Paper Website abstract bibtex
The ever increasing amount of data being handled in data centers causes an intrinsic inefficiency: moving data around is expensive in terms of bandwidth, latency, and power con-sumption, especially given the low computational complex-ity of many database operations. In this paper we explore near-data processing in database engines, i.e., the option of offloading part of the compu-tation directly to the storage nodes. We implement our ideas in Caribou, an intelligent distributed storage layer in-corporating many of the lessons learned while building sys-tems with specialized hardware. Caribou provides access to DRAM/NVRAM storage over the network through a simple key-value store interface, with each storage node providing high-bandwidth near-data processing at line rate and fault tolerance through replication. The result is a highly effi-cient, distributed, intelligent data storage that can be used to both boost performance and reduce power consumption and real estate usage in the data center thanks to the micro-server architecture adopted.