Abstract: Systems, methods, and computer-readable media for operating a dedicated virtual machine host cluster within cloud computing infrastructure are described herein. In one embodiment, an instance principal certificate that includes a virtual network identifier for a dedicated virtual network for the cluster is retrieved by a host computing device. The instance principal certificate is authenticated by the host against a switch to grant access of the host to a virtual network indicated by the virtual network identifier through a network interface card of the host. A virtual function of the network interface card of the host is allocated to a guest virtual machine. The guest virtual machine is launched in the host with the virtual function as a network device of the guest virtual machine.

Abstract: Systems, methods, and computer-readable media for operating a dedicated virtual machine host cluster within cloud computing infrastructure are described herein. In one embodiment, an instance principal certificate that includes a virtual network identifier for a dedicated virtual network for the cluster is retrieved by a host computing device. The instance principal certificate is authenticated by the host against a switch to grant access of the host to a virtual network indicated by the virtual network identifier through a network interface card of the host. A virtual function of the network interface card of the host is allocated to a guest virtual machine. The guest virtual machine is launched in the host with the virtual function as a network device of the guest virtual machine.

Abstract: Systems, methods, and other embodiments associated with cloud computing cluster isolation with strong authentication and automatic configuration deployment are described. In one embodiment, a method includes, in response to receiving an authentication frame through the uncontrolled port, (i) extracting data from the authentication frame for performing authentication, and (ii) parsing the authentication frame to identify a piggybacked virtual network identifier that functions as an instruction to create an isolated connection. If the authentication was successful and the piggybacked virtual network identifier was identified, creating the isolated connection between the controlled port and a virtual network that is identified by the piggybacked virtual network identifier.

Abstract: Systems, methods, and other embodiments associated with cloud computing cluster isolation with strong authentication and automatic configuration deployment are described. In one embodiment, a method includes, in response to receiving an authentication frame through the uncontrolled port, (i) extracting data from the authentication frame for performing authentication, and (ii) parsing the authentication frame to identify a piggybacked virtual network identifier that functions as an instruction to create an isolated connection. If the authentication was successful and the piggybacked virtual network identifier was identified, creating the isolated connection between the controlled port and a virtual network that is identified by the piggybacked virtual network identifier.

Abstract: Some embodiments facilitate high performance packet-processing by enabling one or more processors that perform packet-processing to determine whether to enter an idle state or similar state. As network packets usually arrive or are transmitted in batches, the processors of some embodiments determine that more packets may be coming down a multi-stage pipeline upon receiving a first packet for processing. As a result, the processors may stay awake for a duration of time in anticipation of an incoming packet. Some embodiments keep track of the last packet that entered the first stage of the pipeline and compare that with a packet that the processor just processed in a pipeline stage to determine whether there may be more packets coming that need processing. In some embodiments, a processor may also look at a queue length of a queue associated with an upstream stage to determine whether more packets may be coming.

Abstract: Some embodiments facilitate high performance packet-processing by enabling one or more processors that perform packet-processing to determine whether to enter an idle state or similar state. As network packets usually arrive or are transmitted in batches, the processors of some embodiments determine that more packets may be coming down a multi-stage pipeline upon receiving a first packet for processing. As a result, the processors may stay awake for a duration of time in anticipation of an incoming packet. Some embodiments keep track of the last packet that entered the first stage of the pipeline and compare that with a packet that the processor just processed in a pipeline stage to determine whether there may be more packets coming that need processing. In some embodiments, a processor may also look at a queue length of a queue associated with an upstream stage to determine whether more packets may be coming.

Abstract: Some embodiments facilitate high performance packet-processing by enabling one or more processors that perform packet-processing to determine whether to enter an idle state or similar state. As network packets usually arrive or are transmitted in batches, the processors of some embodiments determine that more packets may be coming down a multi-stage pipeline upon receiving a first packet for processing. As a result, the processors may stay awake for a duration of time in anticipation of an incoming packet. Some embodiments keep track of the last packet that entered the first stage of the pipeline and compare that with a packet that the processor just processed in a pipeline stage to determine whether there may be more packets coming that need processing. In some embodiments, a processor may also look at a queue length of a queue associated with an upstream stage to determine whether more packets may be coming.

Abstract: Some embodiments facilitate high performance packet-processing by enabling one or more processors that perform packet-processing to determine whether to enter an idle state or similar state. As network packets usually arrive or are transmitted in batches, the processors of some embodiments determine that more packets may be coming down a multi-stage pipeline upon receiving a first packet for processing. As a result, the processors may stay awake for a duration of time in anticipation of an incoming packet. Some embodiments keep track of the last packet that entered the first stage of the pipeline and compare that with a packet that the processor just processed in a pipeline stage to determine whether there may be more packets coming that need processing. In some embodiments, a processor may also look at a queue length of a queue associated with an upstream stage to determine whether more packets may be coming.

