scispace - formally typeset
Search or ask a question
Journal ArticleDOI

An adaptive caching algorithm suitable for time-varying user accesses in VOD systems

01 Dec 2015-Multimedia Tools and Applications (Springer US)-Vol. 74, Iss: 24, pp 11117-11137
TL;DR: An approach to choose the cached videos under the time-varying user behavior and a new segmentation approach, which makes the storage granularity independent from the management granularity and can make a better use of the cache space are proposed.
Abstract: With the fast progresses of network technology, Video-On-Demand (VOD) service has found more and more applications. The transmission of multimedia files places heavy burdens on the Internet owing to their large sizes. To resolve this issue, caching servers are deployed at the edge of the Internet to meet most needs of local users by caching some popular videos. This paper provides an approach to choose the cached videos under the time-varying user behavior. Our approach estimates the average access intervals of a video with an Exponential Weighted Moving Average (EWMA) approach and furthermore predicts the video's future popularity based on its historical access intervals. The forgetting and predicting operations enable the algorithm to not only track the change of the time-varying user accesses, but also reduce the effects of the randomness of a single user access on the caching performance. In addition, we propose a new segmentation approach, which makes the storage granularity independent from the management granularity and can make a better use of the cache space. Simulation results show that our segmentation approach has a higher Byte-Hit Ratio than uniform segmentation and chunk segmentation, and our caching algorithm outperforms Least Recently Used (LRU), Least Frequently Used (LFU) and EWMA.
Citations
More filters
Journal Article
TL;DR: In this paper, the LFU-K page replacement algorithm is proposed, which is an improvement to the Least Frequently Used (LFU) algorithm for database disk buffering.
Abstract: This paper introduces a new approach to database disk buffering, called the LFU-K method. The LFU-K page replacement algorithm is an improvement to the Least Frequently Used (LFU) algorithm. The paper proposes a theoretical-probability model for formal description of LFU-K algorithm. Using this model we evaluate estimations for the LFU-K parameters. This paper also describes an implementation of LFU-2 policy. As we demonstrate by trace-driven simulation experiments, the LFU-2 algorithm provides significant improvement over conventional buffering algorithms for the shared-nothing database systems.

36 citations

Proceedings ArticleDOI
01 Jul 2017
TL;DR: Simulations confirm that the adaptive EWMA method significantly outperforms conventional popularity tracking methods and can achieve nearly the best performance without prior knowledge regarding the shifting speed of hot objects requested by users.
Abstract: The users' requests for online objects in mobile networks are characterized by fast varying dynamics caused by user mobility and diversity of users who produce contents. Conventional popularity methods cannot efficiently handle such strong dynamic requests of users, because they pay little attention to the time-varying user behavior which contributes to the growth and fading of online media files. To handle this issues, this paper proposes two schemes, including sensing the popularity variation speed of the hot contents and adaptively adjusting the popularity tracking speed of the cache algorithm. For the first one, we make use of the pseudo popularity acceleration in a time window to calculate the shifting speed of hot objects. With the obtained shifting speed can we adaptively adjust the popularity tracking speed of our previous EWMA method. Note that our adaptive EWMA method can achieve nearly the best performance without prior knowledge regarding the shifting speed of hot objects requested by users. Simulations confirm that our adaptive EWMA method significantly outperforms conventional popularity tracking methods.

9 citations


Cites methods from "An adaptive caching algorithm suita..."

  • ...Our pervious work [8] uses exponential weighted moving average method to achieve the goal of tracking the variation of popularity while depressing the influence of random access noise....

    [...]

  • ...We compare our popularity tracking method with LRU, LFU and the EWMA method in [8]....

    [...]

  • ...First we briefly introduce to our pervious work, the Exponential Weighted Moving Average (EWMA) popularity tracking method [8], which is the base of the adaptive popularity tracking method proposed in this paper....

    [...]

