Erasure coded storage systems for cloud storage — challenges and opportunities

Lee, Ojus Thomas; Kumar, Suresh; Chandran, Priya

doi:10.1109/icdse.2016.7823943

Cited by 10 publications

(8 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This implies that, from a landscape view, the overall most expensive case on dispersion always happened under the combination (15,2), whereas the most expensive case of recovery always happened under the combination (15,14). Furthermore, we must consider that there are alternative ways to address the costs of the algorithm since this study shows the performance for each value of n ∈ [3,15] This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.…”

Section: Discussionmentioning

confidence: 99%

“…Consider, for example, two instances of IDA with parameters (5,3) and (15,13), respectively. Both offer the same tolerance, but the second case also provides a lower cost in redundancy (their stretching values being 0.666 vs 0.153, respectively).…”

Section: Discussionmentioning

confidence: 99%

“…After the original publication of the IDA, Rabin himself proposed extending the scope of his initial work and sug-VOLUME 4, 2016 gested some applications for storage [2]. The use of coding techniques to optimize storage space has gained relevance in recent years, to the extent that systems of greater capacities are being built and the gap between raw capacity and effective capacity is being reduced [3].…”

Section: Related Workmentioning

confidence: 99%

“…(3,2) C 19 :(8,5) C 37 :(11,2) C 55 :(12,11) C 73 :(14,8) C 2 :(4,2) C 20 :(8,6) C 38 :(11,3) C 56 :(13,2) C 74 :(14,9) C 3 :(4,3)C21 :(8,7) C 39 :(11,4) C 57 :(13,3) C 75 :(14,10) C 4 :(5,2) C 22 :(9,2) C 40 :(11,5) C 58 :(13,4) C 76 :(14,11) C 5 :(5,3) C 23 :(9,3) C 41 :(11,6) C 59 :(13,5) C 77 :(14,12) C 6 :(5,4) C 24 :(9,4) C 42 :(11,7) C 60 :(13,6) C 78 :(14,13) C 7 :(6,2) C 25 :(9,5) C 43 :(11,8) C 61 :(13,7) C 79 :(15,2) C 8 :(6,3) C 26 :(9,6) C 44 :(11,9) C 62 :(13,8) C 80 :(15,3) C 9 :(6,4) C 27 :(9,7) C 45 :(11,10) C 63 :(13,9) C 81 :(15,4) C 10 :(6,5) C 28 :(9,8) C 46 :(12,2) C 64 :(13,10) C 82 :(15,5) C 11 :(7,2) C 29 :(10,2) C 47 :(12,3) C 65 :(13,11) C 83 :(15,6) C 12 :(7,3) C 30 :(10,3) C 48 :(12,4) C 66 :(13,12) C 84 :(15,7) C 13 :(7,4) C 31 :(10,4) C 49 :(12,5) C 67 :(14,2) C 85 :(15,8) C 14 :(7,5) C 32 :(10,5) C 50 :(12,6) C 68 :(14,3) C 86 :(15,9) C 15 :(7,6) C 33 :(10,6) C 51 :(12,7) C 69 :(14,4) C 87 :(15,10) C 16 :(8,2) C 34 :(10,7) C 52 :(12,8) C 70 :(14,5) C 88 :(15,11) C 17 :(8,3) C 35 :(10,8) C 53 :(12,9) C 71 :(14,6) C 89 :(15,12) C 18 :(8,4) C 36 :(10,9) C 54 :(12,10) C 72 :(14,7) C 90 :(15…”

mentioning

confidence: 99%

See 3 more Smart Citations

On the Complexity and Performance of the Information Dispersal Algorithm

et al. 2020

View full text Add to dashboard Cite

The Information Dispersal Algorithm (IDA) has become a key component in several faulttolerant massive storage systems. From a theoretical point of view, it is a linear transformation over a finite field on the vectors that make up a given file. Direct transformation adds redundancy, splitting the initial file into a new set of files called dispersals. The inverse transformation recovers the original file from a subset of dispersals. This piece of research demonstrates the impact of input and output (I/O) operations on direct and inverse transformations. Different alternatives to control the exchange of elements between RAM and disk were evaluated, which is the key operation to build a vector in memory and store its entries in a file. First, the impact of the working finite field was tested; second, the impact of the use of a buffer for exchange between the RAM and the hard disk, and finally, several instances of the algorithm with which to evaluate the impact of parallelism were simultaneously deployed. The results demonstrate that the combination of these factors may have an important effect on the speed of both direct and inverse procedures.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

On the Complexity and Performance of the Information Dispersal Algorithm

et al. 2020

View full text Add to dashboard Cite

show abstract

“…En 1990 Michael O. Rabin presenta posibles aplicaciones para IDA [1], desde entonces el algoritmo ha sido estudiado y utilizado para aplicaciones como el encaminamiento en redes de computadoras en paralelo tolerante a fallas [2], almacenamiento en la nube [3] e incluso Afianian utiliza IDA para el procesamiento de archivos en dispositivos móviles [4].…”

Section: Figura 1 Funcionamiento De Idaunclassified

Estudio sobre el desempeño del Algoritmo de Dispersión de Información

Jiménez¹

2018

Productos Del Vínculo Investigación-Docencia

View full text Add to dashboard Cite

El algoritmo de dispersión de información (IDA por sus siglas en inglés), es un componente relevante en varios sistemas de almacenamiento de altas prestaciones. En sus aspectos fundamentales se trata de un conjunto de productos matriz-vector, realizados sobre un campo finito. En este trabajo mostramos el impacto del orden del campo sobre la velocidad del algoritmo. En particular, cuando se realizan sobre campos de característica 2, que pueden entenderse como cadenas de dígitos binarios. En un primer caso, se trabajan con cadenas de 8 bits, y en un segundo caso, con cadenas de 16 bits. Esta diferencia en longitud, tiene un impacto significativo sobre las operaciones de lectura y escritura de archivos.

show abstract