SPDK Vhost-NVMe: Accelerating I/Os in Virtual Machines on NVMe SSDs via User Space Vhost Target

Yang, Ziye; Liu, Changpeng; Zhou, Yanbo; Liu, Xiaodong; Cao, Gang

doi:10.1109/sc2.2018.00016

Cited by 19 publications

(5 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Naturally, the performance benefits of kernel bypass should be retained when virtualizing devices to share them among multiple clients. For instance with storage devices, software-based approaches like SPDK-vhost [61] or MDev-NVMe [49] implement device virtualization by means of dedicated virtualization servers that use shared memory for communicating with their clients, thus avoiding the tax of entering and returning from the kernel. However, such approaches consume a significant amount of CPU resources, as they have to poll on said memory regions.…”

Section: Related Workmentioning

confidence: 99%

Efficient and scalable core multiplexing with M³v

Asmussen

Haas

Weinhold

et al. 2022

Proceedings of the 27th ACM International Conference on Architectural Support for Programming Languages and Operating Systems

View full text Add to dashboard Cite

The M 3 system (ASPLOS '16) proposed a hardware/software codesign that simplifies integration between general-purpose cores and special-purpose accelerators, allowing users to easily utilize them in a unified manner. M 3 is a tiled architecture, whose tiles (cores and accelerators) are partitioned between applications, such that each tile is dedicated to its own application.The M 3 x system (ATC '19) extended M 3 by trading off some isolation to enable coarse-grained multiplexing of tiles among multiple applications. With M 3 x, if source tile 𝑡 1 runs code of application 𝑝 and sends a message 𝑚 to destination tile 𝑡 2 while 𝑡 2 is currently not associated with 𝑝, then 𝑚 is forwarded to the right place through a łslow pathž, via some special OS tile.In this paper, we present M 3 v, which extends M 3 x by further trading off some isolation between applications to support łfast pathž communication that does not require the said OS tile's involvement. Thus, with M 3 v, a tile can be efficiently multiplexed between applications provided it is a general-purpose core. M 3 v achieves this goal by 1) adding a local multiplexer to each such core, and by 2) virtualizing the core's hardware component responsible for cross-tile communications. We prototype M 3 v using RISC-V cores on an FPGA platform and show that it significantly outperforms M 3 x and may achieve competitive performance to Linux. CCS CONCEPTS• Computer systems organization → Architectures; • Security and privacy → Operating systems security; • Software and its engineering → Process management; Communications management.

show abstract

Section: Related Workmentioning

confidence: 99%

Efficient and scalable core multiplexing with M³v

Asmussen

Haas

Weinhold

et al. 2022

Proceedings of the 27th ACM International Conference on Architectural Support for Programming Languages and Operating Systems

View full text Add to dashboard Cite

show abstract

“…Enhancing I/O performance virtualized environments are also studied actively [14][15][16][17][18][25][26][27][28]. [27].…”

Section: Related Workmentioning

confidence: 99%

“…Yang et al propose a novel I/O service, called SPDK-vhost-NVMe, to reduce the I/O virtualization overhead [18]. They identify that the virtualization overhead comes from three parts, namely in guest OS, host OS and context switch (e.g., VM_Exit) between them.…”

Section: Related Workmentioning

confidence: 99%

“…There are a variety of virtualization techniques, which can be classified into three categories: CPU virtualization [8][9][10], memory virtualization [11][12][13] and I/O virtualization [14][15][16][17][18]. For example, Borrowed Virtual Time (BVT) is a scheduling algorithm for allocating CPUs among virtual machines in a fair-share manner while providing fast dispatch [8].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Direct-Virtio: A New Direct Virtualized I/O Framework for NVMe SSDs

2021

View full text Add to dashboard Cite

Virtualization is a core technology for cloud computing, server consolidation and multi-platform support. However, there is a concern regarding performance degradation due to the duplicated I/O stacks virtualization environments. In this paper, we propose a new I/O framework, we refer to it as Direct-Virtio, that manipulates storage directly, which makes it feasible to avoid the duplicated overhead. In addition, we devise two novel mechanisms, called vectored I/O and adaptive polling, to process multiple I/O requests collectively and to check I/O completion efficiently. Real implementation-based evaluation shows that our proposal can enhance performance for both micro and macro benchmarks.

show abstract

“…Many prior studies [8,12,56,71,75,77] use SPDK to directly access the underlying SSD by bypassing all kernel modules in the storage stack. Even though the kernel-bypass is a very promising and attractive approach, it is limited to apply the kernel-bypass to a system that does not have a file system or ACID support.…”

Section: How Much Can Spdk Reduce the System-level Latency Under A Re...mentioning

confidence: 99%

Faster than Flash: An In-Depth Study of System Challenges for Emerging Ultra-Low Latency SSDs

Koh

Jang

Lee

et al. 2019

2019 IEEE International Symposium on Workload Characterization (IISWC)

View full text Add to dashboard Cite

Emerging storage systems with new flash exhibit ultra-low latency (ULL) that can address performance disparities between DRAM and conventional solid state drives (SSDs) in the memory hierarchy. Considering the advanced low-latency characteristics, different types of I/O completion methods (polling/hybrid) and storage stack architecture (SPDK) are proposed. While these new techniques are expected to take costly software interventions off the critical path in ULL-applied systems, unfortunately no study exists to quantitatively analyze system-level characteristics and challenges of combining such newly-introduced techniques with real ULL SSDs.In this work, we comprehensively perform empirical evaluations with 800GB ULL SSD prototypes and characterize ULL behaviors by considering a wide range of I/O path parameters, such as different queues and access patterns. We then analyze the efficiencies and challenges of the polled-mode and hybrid polling I/O completion methods (added into Linux kernels 4.4 and 4.10, respectively) and compare them with the efficiencies of a conventional interrupt-based I/O path. In addition, we revisit the common expectations of SPDK by examining all the system resources and parameters. Finally, we demonstrate the challenges of ULL SSDs in a real SPDK-enabled server-client system. Based on the performance behaviors that this study uncovers, we also discuss several system implications, which are required to take a full advantage of ULL SSD in the future.

show abstract

SPDK Vhost-NVMe: Accelerating I/Os in Virtual Machines on NVMe SSDs via User Space Vhost Target

Cited by 19 publications

References 4 publications

Efficient and scalable core multiplexing with M³v

Efficient and scalable core multiplexing with M³v

Direct-Virtio: A New Direct Virtualized I/O Framework for NVMe SSDs

Faster than Flash: An In-Depth Study of System Challenges for Emerging Ultra-Low Latency SSDs

Contact Info

Product

Resources

About