Journal ArticleDOI
TL;DR: An online algorithm that decides which file to remove from cache in order to allocate capacity to the newly-requested file is proposed and it is proved that for a cache that can store up to k files, the algorithm achieves a competitive ratio of $\mathcal {O}(\log (k)$ , which is the best competitive ratio achieved by any online algorithm as shown in the literature.
Abstract: In recent years, 5G cellular networks utilization has rapidly increased and is expected to grow even more in the near future This will put the current cellular networks operators in a challenge to overcome the network’s limits to satisfy the increasing mobile data traffic and the proliferation of user demands in deploying mobile applications The deployment of cache-enabled small base stations (Femtocells) is a promising solution to reduce the backhaul traffic loading and the file-access latency and therefore decrease the cellular network operational costs Due to the limited cache capacity when compared with the number of files that can be requested by users, in this paper, we formulate the problem of minimizing the cost paid by the cellular network while satisfying the cache capacity as an integer linear program (ILP) Due to the NP-completeness of the ILP formulation and the difficulty of obtaining the file request sequence apriori in real-life scenarios, we propose an online algorithm that decides which file to remove from cache in order to allocate capacity to the newly-requested file The algorithm works on a per-request basis and does not require the knowledge of the file request sequence in advance We prove that for a cache that can store up to $k$ files, the algorithm achieves a competitive ratio of $\mathcal {O}(\log (k))$ , which is the best competitive ratio achieved by any online algorithm as shown in the literature The simulations conducted considering a single cache show that while the proposed algorithm achieves a similar hit ratio compared with widely-used replacement schemes, it can reduce the cost of the cellular network by 25%

5 citations


Cites background or methods from "An adaptive caching algorithm suita..."

  • ...VI. EXPERIMENTAL PERFORMANCE ANALYSIS In this section, we compare four cache replacement schemes: the OCR algorithm described in Section 3, Least Recently Used (LRU), First In First Out (FIFO), and Forgetting History and Predicting Future (FHPF) [36]....

    [...]

  • ...The parameters for FHPF are directly taken from [36]....

    [...]

  • ...If we denote the costs of the OCR algorithm and FHPF for the g-th request sequence as COCR(g) and CFHPF (g) respectively, then the per demand cost savings is computed as ZOCR(g) = (1− COCR(g) CFHPF (g) ) × 100%....

    [...]

  • ...For each demand, we normalize the cost of the OCR algorithm, FIFO, and LRU with respect to the cost of FHPF, and then subtract it from 1....

    [...]

  • ...Moreover, compared to FHPF, our OCR algorithm VOLUME 6, 2018 41185 achieves at least 10% additional cost savings in 95% of the runs....

    [...]

Journal ArticleDOI
TL;DR: A comprehensive survey on QoS management for the video-on-demand systems focuses on load management and replication algorithms in content delivery networks and peer-to-peer (P2P) networks for their shortcomings and deduces that content availability is linked to the characteristics and the performance of the streaming protocols.
Abstract: The amount of online video content is exponentially increasing, which spurs its access demands. Providing optimal quality of service (QoS) for this ever-increasing video data is a challenging task due to the number of QoS constraints. The system resources, the distributed system platform and the transport protocol thus all need to collaborate to guarantee an acceptable level of QoS for the optimal video streaming process. In this paper, we present a comprehensive survey on QoS management for the video-on-demand systems. First, we focus on load management and replication algorithms in content delivery networks and peer-to-peer (P2P) networks for their shortcomings. We also address the problem of admission control and resource allocation with the objectives of congestion avoidance and frame-loss reduction. Besides, we introduce and discuss various replication schemes. For both the client–server architecture and P2P networks, we highlight the need for a specific storage management policy to preserve system reliability and content availability. We also focus on content distribution and streaming protocols scaling. We deduce that content availability is linked to the characteristics and the performance of the streaming protocols. Finally, we create a comparison table that presents the different contributions of the discussed approaches as well as their limitations. We believe that such a comprehensive survey provides useful insights and contributes to the related domains.

