File Caching Research Articles

Cache-Enabled Device-to-Device Communications: Offloading Gain and Energy Cost

By caching files at users, content delivery traffic can be offloaded via device-to-device (D2D) links if a helper user is willing to transmit the cached file to the user who requests the file. In practice, the user device has limited battery capacity, and may terminate the D2D connection when its battery has little energy left. Thus, taking the battery consumption allowed by the helper users to support D2D into account introduces a reduction in the possible amount of offloading. In this paper, we investigate the relationship between offloading gain of the system and energy cost of each helper user. To this end, we introduce a user-centric protocol to control the energy cost for a helper user to transmit the file. Then, we optimize the proactive caching policy to maximize the offloading opportunity and the transmit power at each helper to maximize the offloading probability. Finally, we evaluate the overall amount of traffic offloaded to D2D links and the average energy consumption at each helper, with the optimized caching policy and transmit power. Simulations show that a significant amount of traffic can be offloaded even when the energy cost is kept low.

Design and Realization of Energy-Efficient Hybrid Magneto-optical Filing System

Hierarchical hybrid filing system is a research hotspot in the field of data storage, and also is widely accepted in the industrial circle, which is widely used from personal storage to data center. This paper introduces an energy-efficient optical storage at a lower cost to solve the energy consumption problem of large data center, with an attempt to build a hybrid filing system comprised of array, online optical disk library and off-line optical disk library, discusses the guarantee mechanism of data consistency and the elastic design of system, and focuses on the research of the file caching and prefetching mechanism of hybrid magneto-optical filing system based on the read-write features of optical medium. Through experiments, the system is proved to be capable of effectively saving energy and maintaining the timeliness of data retrieval and query.

A Method of Shared File Cache for File Clone Function to Improve I/O Performance for Virtual Machines

SUMMARYWe propose a method for shared file cache function for cloned files used by virtual machines, called SCC. The file clone function copies the files faster than conventional (read and write) method. Moreover, the function reduces disk spaces. The function is used to deploy virtual machines in virtual desktop infrastructure because of fast copying a lot of virtual machine disk files. SCC uses the file cache of a shared file as a shared cache among cloned files. The cached data on the shared file are returned to application programs on accessing the shared file via cloned files. Therefore, SCC improves the I/O performance of the shared file due to avoiding disk accesses. In this paper, we implement SCC and evaluate the I/O performance. From the evaluation, we have found SCC improves I/O throughput about 38 times in the case of random read and shared cache hit.

Robust fully distributed file caching for delay-tolerant networks: A reward-based incentive mechanism

This article exhibits a reward-based incentive mechanism for file caching in delay-tolerant networks. In delay-tolerant networks, nodes use relay’s store-carry-forward paradigm to reach the final destination. Thereby, relay nodes may store data in their buffer and carry it till an appropriate contact opportunity with destination arises. However, the relays are not always willing to assist data forwarding due to a limited energy or a low storage capacity. Our proposal suggests a reward mechanism to uphold and to sustain cooperation among relay nodes. We model this distributed network interaction as a non-cooperative game. Namely, the source node offers to the relay nodes a positive reward if they accept to cache and to forward a given file successfully to a target destination, whereas the relay nodes may either accept or reject the source deal, depending on the reward attractiveness and on their battery status (their actual energy level). Next, full characterizations of both pure and mixed Nash equilibria are provided. Then, we propose three fully distributed algorithms to ensure convergence to the Nash equilibria (for both pure equilibrium and mixed equilibrium). Finally, we validate our proposal through extensive numerical examples and many learning simulations and draw some conclusions and insightful remarks.

FD device‐to‐device communication for wireless video distribution

Spectrum scarcity and dramatically increasing demand for high data rate and high-quality video live streaming are of future cellular network design challenges. As a solution to this problem, cache-enabled cellular network architecture has been recently proposed. Device-to-device (D2D) communications can be exploited for distributed video content delivery, and devices can be used for caching of the video files. This can increase the capacity and reduce the end-to-end delay in cellular networks. In this study, the authors propose a new scheme for video distribution over cellular networks by exploiting full-duplex (FD) radios for D2D devices in two scenarios: (i) two nodes exchange their desired video files simultaneously and (ii) each node can concurrently transmit to and receive from two different nodes. In the latter case, an intermediate transceiver can serve one or multiple users’ file request(s) whilst capturing its desired file from another device in the vicinity. Mathematical expressions along with extensive simulations are used to compare their proposed scheme with a half-duplex scheme to show the achievable gains in terms of sum throughput, active links, and delay. They will also look into the energy cost for achieving the improvements provided by operation in FD mode.

