论文标题
CNN加速器的可靠且节能的MLC STT-RAM缓冲液
Reliable and Energy Efficient MLC STT-RAM Buffer for CNN Accelerators
论文作者
论文摘要
我们提出了一种轻巧的方案，其中更改数据块的形成方式使其可以忍受软误差明显好于基线。我们工作背后的关键见解是，在每个卷积层之后，CNN权重在-1和1之间进行了归一化，并且在半精确的浮点表示中，这使一点未使用。通过利用未使用的位，我们为最重要的位创建了一个备份，以保护其免受软错误。同样，考虑到在MLC STT-RAMS中，内存操作的成本（读取和写入），并且单元的可靠性取决于内容（某些模式需要较大的当前和更长的时间，而它们更容易容易出现软错误），我们将数据块重新排列以最大程度地减少昂贵的位模式的数量。与无错误的基线相比，这两种技术相比提供了相同的准确性，同时将读写能量分别提高了9％和6％。
We propose a lightweight scheme where the formation of a data block is changed in such a way that it can tolerate soft errors significantly better than the baseline. The key insight behind our work is that CNN weights are normalized between -1 and 1 after each convolutional layer, and this leaves one bit unused in half-precision floating-point representation. By taking advantage of the unused bit, we create a backup for the most significant bit to protect it against the soft errors. Also, considering the fact that in MLC STT-RAMs the cost of memory operations (read and write), and reliability of a cell are content-dependent (some patterns take larger current and longer time, while they are more susceptible to soft error), we rearrange the data block to minimize the number of costly bit patterns. Combining these two techniques provides the same level of accuracy compared to an error-free baseline while improving the read and write energy by 9% and 6%, respectively.
