Gimbal: enabling multi-tenant storage disaggregation on SmartNIC JBOFs

J Min, M Liu, T Chugh, C Zhao, A Wei, IH Doh… - Proceedings of the …, 2021 - dl.acm.org
Emerging SmartNIC-based disaggregated NVMe storage has become a promising storage
infrastructure due to its competitive IO performance and low cost. These SmartNIC JBOFs …

Burstable Cloud Block Storage with Data Processing Units

J Shu, K Qian, E Zhai, X Liu, X Jin - 18th USENIX Symposium on …, 2024 - usenix.org
Cloud block storage (CBS) is a key pillar of public clouds. Today's CBS distinguishes itself
from physical counterparts (eg, SSDs) by offering unique burst capability as well as …

2dfq: Two-dimensional fair queuing for multi-tenant cloud services

J Mace, P Bodik, M Musuvathi, R Fonseca… - Proceedings of the …, 2016 - dl.acm.org
In many important cloud services, different tenants execute their requests in the thread pool
of the same process, requiring fair sharing of resources. However, using fair queue …

Baoverlay: A block-accessible overlay file system for fast and efficient container storage

Y Sun, J Lei, S Shin, H Lu - Proceedings of the 11th ACM Symposium on …, 2020 - dl.acm.org
Container storage commonly relies on overlay file systems to interpose read-only container
images upon backing file systems. While being transparent to and compatible with most …

Effectively Mitigating {I/O} Inactivity in {vCPU} Scheduling

W Jia, C Wang, X Chen, J Shan, X Shang… - 2018 USENIX Annual …, 2018 - usenix.org
In clouds where CPU cores are time-shared by virtual CPUs (vCPU), vCPUs are scheduled
and descheduled by the virtual machine monitor (VMM) periodically. In each virtual machine …

{vSMT-IO}: Improving {I/O} Performance and Efficiency on {SMT} Processors in Virtualized Clouds

W Jia, J Shan, TO Li, X Shang, H Cui… - 2020 USENIX Annual …, 2020 - usenix.org
The paper focuses on an under-studied yet fundamental issue on Simultaneous Multi-
Threading (SMT) processors—how to schedule I/O workloads, so as to improve I/O …

GC-steering: GC-aware request steering and parallel reconstruction optimizations for SSD-based RAIDs

S Wu, W Zhu, Y Han, H Jiang, B Mao… - … on Computer-Aided …, 2020 - ieeexplore.ieee.org
Solid-state disk (SSD)-based redundant array of independent disks (RAIDs) have been
widely deployed in high-end enterprize systems to provide high-performance and highly …

Can i/o variability be reduced on qos-less hpc storage systems?

D Huang, Q Liu, J Choi, N Podhorszki… - IEEE Transactions …, 2018 - ieeexplore.ieee.org
For a production high-performance computing (HPC) system, where storage devices are
shared between multiple applications and managed in a best effort manner, I/O contention is …

BFQ, Multiqueue-Deadline, or Kyber? Performance Characterization of Linux Storage Schedulers in the NVMe Era

Z Ren, K Doekemeijer, N Tehrany… - Proceedings of the 15th …, 2024 - dl.acm.org
Flash SSDs have become the de-facto choice to deliver high I/O performance to modern
data-intensive workloads. These workloads are often deployed in the cloud, where multiple …

Serifos: Workload consolidation and load balancing for SSD based cloud storage systems

Z Yao, I Papapanagiotou, R Griffith - arXiv preprint arXiv:1512.06432, 2015 - arxiv.org
Achieving high performance in virtualized data centers requires both deploying high
throughput storage clusters, ie based on Solid State Disks (SSDs), as well as optimally …