About this Digital Document
We present KVCG, a novel heterogeneous key-value store whose primary objective is to serve client requests targeting frequently accessed (hot) keys at sub-millisecond latency and requests targeting less frequently accessed (cold) keys with high throughput. To accomplish this goal, KVCG deploys an architecture where requests on hot keys are routed to a software cache operated by CPU threads, while the remainder are offloaded to a data repository optimized for execution on modern GPU devices. Cold/hot partitioning is done at runtime through a model trained with the incoming workload. Against a state-of-the-art competitor, we obtain up to 34x improvement in latency.
Full Title
KVCG: A Heterogeneous Key-Value Store for Skewed Workloads
Member of
Contributor(s)
Date Issued
2021-06-14
Type
Genre
Form
electronic documents
Department name
Computer Science
Digital Format
electronic documents
Media type
Creator role
Graduate Student
Faculty
Miller, . dePaul, Nelson, . J., Hassan, . A., & Palmieri, . R. (2021). KVCG: A Heterogeneous Key-Value Store for Skewed Workloads (1–). https://preserve.lehigh.edu/lehigh-scholarship/faculty-staff-publications/faculty-publications/kvcg-heterogeneous-key-value
Miller, dePaul, Jacob Nelson, Ahmed Hassan, and Roberto Palmieri. 2021. “KVCG: A Heterogeneous Key-Value Store for Skewed Workloads”. https://preserve.lehigh.edu/lehigh-scholarship/faculty-staff-publications/faculty-publications/kvcg-heterogeneous-key-value.
Miller, dePaul, et al. KVCG: A Heterogeneous Key-Value Store for Skewed Workloads. 14 June 2021, https://preserve.lehigh.edu/lehigh-scholarship/faculty-staff-publications/faculty-publications/kvcg-heterogeneous-key-value.