摘要:
A calculating apparatus, or system, having a plurality of stages, such as in a pipeline arrangement, has the clocking rail or conductor positioned alongside the stages. With a large number, i.e., hundreds, of stages arranged in parallel sub-arrays, the clocking conductor is snaked alongside the sub-arrays. In individual stages it is arranged that the shortest of the two calculations taking place in a stage, takes place in the return path. An array can be divided into separate sections for independent processing.
摘要:
A plurality of entities are stored in a single addressable location in a Content Addressable Memory (CAM). A column in a CAM entry is selected for storing an entity based on the property of the entity to distribute the entities among the columns to maximize memory utilization. A match for a search key stored in one of the plurality of columns can be found in a single search operation.
摘要:
A Content Addressable Memory array includes a plurality of coupled sub-blocks arranged in rows and columns. Search data received by a first sub-block in a first column of the CAM is propagated to each sub-block in the row to the last sub-block in the last column of the CAM. The search result for the CAM is selected based on the propagated search results for each row of sub-blocks and output on a side of the array opposite from the side receiving the search data.
摘要:
A lookup unit matrix combines a plurality of lookup units to provide a longest prefix match for a search key longer than the lookup unit's mapper key. A portion of the search key is provided to each of the plurality of lookup units in a single search request issued to the lookup unit matrix. Each lookup unit in the lookup unit matrix performs a multi-level search for the result value based on the portion of the search key forwarded as the mapper key and the result of a multilevel search in the previous lookup unit. The search results in a value corresponding to the search key stored in a single location in one of the lookup units.
摘要:
A load balancing mechanism maps a binary tree representation of a routing table into a set of fixed size memories. The mechanism efficiently utilizes the memory in the routing table without violating the tree precedence constraints and the memory access requirements of a pipelined system. The mechanism stores a subtree associated with a densely populated level of the binary tree in memory associated with lower levels.
摘要:
We present a method and apparatus for performing adaptive data compression. An alphabet and vocabulary in the encoder and decoder is built adaptively and stored in a dictionary as symbols are to be encoded and decoded. Each time an unknown symbol is to be encoded by the encoder, the encoder adds the symbol to the dictionary and transmits it in plain in the encoded string. The code words transmitted by the encoder include symbols and indexes. The state of a prefix bit preceding the code word indicates whether the code word is a plain symbol or an index of a symbol or string of symbols stored in the dictionary. The decoder examines the prefix bit of each code word as it is received to determine if the code word stores a symbol in plain or in index. If the code word stores a symbol in plain, the decoder learns the symbol by adding a sequence of symbols resulting from the concatenation of previously decoded symbols and the first symbol of the currently decoded symbol and by adding the symbol to its dictionary. If the code word stores an index, the decoder decodes the code word by extracting the symbol or sequence of symbols stored in the dictionary at the respective index in the dictionary.
摘要:
A number of hash tables are accessed concurrently with a different computed index based on a single search key for each hash table. Each index identifies a location in one of the hash tables capable of storing at least one entry. If all indexed locations are used, the entries stored in the lookup table can be reordered so that the new entry can be inserted in one of the locations identified by the computed indexes.
摘要:
A method and apparatus for minimizing memory required for storing non-binary width data structures is disclosed. The non-binary width data structure is segmented into plural segments. The segments are stored in a plurality of memory blocks. Mapper logic maps a logical address to a physical address in the memory blocks to access non-binary width entries in the non-binary width data structure stored in the memory blocks.
摘要:
An initialization circuit in a delay locked loop ensures that after power up or other reset clock edges are received by a phase detector in the appropriate order for proper operation. After reset of the delay locked loop, the initialization circuit assures that at least one edge of a reference clock is received prior to enabling the phase detector to increase (or decrease) the delay in a delay line. After at least one edge of a feedback clock is received, the initialization circuit enables the phase detector to decrease (or increase) the delay in a delay line.
摘要:
A circuit and method for accelerating bus line communication in an integrated circuit is disclosed. High speed transmission of signals along a bus line is achieved by driving a series of bus line segments with their own bi-directional bus amplification circuits. Because each bus line segment has less capacitive loading than longer non-segmented bus lines, voltage reversal, or data inversion of a pair of complementary lines of a bus line segment is accomplished at high speed. Each bi-directional bus amplification circuit includes a precharge circuit for precharging each complementary pair of lines to known logic levels, and a drive circuit for changing the logic level of each line. The bi-directional bus amplification circuit of the present invention logically connects two lines to each other while actively amplifying the signal in either direction without prior knowledge of which direction the signal must be driven, and without additional control overhead than that required for a conventional precharged bus line.