FliT

Wei, Yuanhao; Ben-David, Naama; Friedman, Michal; Blelloch, Guy E.; Petrank, Erez

doi:10.1145/3503221.3508436

Cited by 10 publications

(10 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The update may have only reached the volatile CPU cache and can be lost upon a power failure or crash. To guarantee atomicity and durability of a pointer update, we update the forward pointer in HSIT using the flush-on-read technique [26,84,85]. Prism encodes a dirty bit to an unused bit into a forward pointer in HSIT entry.…”

Section: Cross-media Concurrency Controlmentioning

confidence: 99%

Prism: Optimizing Key-Value Store for Modern Heterogeneous Storage Devices

Song

Kim

Monga

et al. 2023

Proceedings of the 28th ACM International Conference on Architectural Support for Programming Languages and Operating Systems,

View full text Add to dashboard Cite

As data generation has been on an upward trend, storing vast volumes of data cost-effectively as well as efficiently accessing them is paramount. At the same time, today's storage landscape continues to diversify, from high-bandwidth storage devices such as NVMe SSDs to low-latency non-volatile memory (e.g., Intel Optane DCPMM). These heterogeneous storage devices have the potential to deliver high performance in terms of bandwidth and latency with cost efficiency, while achieving the performance and cost targets together still remains a challenging problem.We provide our solution, Prism, a novel key-value store that utilizes modern heterogeneous storage devices. Prism uses heterogeneous storage devices synergistically to harness the advantages of each storage device while suppressing their downsides. We devise new techniques to balance the latency-bandwidth tradeoff when reading from SSD. For ensuring multicore scalability and crash consistency of data across heterogeneous storage media, Prism proposes cross-storage concurrency control and cross-storage crash consistency protocols. Our evaluation shows that Prism outperforms state-of-the-art key-value stores by up to 13.1× with significantly lower tail latency.

show abstract

Section: Cross-media Concurrency Controlmentioning

confidence: 99%

Prism: Optimizing Key-Value Store for Modern Heterogeneous Storage Devices

Song

Kim

Monga

et al. 2023

Proceedings of the 28th ACM International Conference on Architectural Support for Programming Languages and Operating Systems,

View full text Add to dashboard Cite

show abstract

“…The ideas above have been implemented by libraries such as FliT [30] by providing read/write methods that efficiently track read-write dependencies between client operations. A programmer can gain durable linearizability guarantees by calling reads/writes of such a library in place of standard reads and writes, together with a call to a complete method before returning from an operation.…”

Section: Motivationmentioning

confidence: 99%

“…Our second contribution is a proof of correctness of the FliT library [30]. This proof leverages correctness of PLib by showing FliT in turn to refine PLib.…”

Section: Introductionmentioning

confidence: 98%

“…In recent years, there have been several proposals [5,13,26,30] that aim to help programmers with the transition to NVM, i.e., correct durability of concurrent programs developed for systems without NVM. Many of these approaches provide mechanisms that support transformations to durability, whereby program operations such as reads and writes to memory are replaced by calls to library operations that manage durability.…”

Section: Introductionmentioning

confidence: 99%

“…Mirror [13] transforms linearizable [14] objects (viz., data structures like stacks or queues) into durably linearizable [15] objects, provided that the object in question is non-blocking. FliT [30] is a generic proposal that ensures correct durability by tracking read-write dependencies on behalf of a programmer using persistent-write-back (pwb) instructions. Neither FliT nor Mirror have a fully formalised correctness proof of such transformations.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Fully Verified Persistency Library

Bodenmüller,

Derrick,

Dongol

et al. 2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Non-volatile memory (NVM) technologies offer DRAM-like speeds with the added benefit of failure resilience. However, developing concurrent programs for NVM can be challenging since programmers must consider both inter-thread synchronisation and durability aspects at the same time. To alleviate this, libraries such as FliT have been developed to manage transformations to durability, allowing a linearizable concurrent object to be converted into a durably linearizable one by replacing the reads/writes to memory by calls to corresponding operations of the FliT library. However, a formal proof of correctness for FliT is missing, and standard proof techniques for durable linearizability are challenging to apply, since FliT itself is not durably linearizable. In this paper, we study the problem of proving correctness of transformations to durability. First, we develop an abstract persistency library (called PLib) that operationally characterises transformations to durability. We prove soundness of PLib via a forward simulation coupled with a prophecy variable used as an oracle about future behaviour. Second, we show correctness of the library FliT by proving that FliTrefines PLib under the realistic PTSO memory model, i.e., the persistent version of TSO memory model implemented by Intel architectures. The proof of refinement between FliT and PLib has been mechanised within the theorem prover KIV. Taken together, these proofs guarantee that FliT is also sound wrt transformations to durability.

show abstract

SMT-Based Verification of Persistency Invariants of Px86 Programs

Marmanis

Vafeiadis

2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

FliT

Cited by 10 publications

References 17 publications

Prism: Optimizing Key-Value Store for Modern Heterogeneous Storage Devices

Prism: Optimizing Key-Value Store for Modern Heterogeneous Storage Devices

A Fully Verified Persistency Library

SMT-Based Verification of Persistency Invariants of Px86 Programs

Contact Info

Product

Resources

About