摘要:
Methods and apparatus implementing Hardware/Software co-optimization to improve performance and energy for inter-VM communication for NFVs and other producer-consumer workloads. The apparatus include multi-core processors with multi-level cache hierarchies including and L1 and L2 cache for each core and a shared last-level cache (LLC). One or more machine-level instructions are provided for proactively demoting cachelines from lower cache levels to higher cache levels, including demoting cachelines from L1/L2 caches to an LLC. Techniques are also provided for implementing hardware/software co-optimization in multi-socket NUMA architecture system, wherein cachelines may be selectively demoted and pushed to an LLC in a remote socket. In addition, techniques are disclosure for implementing early snooping in multi-socket systems to reduce latency when accessing cachelines on remote sockets.
摘要:
A processor includes a cache, a prefetcher module to select information according to a prefetcher algorithm, and a prefetcher algorithm selection module. The prefetcher algorithm selection module includes logic to select a candidate prefetcher algorithm determine and store memory addresses of predicted memory accesses of the candidate prefetcher algorithm when performed by the prefetcher module, determine cache lines accessed during memory operations, and evaluate whether the determined cache lines match the stored memory addresses. The prefetcher algorithm selection module further includes logic to adjust an accuracy ratio of the candidate prefetcher algorithm, compare the accuracy ratio with a threshold accuracy ratio, and determine whether to apply the first candidate prefetcher algorithm to the prefetcher module.
摘要:
A technology for implementing a method for distributed memory operations. A method of the disclosure includes obtaining distributed channel information for an algorithm to be executed by a plurality of spatially distributed processing elements. For each distributed channel in the distributed channel information, the method further associates one or more of the plurality of spatially distributed processing elements with the distributed channel based on the algorithm.
摘要:
A thread scheduler may be used in a chip multiprocessor or symmetric multiprocessor system to schedule threads to processors. The scheduler may determine the bandwidth utilization of the two threads in combination and whether that utilization exceeds the threshold value. If so, the threads may be scheduled on different processor clusters that do not have the same paths between the common memory and the processors. If not, then the threads may be allocated on the same processor cluster that shares cache among processors.
摘要:
An apparatus and method is described herein for intelligently spilling cache lines. Usefulness of cache lines previously spilled from a source cache is learned, such that later evictions of useful cache lines from a source cache are intelligently selected for spill. Furthermore, another learning mechanism—cache spill prediction—may be implemented separately or in conjunction with usefulness prediction. The cache spill prediction is capable of learning the effectiveness of remote caches at holding spilled cache lines for the source cache. As a result, cache lines are capable of being intelligently selected for spill and intelligently distributed among remote caches based on the effectiveness of each remote cache in holding spilled cache lines for the source cache.
摘要:
An apparatus and method is described herein for intelligently spilling cache lines. Usefulness of cache lines previously spilled from a source cache is learned, such that later evictions of useful cache lines from a source cache are intelligently selected for spill. Furthermore, another learning mechanism—cache spill prediction—may be implemented separately or in conjunction with usefulness prediction. The cache spill prediction is capable of learning the effectiveness of remote caches at holding spilled cache lines for the source cache. As a result, cache lines are capable of being intelligently selected for spill and intelligently distributed among remote caches based on the effectiveness of each remote cache in holding spilled cache lines for the source cache.
摘要:
A thread scheduler may be used in a chip multiprocessor or symmetric multiprocessor system to schedule threads to processors. The scheduler may determine the bandwidth utilization of the two threads in combination and whether that utilization exceeds the threshold value. If so, the threads may be scheduled on different processor clusters that do not have the same paths between the common memory and the processors. If not, then the threads may be allocated on the same processor cluster that shares cache among processors.
摘要:
Systems and methods for managing shared cache by multi-core processor. An example processing system comprises: a plurality of processing cores, each processing core communicatively coupled to a last level cache (LLC) slice; and a cache control logic coupled to the plurality of processing cores, the cache control logic configured to perform one of: making an LLC slice of an inactive processing core available to an active processing core or power gating the LLC slice, based on estimating cache requirements by active processing cores.
摘要:
A processor is described having an out-of-order core to execute a first thread and a non-out-of-order core to execute a second thread. The processor also includes statistics collection circuitry to support calculation of the following: the first thread's performance on the out-of-order core; an estimate of the first thread's performance on the non-out-of-order core; the second thread's performance on the non-out-of-order core; an estimate of the second thread's performance on the out-of-order core.
摘要:
A thread scheduler may be used in a chip multiprocessor or symmetric multiprocessor system to schedule threads to processors. The scheduler may determine the bandwidth utilization of the two threads in combination and whether that utilization exceeds the threshold value. If so, the threads may be scheduled on different processor clusters that do not have the same paths between the common memory and the processors. If not, then the threads may be allocated on the same processor cluster that shares cache among processors.