arXiv subjects

Hengfeng Wei

Publications and source records attributed to Hengfeng Wei.

22 records · Page 2Linked to original sources

arXiv2019

Inversion-based Measurement of Data Consistency for Read/Write Registers

Both providers and consumers of distributed storage services benefit from the quantification of the severity of consistency violations. However, existing methods fail to capture a typical pattern of violation - the disorder among operations with reference to the ideally strong atomic execution. Such disorder are often seen in Internet applications based on distributed storage services, such as instant messaging. To this end, we use inversions in a permutation with reference to the totally-ordered permutation to measure the consistency of data. The $i$-atomicity model is proposed and a (pseudo-)polynomial verification algorithm for a restricted type of histories is proposed. The basic idea of the verification algorithm is brute-force search of the permutation with less than $i$ inversions, which requires exponential cost. Reasonable assumptions abstracted from applications scenarios can be leveraged to prune the search space, which yields an efficient polynomial verification algorithm.

cs.DC↗

arXiv2018

Specification and Implementation of Replicated List: The Jupiter Protocol Revisited

The replicated list object is frequently used to model the core functionality of replicated collaborative text editing systems. Since 1989, the convergence property has been a common specification of a replicated list object. Recently, Attiya et al. proposed the strong/weak list specification and conjectured that the well-known Jupiter protocol satisfies the weak list specification. The major obstacle to proving this conjecture is the mismatch between the global property on all replica states prescribed by the specification and the local view each replica maintains in Jupiter using data structures like 1D buffer or 2D state space. To address this issue, we propose CJupiter (Compact Jupiter) based on a novel data structure called $n$-ary ordered state space for a replicated client/server system with $n$ clients. At a high level, CJupiter maintains only a single $n$-ary ordered state space which encompasses exactly all states of each replica. We prove that CJupiter and Jupiter are equivalent and that CJupiter satisfies the weak list specification, thus solving the conjecture above.

cs.DC↗

arXiv2015

Almost Strong Consistency: "Good Enough" in Distributed Storage Systems

A consistency/latency tradeoff arises as soon as a distributed storage system replicates data. For low latency, modern storage systems often settle for weak consistency conditions, which provide little, or even worse, no guarantee for data consistency. In this paper we propose the notion of almost strong consistency as a better balance option for the consistency/latency tradeoff. It provides both deterministically bounded staleness of data versions for each read and probabilistic quantification on the rate of "reading stale values", while achieving low latency. In the context of distributed storage systems, we investigate almost strong consistency in terms of 2-atomicity. Our 2AM (2-Atomicity Maintenance) algorithm completes both reads and writes in one communication round-trip, and guarantees that each read obtains the value of within the latest 2 versions. To quantify the rate of "reading stale values", we decompose the so-called "old-new inversion" phenomenon into concurrency patterns and read-write patterns, and propose a stochastic queueing model and a "timed balls-into-bins model" to analyze them, respectively. The theoretical analysis not only demonstrates that "old-new inversions" rarely occur as expected, but also reveals that the read-write pattern dominates in guaranteeing such rare data inconsistencies. These are further confirmed by the experimental results, showing that 2-atomicity is "good enough" in distributed storage systems by achieving low latency, bounded staleness, and rare data inconsistencies.

cs.DC↗

arXiv2013

Verifying PRAM Consistency over Read/Write Traces of Data Replicas

Data replication technologies enable efficient and highly-available data access, thus gaining more and more interests in both the academia and the industry. However, data replication introduces the problem of data consistency. Modern commercial data replication systems often provide weak consistency for high availability under certain failure scenarios. An important weak consistency is Pipelined-RAM (PRAM) consistency. It allows different processes to hold different views of data. To determine whether a data replication system indeed provides PRAM consistency, we study the problem of Verifying PRAM Consistency over read/write traces (or VPC, for short). We first identify four variants of VPC according to a) whether there are Multiple shared variables (or one Single variable), and b) whether write operations can assign Duplicate values (or only Unique values) for each shared variable; the four variants are labeled VPC-SU, VPC-MU, VPC-SD, and VPC-MD. Second, we present a simple VPC-MU algorithm, called RW-CLOSURE. It constructs an operation graph $\mathcal{G}$ by iteratively adding edges according to three rules. Its time complexity is $O(n^5)$, where n is the number of operations in the trace. Third, we present an improved VPC-MU algorithm, called READ-CENTRIC, with time complexity $O(n^4)$. Basically it attempts to construct the operation graph $\mathcal{G}$ in an incremental and efficient way. Its correctness is based on that of RW-CLOSURE. Finally, we prove that VPC-SD (so is VPC-MD) is $\sf{NP}$-complete by reducing the strongly $\sf{NP}$-complete problem 3-PARTITION to it.

cs.DC↗