“What customers really want is high throughput per dollar,” said Geoff Tate, CEO of AI accelerator company Flex Logix.
Tate explained that having more TOPS doesn’t necessarily correlate with higher throughput. This is particularly true in edge applications where the batch size is 1. Applications such as data centers may increase their throughput by processing multiple inputs in parallel using larger batches (since they have TOPS to spare), but this is not often suitable for edge devices.
The problem with using TOPS to measure AI accelerator performance
Posted on Tuesday, December 10 2019 @ 11:47 CET by Thomas De Maesschalck
EE Times has an interesting article about the measurement of performance of AI accelerators. The piece, which you can read over here, argues that the frequently cited TOPS is not a good measure for performance: