ECSSD: Hardware/Data Layout Co-Designed In-Storage-Computing Architecture for Extreme Classification
S.F. Li, Fengbin Tu, Liu Liu, Jilan Lin, Zheng Wang, Yangwook Kang, Yufei Ding, Yuan Xie
Abstract
With the rapid growth of classification scale in deep learning systems, the final classification layer becomes extreme classification with a memory footprint exceeding the main memory capacity of the CPU or GPU. The emerging in-storage-computing technique offers an opportunity on account of the fact that SSD has enough storage capacity for the parameters of extreme classification. However, the limited performance of naive in-storage-computing schemes is insufficient to support the heavy workload of extreme classification.
Topics & Concepts
Computer scienceWorkloadMemory footprintComputer data storageArchitectureStorage managementMemory managementComputer architectureParallel computingOperating systemEmbedded systemDistributed computingSemiconductor memoryArtVisual artsAdvanced Data Storage TechnologiesFerroelectric and Negative Capacitance DevicesAdvanced Memory and Neural Computing