Abstract: A method and an apparatus for color mapping of flow power data in which the flow states containing information of most interest to the user are enhanced, while flow states not containing information of interest are suppressed. This is accomplished using a color mapping having three segments which are connected at upper and lower knee points, the latter being settable either automatically or via operator inputs. In the automatic mode, the host computer determines the positions of the knee points based on analysis of a single or a mean image frame of power Doppler data or based on system parameters (e.g., dynamic range) set by the operator. The host computer then constructs a color mapping having the determined knee points and loads that color mapping into the video processor.

Abstract: An ultrasound imaging system having an adaptive spatial filter the filter coefficients of which, for particular image parameter sample, are determined by counting the number of neighboring image parameter samples having zero or near-zero values. If the number of zero or near-zero values in a data window is greater than a predetermined threshold, the data in the window is passed, not filtered. This filter has two advantages over other spatial filters. First, image parameter data samples having only zero or near-zero neighboring values (i.e., isolated "point noise") are not smeared. Second, boundaries such as the edge of color in a vessel (where the surrounding area is black, i.e., the color image parameter values are zero or near zero) are not smoothed as much as in conventional filters, preserving the sharpness of the edge.

Abstract: An analytical method for automatic tracing of a Doppler time-velocity waveform envelope. At the core of this algorithm is a maximum frequency detection mechanism which is based on searching for the highest frequency bin whose spectral amplitude exceeds a certain noise threshold. The method is based on a theoretical noise amplitude distribution in the video spectral domain. In particular, the method uses a precise model of the statistical distribution of the video spectral power of white noise to establish a threshold for maximum frequency detection. Input to the noise model is the average white noise level in the video spectral display, which can be computed using either of two analytical methods.

Abstract: A method and an apparatus for increasing the spatial resolution and sensitivity of a color flow image while maintaining a desired acoustic frame rate. The ultrasound energy is concentrated at a more narrowly defined focal region, which allows for increased flow sensitivity and vessel filling. Better flow uniformity across the color region of interest is also achieved. The method uses multiple transmit focal zones, and transmit and receive apertures having low f-numbers. Using multiple focal zones with low f-numbers allows for tight focusing over a larger depth-of-field. Unique waveforms and unique gain curves are used for each focal zone. Each focal zone is fired on a separate acoustic frame. An adaptive frame averaging algorithm is used to blend together the in-focus data from each of these acoustic frames before the data is displayed.

Abstract: A method and an apparatus for mitigating aliasing when imaging moving fluid or tissue using velocity Doppler shift data. To eliminate the effects of slight aliasing in the velocity mode, symmetrical (or non-directional) velocity/color maps are used to map positive and negative velocity data of the same magnitude to the same color and same display intensity. If the pulse repetition frequency is adjusted so that only subtle aliasing is present, an optimal frame averaging of the data can be achieved by removing the sign of the velocity data either before or during frame averaging.

Abstract: A method and an apparatus for three-dimensional imaging of ultrasound data by constructing projections of data from a volume of interest. An ultrasound scanner collects B-mode or color flow images in a cine memory, i.e., for a multiplicity of slices. A multi-row transducer array having a uniform elevation beamwidth is used to provide reduced slice thickness. The data from a respective region of interest for each of a multiplicity of stored slices is sent to a master controller, such data forming a volume of interest. The master controller performs an algorithm that projects the data in the volume of interest onto a plurality of rotated image planes using a ray-casting technique. The data for each projection is stored in a separate frame in the cine memory. These reconstructed frames are then displayed selectively by the system operator. Segmentation of three-dimensional projection images is enhanced by decreasing the thickness and increasing the resolution (i.e.

Abstract: A method and an apparatus for frame averaging ultrasound imaging data. A one-tap IIR filter is used to average the corresponding pixel data of two frames. The frame averaging is a function of a normalized difference between the pixel data of the two frames. This is achieved by taking the absolute difference between the signal levels of the current frame and the previous frame and dividing the result by the arithmetic (or geometric) mean of the two data. A multitude of look-up tables of output values are generated off-line. Each look-up table is designed to be used under a specific set of operating parameters. In response to the user's selection of these operating parameters, the system downloads a selected frame-averaging look-up table from system memory. During subsequent system operation, the downloaded look-up table outputs values which are a function of the normalized difference between the previous and current frame data used to address the look-up table.

Abstract: A method for optimizing the beam distribution to obtain the highest frame rate together with the least amount of acoustic noise while limiting the spatial aliasing to an acceptable level. The beam distribution is adjusted according to the scan format and as a function of the beam width, which is proportional to the product of the F number and the operating wavelength. As the F number changes in going from the shallow depth to a deeper depth, the number of beams fired to construct an image would change correspondingly. In addition, if the F number is changing in going from the center of image to the edge of the image, the separation of the fired beams would change. Alternatively, if the operating frequency is being changed in going from shallower depths to deeper depths or in going from the center of image to the edge of the image, the beam spacing is changed in order to fire the optimum number of beams at all times, and thereby obtain the highest frame rate.