3 citations

Journal ArticleDOI
TL;DR: Simulation results illustrate that data required by clients could be better predicted and lower disk I/O could be achieved by using the model proposed in this paper.
Abstract: In contrast to other services on the Internet, streaming media service needs to fetch data from local disks more frequently, since it always lasts long and the bit rate is quite high. In addition, because of the much slower reading/writing speed of disk than random access memory (RAM), adopting advisable RAM caching policy can efficiently reduce disk I/O. In this paper, we study the problem of reducing disk I/O by using a novel approach. We first provide a new popularity estimate algorithm. Then a formal optimization problem about average disk I/O is presented, and a suboptimal caching algorithm for a special case of the problem is given. Furthermore, a partially observable Markov decision process (POMDP) model is constructed for the caching problem. Based on the model, popularity is taken advantage of to predict clients’ randomized behaviors, data replacing decisions are made when the defined observations occur, and the impact of caching actions on disk performance for future infinite steps is assessed. The method of event-based optimization is applied in search of the optimal stochastic policy. Disk I/O, as the long-run average performance measure, is optimized by applying the policy-gradient algorithm. The simulation results illustrate that data required by clients could be better predicted and lower disk I/O could be achieved by using the model proposed in this paper.

1 citations

References
More filters
Proceedings ArticleDOI
21 Mar 1999
TL;DR: This paper investigates the page request distribution seen by Web proxy caches using traces from a variety of sources and considers a simple model where the Web accesses are independent and the reference probability of the documents follows a Zipf-like distribution, suggesting that the various observed properties of hit-ratios and temporal locality are indeed inherent to Web accesse observed by proxies.
Abstract: This paper addresses two unresolved issues about Web caching. The first issue is whether Web requests from a fixed user community are distributed according to Zipf's (1929) law. The second issue relates to a number of studies on the characteristics of Web proxy traces, which have shown that the hit-ratios and temporal locality of the traces exhibit certain asymptotic properties that are uniform across the different sets of the traces. In particular, the question is whether these properties are inherent to Web accesses or whether they are simply an artifact of the traces. An answer to these unresolved issues will facilitate both Web cache resource planning and cache hierarchy design. We show that the answers to the two questions are related. We first investigate the page request distribution seen by Web proxy caches using traces from a variety of sources. We find that the distribution does not follow Zipf's law precisely, but instead follows a Zipf-like distribution with the exponent varying from trace to trace. Furthermore, we find that there is only (i) a weak correlation between the access frequency of a Web page and its size and (ii) a weak correlation between access frequency and its rate of change. We then consider a simple model where the Web accesses are independent and the reference probability of the documents follows a Zipf-like distribution. We find that the model yields asymptotic behaviour that are consistent with the experimental observations, suggesting that the various observed properties of hit-ratios and temporal locality are indeed inherent to Web accesses observed by proxies. Finally, we revisit Web cache replacement algorithms and show that the algorithm that is suggested by this simple model performs best on real trace data. The results indicate that while page requests do indeed reveal short-term correlations and other structures, a simple model for an independent request stream following a Zipf-like distribution is sufficient to capture certain asymptotic properties observed at Web proxies.

3,582 citations

Proceedings Article
08 Dec 1997
TL;DR: GreedyDual-Size as discussed by the authors incorporates locality with cost and size concerns in a simple and nonparameterized fashion for high performance, which can potentially improve the performance of main-memory caching of Web documents.
Abstract: Web caches can not only reduce network traffic and downloading latency, but can also affect the distribution of web traffic over the network through cost-aware caching. This paper introduces GreedyDual-Size, which incorporates locality with cost and size concerns in a simple and non-parameterized fashion for high performance. Trace-driven simulations show that with the appropriate cost definition, GreedyDual-Size outperforms existing web cache replacement algorithms in many aspects, including hit ratios, latency reduction and network cost reduction. In addition, GreedyDual-Size can potentially improve the performance of main-memory caching of Web documents.

1,048 citations


"An adaptive caching algorithm suita..." refers background in this paper

  • ...Greedy Dual-size (GDS) in [26] considers the effects of the object size and fetching cost on the popularity function....

    [...]

