Explicit constructions of MSR codes for clustered distributed storage: The rack-aware storage model

Chen, Zitan; Barg, Alexander

doi:10.48550/arxiv.1901.04419

Cited by 1 publication

(5 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In short, the code in [12] and the code in the present paper achieve different tradeoff curves. Another recent work [27] suggested MSR codes for clustered DSS. On one hand, the authors of [27] assumed that intra-cluster repair bandwidth does not incur any cost, and only concerned with minimizing the amount of information transmitted across different clusters.…”

Section: B Related Workmentioning

confidence: 99%

“…Another recent work [27] suggested MSR codes for clustered DSS. On one hand, the authors of [27] assumed that intra-cluster repair bandwidth does not incur any cost, and only concerned with minimizing the amount of information transmitted across different clusters. On the other hand, the present paper considers both the intra-and cross-cluster repair burdens, and proposes MBR and MSR codes which minimize the overall repair burden for a given = β c /β I , the ratio of cross-to intra-cluster repair bandwidth.…”

Section: B Related Workmentioning

confidence: 99%

“…On the other hand, the present paper considers both the intra-and cross-cluster repair burdens, and proposes MBR and MSR codes which minimize the overall repair burden for a given = β c /β I , the ratio of cross-to intra-cluster repair bandwidth. Thus, the code in [27] and the code in this paper achieve different tradeoff curves. Compared to the conference version [1] of the current work, this paper adds an MSR code construction rule for ∈ [ 1 n−k , 1], and provides another MSR code for = 1 n−k , n = kL having a much smaller field size.…”

Section: B Related Workmentioning

confidence: 99%

“…Moreover, M = 18 and θ = 27 from ( 18), (19). First, we apply a [θ, M] = [27,18]−MDS code to the original source symbol s…”

Section: Examplementioning

confidence: 99%

“…6. Note that this contact can retrieve 18 coded symbols [27,18] code used in Algorithm 2, we obtain {s i } 18 i=1 from the retrieved 18 coded symbols. The same can be shown for any arbitrary contact of k = 3 nodes.…”

Section: Examplementioning

confidence: 99%

See 4 more Smart Citations

Explicit Constructions of MBR and MSR Codes for Clustered Distributed Storage

Sohn¹,

Choi²,

Moon³

2018

Preprint

View full text Add to dashboard Cite

This paper considers capacity-achieving coding for the clustered form of distributed storage that reflects practical storage networks. To reflect the clustered structure with limited cross-cluster communication bandwidths, nodes in the same cluster are set to communicate βI symbols, while nodes in other clusters can communicate βc ≤ βI symbols with one another. We provide two types of exact regenerating codes which achieve the capacity of clustered distributed storage: the minimum-bandwidth-regenerating (MBR) codes and the minimum-storage-regenerating (MSR) codes. First, we construct MBR codes for general parameter settings of clustered distributed storage. The suggested MBR code is a generalization of an existing code proposed by Rashmi et al., for scenarios where storage nodes are dispersed into L > 1 clusters. The proposed MBR code for the βc = 0 case requires a much smaller field size compared to existing local MBR codes. Secondly, we devise MSR codes for clustered distributed storage. Focus is given on two important cases: = 0 and ∈ [ 1 n−k , 1], where = βc/βI is the ratio of the available cross-to intra-cluster repair bandwidths, n is the total number of distributed nodes and k is the number of contact nodes in data retrieval. The former represents the scenario where cross-cluster communication is not allowed, while the latter corresponds to the case of minimum node storage overhead. For = 0, two existing locally repairable codes are proven to be MSR codes for the clustered model. For ∈ [ 1 n−k , 1], existing MSR codes for the non-clustered model are applicable to clustered scenarios with a simple modification. Finally, under the settings of = 1 n−k and n = kL, an MSR code is suggested which is based on simple MDS codes and requires a smaller field size for symbols than the existing code for L ≥ 3.

show abstract

Section: B Related Workmentioning

confidence: 99%

Section: B Related Workmentioning

confidence: 99%