Optimal Caching and Scheduling for Cache-Enabled D2D Communications

To maximize offloading gain of cache-enabled device-to-device (D2D) communications, content placement and delivery should be jointly designed. In this letter, we jointly optimize caching and scheduling policies to maximize successful offloading probability, defined as the probability that a user can obtain desired file in local cache or via D2D link with data rate larger than a given threshold. We obtain the optimal scheduling factor for a random scheduling policy that can control interference in a distributed manner, and a low complexity solution to compute caching distribution. We show that the offloading gain can be remarkably improved by the joint optimization.

Transparently Exploiting Device-Reserved Memory for Application Performance in Mobile Systems

Most embedded systems require contiguous memory space to be reserved for devices, which may lead to memory under-utilization. Although several approaches have been proposed to address this issue, they have limitations of either inefficient memory usage or long latency for switching the reserved memory space between a device and general-purpose uses. Our scheme, on the other hand, utilizes reserved memory as an eviction-based file cache. It guarantees contiguous memory allocation to devices while providing idle device memory as an additional file cache called eCache for general-purpose usage. Because eCache stores only evicted data from the in-kernel page cache, the memory efficiency is preserved and the allocation time for devices is minimized. Cost-based region selection also minimizes additional read I/O operations by carefully discarding cached data from eCache. The additional indexing cost incurred by adding eCache is minimized by integrating its index structure with the kernel page cache. The prototype is implemented on the Nexus S smartphone and is evaluated using popular Android applications. The evaluation results show that our scheme outperforms previous approaches in terms of the application launch performance. The device memory reallocation time is also limited to a few milliseconds, which is sufficiently small to make our scheme transparent.

Performance Improvement in Ontology based Semantic Web using Multi-level Cache

Objective: The objective of the paper is to improve the performance of semantic web in a Hadoop environment. An analysis of the ontology and the improvement in the role of cache played a major part in the performance improvement. Methods: Ontology describes a formal specification of certain domain. Existing platform and servers are unable to process different types of data. Hence, it is better to have an effective large number of ontology in the web. Hadoop platform helps in implementing the incremental and distributed ontologies over and above existing ontologies in Semantic Web. One of the major challenges recorded while deploying the semantic technologies is the performances degradation of triple stores. Findings: A framework is proposed in order to improve the performance of Web applications which is relational databasebacked. When implemented effectively, it provides a strong base to semantic computations. Application: We all know that inherent speed difference between disk and the processor can be reduced by File caching mechanism and hence Multilevel cache helped in improving the performance of semantic web.

Dynamic Process Migration Based on Block Access Patterns Occurring in Storage Servers

An emerging trend in developing large and complex applications on today’s high-performance computers is to couple independent components into a comprehensive application. The components may employ the global file system to exchange their data when executing the application. In order to reduce the time required for input/output (I/O) data exchange and data transfer in the coupled systems or other applications, this article proposes a dynamic process migration mechanism on the basis of block access pattern similarity for utilizing the local file cache to exchange the data. We first introduce the scheme of the block access counting diagram to profile the process access pattern during a time period on the storage server. Next, we propose an algorithm that compares the access patterns of processes running on different computing nodes. Last, processes are migrated in order to group processes with similar access patterns. Consequently, the processes on the computing node can exchange their data by accessing the local file cache, instead of the global file system. The experimental results show that the proposed process migration mechanism can reduce the execution time required by the application because of the shorter I/O time, as well as yield attractive I/O throughput. In summary, this dynamic process migration technique can work fairly well for distributed applications whose data dependency rely on distributed file systems.

Wireless Content Caching for Small Cell and D2D Networks

The fifth generation wireless networks must provide fast and reliable connectivity while coping with the ongoing traffic growth. It is of paramount importance that the required resources, such as energy and bandwidth, do not scale with traffic. While the aggregate network traffic is growing at an unprecedented rate, users tend to request the same popular contents at different time instants. Therefore, caching the most popular contents at the network edge is a promising solution to reduce the traffic and the energy consumption over the backhaul links. In this paper, two scenarios are considered, where caching is performed either at a small base station, or directly at the user terminals, which communicate using \ac{D2D} communications. In both scenarios, joint design of the transmission and caching policies is studied when the user demands are known in advance. This joint design offers two different caching gains, namely, the \textit{pre-downloading} and \textit{local caching gains}. It is shown that the finite cache capacity limits the attainable gains, and creates an inherent tradeoff between the two types of gains. In this context, a continuous time optimization problem is formulated to determine the optimal transmission and caching policies that minimize a generic cost function, such as energy, bandwidth, or throughput. The jointly optimal solution is obtained by demonstrating that caching files at a constant rate is optimal, which allows to reformulate the problem as a finite-dimensional convex program. The numerical results show that the proposed joint transmission and caching policy dramatically reduces the total cost, which is particularised to the total energy consumption at the \ac{MBS}, as well as to the total economical cost for the service provider, when users demand economical incentives for delivering content to other users over the D2D links.

