatile storage is turned off. During operation, the system reanimates a computing device from a hibernation image by restoring reanimation code from the hibernation image and then executing the reanimation code. While executing this reanimation code, the system restores the rest of the hibernation image by, reading compressed data containing the rest of the hibernation image, and decompressing the compressed data using computational circuitry within the computing device. During this process, the decompression operations are overlapped with the reading operations to improve performance.

Abstract: One embodiment provides a system that receives a packet with a hierarchically structured variable-length identifier (HSVLI). An HSVLI indicates a piece or collection of content and may be hierarchically structured, comprising contiguous components ordered from a most general level to a most specific level. The length of a respective identifier is not fixed. During operation, the system converts the HSVLI into a fixed-length key. Subsequently, the system obtains forwarding information based on one or more longest-prefix matches with a longest-prefix-match lookup engine using the fixed-length key. Next, the system forwards the packet to an output port based on the forwarding information.

Abstract: One embodiment provides a system that verifies a user's identity. The system generates a list including a plurality of items and formulates a substantially large set of security questions base on the plurality of items. The number of questions in the set is significantly larger than a subset of security questions presented to the user to reduce the likelihood of the same questions being asked repeatedly. During account creation, the system presents to the user the subset of questions, and receives and stores a response from the user. At least one question in the subset is selected based on user information that is automatically extracted from devices associated with the user. Subsequently, the system receives a request to reset the user's password and presents the subset of questions to the requester. The system determines whether the requester is the user by comparing the requester's response with the stored user response.

Abstract: One embodiment of the present invention provides a system that implements a long-lived query (LLQ) at a name server. During operation, the system receives an LLQ from a client at the name server, wherein the LLQ requests information related to one or more data items stored on the name server. In response to the LLQ, the system informs the client of updates to the one or more data items.

Abstract: Some embodiments provide a system for accurately and efficiently modeling chemically amplified resist. During operation, the system can determine a quenched acid profile from an initial acid profile by applying multiple quenching models which are associated with different acid concentration ranges to the initial acid profile. One quenching model may be expressed as H=H0?B0, where H is an acid profile after quenching, H0 is an acid profile before quenching, and B0 is an initial base quencher profile. Another quenching model may be expressed as H=k·H0, where k is a constant. Next, the system can apply a smoothing kernel to the quenched acid profile to obtain a quenched-and-diffused acid profile. The smoothing kernel can generally be any weighted averaging function. The quenched-and-diffused acid profile can then be used to predict shapes that are expected to print on the wafer and to perform resolution enhancement techniques on a layout.