Large-scale Graph Neural Network Services through Computational SSD and In-Storage Processing Architectures

Kwon, Miryeong; Gouk, Donghyun; Lee, Sang‐Won

doi:10.1109/hcs55958.2022.9895623

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although several works on IOD have begun to appear [14,62], they mainly target hardwarelevel partitioning for better workload isolation, and none of them address IOD-PLM challenges. We present more detailed comparisons between IODA and related work in Section 6.2, qualitatively and quantitatively.…”

Section: :7mentioning

confidence: 99%

Extending and Programming the NVMe I/O Determinism Interface for Flash Arrays

Putra

Shi

et al. 2023

ACM Trans. Storage

View full text Add to dashboard Cite

Predictable latency on flash storage is a long-pursuit goal, yet, unpredictability stays due to the unavoidable disturbance from many well-known SSD internal activities. To combat this issue, the recent NVMe IO Determinism (IOD) interface advocates host-level controls to SSD internal management tasks. While promising, challenges remain on how to exploit it for truly predictable performance. We present IODA , an I/O deterministic flash array design built on top of small but powerful extensions to the IOD interface for easy deployment. IODA exploits data redundancy in the context of IOD for a strong latency predictability contract. In IODA , SSDs are expected to quickly fail an I/O on purpose to allow predictable I/Os through proactive data reconstruction. In the case of concurrent internal operations, IODA introduces busy remaining time exposure and predictable-latency-window formulation to guarantee predictable data reconstructions. Overall, IODA only adds 5 new fields to the NVMe interface and a small modification in the flash firmware, while keeping most of the complexity in the host OS. Our evaluation shows that IODA improves the 95–99.99þ latencies by up to 75 ×. IODA is also the nearest to the ideal, no disturbance case compared to 7 state-of-the-art preemption, suspension, GC coordination, partitioning, tiny-tail flash controller, prediction, and proactive approaches.

show abstract

Section: :7mentioning

confidence: 99%