English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

We optimize pipeline parallelism for deep neural network (DNN) inference bypartitioning model graphs into $k$ stages and minimizing the running time ofthe bottleneck stage, including communication. We give practical and effectivealgorithms for this NP-hard problem, but our emphasis is on tackling thepractitioner's dilemma of deciding when a solution is good enough. To this end,we design novel mixed-integer programming (MIP) relaxations for proving lowerbounds. Applying these methods to a diverse testbed of 369 production models,for $k \in \{2, 4, 8, 16, 32, 64\}$, we empirically show that these lowerbounds are strong enough to be useful in practice. Our lower bounds aresubstantially stronger than standard combinatorial bounds. For example,evaluated via geometric means across a production testbed with $k = 16$pipeline stages, our MIP formulations raise the lower bound from 0.4598 to0.9452, expressed as a fraction of the best partition found. In other words,our improved lower bounds close the optimality gap by a factor of 9.855x.

Practical Performance Guarantees for Pipelined DNN Inference