摘要:
A method for generating and implementing a three-dimensional (3D) computer processing chip stack plan that includes receiving system requirements from a plurality of clients, identifying common processing structures and technologies from the system requirements, and assigning the common processing structures and technologies to a layer in the 3D computer processing chip stack plan. The method also includes identifying uncommon processing structures and technologies from the system requirements and assigning the uncommon processing structures and technologies to a host layer in the 3D computer processing chip stack plan. The method further includes determining placement and wiring of the uncommon structures on the host layer, storing placement information in the plan, and transmitting the plan to manufacturing equipment. The manufacturing equipment generates and integrates both the layer including the common structures and technologies and the host layer including the uncommon structures and technologies to form the 3D computer processing chip stack.
摘要:
A semiconductor chip includes a plurality of multi-core clusters each including a plurality of cores and a cluster controller unit. Each cluster controller unit is configured to control thread assignment within the multi-core cluster to which it belongs. The cluster controller unit monitors various parameters measured in the plurality of cores within the multi-core cluster to estimate the computational demand of each thread that runs in the cores. The cluster controller unit may reassign the threads within the multi-core cluster based on the estimated computational demand of the threads and transmit a signal to an upper-level software manager that controls the thread assignment across the semiconductor chip. When an acceptable solution to thread assignment cannot be achieved by shuffling of threads within the multi-core cluster, the cluster controller unit may also report inability to solve thread assignment to the upper-level software manager to request a system level solution.
摘要:
Block placement within each device-containing layer is optimized under the constraint of a simultaneous optimization of interlayer connectivity between the device-containing layer and immediately adjacent device-containing layers. For each functional block within the device-containing layer, lateral heat flow is calculated to laterally adjacent functional blocks. If the lateral heat flow is less than a threshold value for a pair of adjacent functional blocks, placement of the functional blocks and/or interlayer interconnect structure array therebetween or modification of the interlayer interconnect structure array is performed. This routine is repeated for all adjacent pairs of functional blocks in each of the device-containing layers. Subsequently, block placement within each device-containing layer may be optimized under the constraint of a simultaneous optimization of interlayer connectivity across all device-containing layers. This method provides a design having sufficient lateral heat flow in each of the device-containing layers in a semiconductor chip.
摘要:
A method for branch prediction, the method comprising, receiving a branch wrong guess instruction having a branch wrong guess instruction address and data including an opcode and a branch target address, determining whether the branch wrong guess instruction was predicted by a branch prediction mechanism, sending the branch wrong guess instruction to an execution unit responsive to determining that the branch wrong guess instruction was predicted by the branch prediction mechanism, and receiving and decoding instructions at the branch target address.
摘要:
An optically connectable circuit board and optical components mounted thereon. At least one component includes optical transceivers and provides an optical connection to the board. Electronic components may be directly connected to the board electrically or optically. Also, some electronic components may be indirectly connected optically to the board through intermediate optical components.
摘要:
A hardware description language (HDL) design structure embodied on a machine-readable data storage medium includes elements that when processed in a computer aided design system generates a machine executable representation of a device for implementing dynamic refresh protocols for DRAM based cache. The HDL design structure further includes a DRAM cache partitioned into a refreshable portion and a non-refreshable portion; and a cache controller configured to assign incoming individual cache lines to one of the refreshable portion and the non-refreshable portion of the cache based on a usage history of the cache lines; wherein cache lines corresponding to data having a usage history below a defined frequency are assigned by the controller to the refreshable portion of the cache, and cache lines corresponding to data having a usage history at or above the defined frequency are assigned to the non-refreshable portion of the cache.
摘要:
A memory cache comprising, a data sector having a sector ID, wherein the data sector stores a data entry, a primary directory having a primary directory entry, wherein a position of the primary directory entry is defined by a congruence class value and a way value, and a secondary directory corresponding to the data sector having a secondary directory entry corresponding to the data sector, wherein the secondary directory entry include, a primary ID field corresponding to the way value, and a sector ID field operative to identify the sector ID.
摘要:
A high speed optical channel including an optical driver and a photodetector in a CMOS photoreceiver. The optical channel driver includes a FET driver circuit driving a passive element (e.g., an integrated loop inductor) and a vertical cavity surface emitting laser (VCSEL) diode. The VCSEL diode is biased by a bias supply. The integrated loop inductor may be integrated in CMOS technology and on the same IC chip as either/both of the FET driver and the VCSEL diode. The photodetector is in a semiconductor (silicon) layer that may be on an insulator layer, i.e., SOI. One or more ultrathin metal electrodes (
摘要:
A computer chip is structured to have at least one single-layered chip, at least one multi-layered chip stack, and a carrier package characterized by electrical interconnections of less than 100 microns diameter, wherein the single-layered chip and the multi-layered chip stack are each electrically coupled to the electrical interconnections of the carrier package, and the single-layered chip is communicatively coupled to the multi-layered chip stack through the carrier package so that an electrical signal propagates over a given distance between the single-layered chip and the multi-layered chip stack at substantially a speed of propagation for a single layer chip over the given distance. The single-layered chip can be a processor having multi-cores and the multi-layered chip stack can be a memory cache stack. Interconnect vias, having a density at least as great as 2500 interconnects/cm2 electrically couple the single-layered chip and the multi-layered chip stack to the carrier package.
摘要翻译:计算机芯片被构造成具有至少一个单层芯片,至少一个多层芯片堆叠和以小于100微米直径的电互连为特征的载体封装,其中单层芯片和多层芯片 芯片堆叠都电耦合到载体封装的电互连,并且单层芯片通过载体封装通信地耦合到多层芯片堆叠,使得电信号在单层之间传递给定距离 芯片和多层芯片堆栈,在给定距离内的单层芯片基本上是传播速度。 单层芯片可以是具有多核的处理器,并且多层芯片堆栈可以是存储器高速缓存堆栈。 具有至少高达2500个互连/ cm 2的密度的互连通孔将单层芯片和多层芯片堆叠电耦合到载体封装。
摘要:
An electronic system with components communicating over optical channels, a board initialization and continuity check and a method of transferring data over the optical channels. The system include a backplane with board to board signal wiring and a shared optical bus. Optical gratings are attached to the backplane and to circuit boards to pass optical energy between an optical transceiver and board/backplane. An optical transceiver at each end of each optical jumper relays optical signals between the optical jumpers and the connected circuit board or the backplane. Optical jumpers optically connect the circuit boards to the backplane.