Abstract: A method is provided for a network-attached storage (NAS) server to directly write data to a disk or block device within a storage subsystem. A NAS server Ethernet interface receives a file, and writes the file data into kernel space as PDU segments. A TCP/IP stack maps the file data in kernel space RAM as sequentially ordered segments. The NAS/CIFS server application sends a call specifying file storage data. A zero-copy DMA application receives the call, maps a file offset to a Logical Block Address (LBA) in the block device, and requests that the block device DMA application transfer the file data. Without rewriting the file data in the system RAM, the block driver DMA application transfers the file data, in units of file system blocks, directly from kernel space RAM to the block device, with each file system block written in a single write operation.

Abstract: Various aspects provide for modifying a data stream for rate adaptation. A clock component receives a data stream at a first clock rate. In an aspect, a rate adaptation component inserts a first identification codeword into a particular location in the data stream based on a set of encoding rules in response to a determination that the first clock rate is lower than a second clock rate associated with a device configured for receiving a rate-adapted version of the data stream. In another aspect, the rate adaptation component removes a predefined codeword from the data stream and transforms another predefined codeword in the data stream into a second identification codeword in response to a determination that the first clock rate is greater than the second clock rate.

Abstract: One or more processing units are programmed to select from among M tones in a frequency domain representation of a signal, a set of tones including at least a strongest tone (relative to background noise) and a tone adjacent thereto. From among M complex numbers in the frequency domain representation of the signal, a set of complex numbers are identified and denoted as a vector Z, corresponding to the selected set of tones. Vector Z is then multiplied with each of M columns of a matrix G which is predetermined to identify a sub-resolution maxima in Z. The M products that result from the vector multiplication of Z and G are used to determine and store in memory at least one or both of: (A) a flag indicating presence or absence of narrowband interference in the signal; and (B) an estimate of a frequency of the narrowband interference.