Block management in solid-state devices

Abhishek Rajimwale, Vijayan Prabhakaran, John D. Davis

2009 · 60 citations · 16 references

Concepts

TL;DR

Solid‑state devices promise to replace hard disk drives, yet legacy spinning‑media assumptions remain entrenched in the software stack. The authors aim to reexamine these system‑level assumptions for SSDs and argue that block management should be removed from the file system and delegated to the SSD. They propose delegating block management to the SSD to eliminate storage‑specific assumptions that have accumulated over decades. Their findings indicate that legacy assumptions are invalidated by SSDs, causing performance and lifetime problems due to an incorrect division of labor between file systems and storage, and that object‑based storage can address this issue.

Abstract

Solid-state devices (SSDs) have the potential to replace traditional hard disk drives (HDDs) as the de facto storage medium. Unfortunately, there are several decades of spinning-media assumptions embedded in the software stack as an [20]. In this paper, we revisit these system-level assumptions in light of SSDs and find that several of them are invalidated by SSDs, breaking the unwritten contract and resulting in poor performance and lifetime. The underlying cause is the incorrect division of labor between file systems and storage. Block management must be removed from the file system and delegated to the SSD to prevent further accumulation of storage-specific assumptions. We find that object-based storage is an appropriate way to achieve this.

References

16