arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2201.10095cs.LGcs.ARcs.DCcs.PF
收藏

RecShard:基于统计学习的工业规模神经网络推荐内存优化

RecShard: Statistical Feature-Based Memory Optimization for Industry-Scale Neural Recommendation

Stanford University(斯坦福大学);Meta

Geet Sethi, Bilge Acun, Niket Agarwal, Christos Kozyrakis, Caroline Trippel, Carole-Jean Wu

摘要:

我们提出了RecShard,一种用于深度学习推荐模型(DLRM)的细粒度嵌入表(EMB)分区和放置技术。RecShard的设计基于两个关键观察。首先,并非所有EMB都是相等的,EMB内的所有行在访问模式方面也不是相等的。EMB表现出不同的存储器特性,为跨分层存储层次结构的智能EMB分区和放置提供了性能优化机会。其次,在现代DLRM中,EMB用作哈希表。结果,EMB显示出有趣的现象,如生日悖论,使EMB严重利用不足。RecShard基于训练数据分布和模型特征以及底层分层存储层次结构的带宽特征,为一组EMB确定最佳EMB分片策略。通过这样做,RecShard在容量受限的DLRM中实现了平均6倍以上的EMB训练吞吐量。吞吐量的增加来自于EMB负载平衡提高了12倍以上,以及对较慢内存的访问减少了87倍以上。

英文摘要:

We propose RecShard, a fine-grained embedding table (EMB) partitioning and placement technique for deep learning recommendation models (DLRMs). RecShard is designed based on two key observations. First, not all EMBs are equal, nor all rows within an EMB are equal in terms of access patterns. EMBs exhibit distinct memory characteristics, providing performance optimization opportunities for intelligent EMB partitioning and placement across a tiered memory hierarchy. Second, in modern DLRMs, EMBs function as hash tables. As a result, EMBs display interesting phenomena, such as the birthday paradox, leaving EMBs severely under-utilized. RecShard determines an optimal EMB sharding strategy for a set of EMBs based on training data distributions and model characteristics, along with the bandwidth characteristics of the underlying tiered memory hierarchy. In doing so, RecShard achieves over 6 times higher EMB training throughput on average for capacity constrained DLRMs. The throughput increase comes from improved EMB load balance by over 12 times and from the reduced access to the slower memory by over 87 times.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