Proceedings ArticleDOI
18 Apr 2006
TL;DR: This study focuses on user behavior, content access patterns, and their implications on the design of multimedia streaming systems, and introduces a modified Poisson distribution that more accurately models the observations.
Abstract: Video-on-demand over IP (VOD) is one of the best-known examples of "next-generation" Internet applications cited as a goal by networking and multimedia researchers. Without empirical data, researchers have generally relied on simulated models to drive their design and developmental efforts. In this paper, we present one of the first measurement studies of a large VOD system, using data covering 219 days and more than 150,000 users in a VOD system deployed by China Telecom. Our study focuses on user behavior, content access patterns, and their implications on the design of multimedia streaming systems. Our results also show that when used to model the user-arrival rate, the traditional Poisson model is conservative and overestimates the probability of large arrival groups. We introduce a modified Poisson distribution that more accurately models our observations. We also observe a surprising result, that video session lengths has a weak inverse correlation with the video's popularity. Finally, we gain better understanding of the sources of video popularity through analysis of a number of internal and external factors.

728 citations


"An adaptive caching algorithm suita..." refers background in this paper

  • ...As shown in [2], some videos are very popular and accessed by many people while some others are barely accessed....

    [...]

  • ...For example, in [2] the popularity distribution curve has a long tail and lots of videos at the tail are requested only few times....

    [...]

  • ..., 90 % of the user sessions target for 14 % of the files in [2]....

    [...]

Proceedings ArticleDOI
21 Mar 1999
TL;DR: This work proposes a prefix caching technique whereby a proxy stores the initial frames of popular clips, and describes how to construct a smooth transmission schedule, based on the size of the prefix, smoothing, and playback buffers, without increasing client playback delay.
Abstract: High latency and loss rates in the Internet make it difficult to stream audio and video without introducing a large playback delay. To address these problems, we propose a prefix caching technique whereby a proxy stores the initial frames of popular clips. Upon receiving a request for the stream, the proxy initiates transmission to the client and simultaneously requests the remaining frames from the server. In addition to hiding the delay, throughput, and loss effects of a weaker service model between the server and the proxy, this novel yet simple prefix caching technique aids the proxy in performing workahead smoothing into the client playback buffer. By transmitting large frames in advance of each burst, workahead smoothing substantially reduces the peak and variability of the network resource requirements along the path from the proxy to the client. We describe how to construct a smooth transmission schedule, based on the size of the prefix, smoothing, and playback buffers, without increasing client playback delay. Experiments with MPEG traces show how a few megabytes of buffer space at the proxy can substantially reduce the bandwidth requirements of variable-bit-rate video. Drawing on these results, we present guidelines for allocating buffer space for each stream, and how to effectively share buffer and bandwidth resources among multiple clients and streams.

612 citations


Additional excerpts

  • ...Prefix-caching[9, 10] caches only the beginning frames of popular video clips to minimize the average initial delay....

    [...]

Proceedings ArticleDOI
01 Apr 1990
TL;DR: A replacement algorithm based on the concept of maintaining reference counts in which locality has been “factored out” is described, which can offer up to 34% performance improvement over LRU replacement.
Abstract: We propose a new frequency-based replacement algorithm for managing caches used for disk blocks by a file system, database management system, or disk control unit, which we refer to here as data caches. Previously, LRU replacement has usually been used for such caches. We describe a replacement algorithm based on the concept of maintaining reference counts in which locality has been “factored out”. In this algorithm replacement choices are made using a combination of reference frequency and block age. Simulation results based on traces of file system and I/O activity from actual systems show that this algorithm can offer up to 34% performance improvement over LRU replacement, where the improvement is expressed as the fraction of the performance gain achieved between LRU replacement and the theoretically optimal policy in which the reference string must be known in advance. Furthermore, the implementation complexity and efficiency of this algorithm is comparable to one using LRU replacement.

440 citations


"An adaptive caching algorithm suita..." refers methods in this paper

  • ...Some algorithms are proposed to improve LRU and LFU, such as LRU-K [22] and LFUaging[23]....

    [...]