Data-Pattern Enabled Self-Recovery Low-Power Storage System for Big Video Data

Edstrom, Jonathon; Chen, Dongliang; Gong, Yifu; Wang, Jinhui

doi:10.1109/tbdata.2017.2750699

Cited by 16 publications

(8 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This work has presented a comprehensive representation scheme of the involuntary regulation of the behavior of student. Apart from this, there are various other schemes towards big data analytics e.g., tensor-based scheme [27], compression based on context [28], pattern analysis [29], deep learning [30], clustering technique [31]. Existing system has also witnessed an extensive usage of Hadoop framework.…”

Section: Related Workmentioning

confidence: 99%

A Comprehensive Framework for Big Data Analytics in Education

Shidaganti¹,

Prakash²

2021

IJACSA

View full text Add to dashboard Cite

With the adoption of cloud services for hosting knowledge delivery system in educational domain, there is a surplus quantity of education data being generated every day by current learning management system. Such data are associated with certain typical complexities that impose significant challenges for existing database management and analytics. Review of existing approaches towards educational data highlights that they do not offer full-fledged solution towards analytics and still there is an open-end problem. Therefore, the proposed system introduces a comprehensive framework which offers integrated operation of transformation, data quality, and predictive analytics. The emphasis is more towards achieving distributed analytical operation towards educational data in cloud. Implemented using analytical research methodology, the proposed system shows better analytical performance with respect to frequently used educational data analytical approaches.

show abstract

Section: Related Workmentioning

confidence: 99%

A Comprehensive Framework for Big Data Analytics in Education

Shidaganti¹,

Prakash²

2021

IJACSA

View full text Add to dashboard Cite

show abstract

“…and let E(i 1 : i 2 ) denote the sum of squares of deviations from the mean of the entries of x(i 1 : i 2 ) as defined in Equ. (4).…”

Section: Approximate Subsequencementioning

confidence: 99%

“…In many practical applications mining of data that is in the form of time series [5,10] is of interest and this has led to the development of bespoke approaches for tasks such as pattern discovery and clustering [37,21,9], classification [7,20], rule discovery [30,34], and summarisation [13]. As with standard data mining, indexing is a fundamental technique for efficiently accessing and querying data when performing these tasks [6,4]. However, when indexing time series data the choice of similarity measurement is a key consideration [23], particularly when they are not aligned temporally.…”

Section: Introductionmentioning

confidence: 99%

Time series indexing by dynamic covering with cross-range constraints

et al. 2020

View full text Add to dashboard Cite

Time series indexing plays an important role in querying and pattern mining of big data. This paper proposes a novel structure for tightly covering a given set of time series under the dynamic time warping similarity measurement. The structure, referred to as Dynamic Covering with cross-Range Constraints (DCRC), enables more efficient and scalable indexing to be developed than current hypercube based partitioning approaches. In particular, a lower bound of the DTW distance from a given query time series to a DCRC-based cover set is introduced. By virtue of its tightness, which is proven theoretically, the lower bound can be used for pruning when querying on an indexing tree. If the DCR-C based Lower Bound (LB DCRC) of an upper node in an index tree is larger than a given threshold, all child nodes can be pruned yielding a significant reduction in

show abstract

“…The explosion of multimedia data such as audio, images and video puts stringent delay requirements in I/O access in today's storage systems [1,2]. Multi-tiered storage systems are widely used to meet such increasing demands with a reasonable cost [3][4][5].…”

Section: Introductionmentioning

confidence: 99%

“…However, it does not perform well on long-term access frequency predictions. As shown in Figure 1, the block access patterns change a lot at different time periods 1 , which brings a great curse for the above methods. In recent years, deep learning approaches have achieved great success in natural language processing [17,18], time series prediction [19,20] and related areas.…”

Section: Introductionmentioning

confidence: 99%

Block Popularity Prediction for Multimedia Storage Systems Using Spatial-Temporal-Sequential Neural Networks

Cheng

Zhang

et al. 2021

Proceedings of the 29th ACM International Conference on Multimedia

View full text Add to dashboard Cite

Predicting block popularity is of crucial importance for data placement in multi-tiered multimedia storage systems. Traditional methods, such as least recently used and exponential smoothing, are commonly employed to predict future block access frequencies and fail to achieve good performance for complex and changing access patterns. Recently, deep neural networks have brought great success to pattern recognition and prediction, which motivates us to introduce deep learning to solve the problem of block popularity prediction. In this paper, we first analyze and verify the temporal and spatial correlations among the multimedia I/O traces. Then, we design a multi-dimension feature to capture such correlations, which serves as the input of the designed deep neural network. A spatial-temporal-sequential neural network (STSNN) and its variants that capture the locality information, time dependency information, and block sequential information are proposed to predict the block popularity. We systematically evaluate our STSNN models against six baseline models from three different categories, i.e., heuristic methods, regression methods and neural network-based methods. Experiment results show that our proposed STSNN models are very promising for predicting block access frequencies under some of Huawei and Microsoft datasets and particularly achieve 2-6 times better performance compared with the baselines in terms of the I/O hit ratio, I/O recall rate and I/O prediction ratio under the Microsoft 64 MB-block dataset.

show abstract

Data-Pattern Enabled Self-Recovery Low-Power Storage System for Big Video Data

Cited by 16 publications

References 25 publications

A Comprehensive Framework for Big Data Analytics in Education

A Comprehensive Framework for Big Data Analytics in Education

Time series indexing by dynamic covering with cross-range constraints

Block Popularity Prediction for Multimedia Storage Systems Using Spatial-Temporal-Sequential Neural Networks

Contact Info

Product

Resources

About