HDCache: A Distributed Cache System for Real-Time Cloud Services

Providing a real-time cloud service requires simultaneously retrieving a large amount of data. How to improve the performance of file access becomes a great challenge. This paper first addresses the preconditions of dealing with this problem considering the requirements of applications, hardware, software, and network environments in the cloud. Then, a novel distributed layered cache system named HDCache is proposed. HDCahe is built on the top of Hadoop Distributed File System (HDFS). Applications can integrate the client library of HDCache to access the multiple cache services. The cache services are built up with three access layers an in-memory cache, a snapshot of the local disk, and a network disk provided by HDFS. The files loaded from HDFS are cached in a shared memory which can be directly accessed by the client library. In order to improve robustness and alleviate workload, the cache services are organized in a peer-to-peer style using a distributed hash table and every cached file has three replicas scattered in different cache service nodes. Experimental results show that HDCache can store files with a wide range in their sizes and has the access performance in a millisecond level under highly concurrent environments. The tested hit ratio obtained from a real-world cloud serviced is higher than 95 %.

Implementation of a deduplication cache mechanism using content-defined chunking

Many application programs in data-intensive science read and write large files. Large data consume significant memory because the data is loaded into the page cache. Since memory resources are critically valuable in data-intensive computing, reducing the memory footprint consumed by file data is essential. In this paper, we propose a cache deduplication mechanism with content-defined chunking CDC for the Gfarm distributed file system. CDC divides a file into variable-size blocks chunks based on the contents of the file. The client stores the chunks in the local file system as cache files and reuses them during subsequent file accesses. Deduplication of chunks reduces the amount of transmitted data between clients and servers, and reduces storage and memory requirements. The experimental results demonstrate that the proposed mechanism significantly improves the performance of file-read operations and that the introduction of parallelism reduces the overhead of file-write operations.

Block-Precise Processors: Low-Power Processors with Reduced Operand Store Accesses and Result Broadcasts

Power is a first order design constraint for most processors today. Benefits of low power designs include lower manufacturing and operating costs and a longer battery life. In this work we propose an out of order processor architecture called Block-precise processor (B-Processor) that is designed for low power consumption. The B-Processor consumes lower power than typical processor designs by eliding the write of results of many instructions to the reorder buffer and to the register file, which are power hungry structures. The B-Processor reduces power consumption even further by omitting the broadcast of certain results over multiple levels of the bypass network. Experimental results show that on average the B-Processor spends 15.1 percent less power on register file and reorder buffer accesses and 14.5 percent less power on broadcasting results. In combination with register file caching, on average the B-Processor saves 28.7 percent power for accessing the register file and the reorder buffer.

데이터 망각을 활용한 비휘발성 메모리 기반 파일 캐시 관리 기법

Non-volatile memory (NVM) supports both byte addressability and non-volatility. These characteristics make it feasible for NVM to be employed at any layer of the memory hierarchy such as cache, memory and disk. An interesting characteristic of NVM is that, even though it supports non-volatility, its retention capability is limited. Furthermore NVM has tradeoff between its retention capability and write latency. In this paper, we propose a novel NVM-based file cache management scheme that makes use of the limited retention capability to improve the cache performance. Experimental results with real-workloads show that our scheme can reduce access latency by up to 31% (24.4% average) compared with the conventional LRU based cache management scheme.

Mining-based File Caching in a Hybrid Storage System

In this work, we propose a new mining-based file caching scheme for a hybrid storage disk system. In particular, we focus our efforts on reducing the latency of launching applications. The proposed scheme identifies correlated file accesses in a file access sequence via sequential pattern mining algorithm. Our scheme caches correlated files together to maximize the caching efficiency. The correlated files are extracted from the access patterns through the proposed mining scheme, which consists of three steps: frequent pattern based file extraction, cluster moving gap based file sort, and frequency and size based file prioritization. The extracted correlated files are relocated to an SSD during idle time. DiskSim and NANDSim are used to evaluate the proposed scheme, called Informed Mining. The proposed scheme is compared with a disk only scheme and five other mining based file relocation schemes: Mining based file relocation scheme (Miner), minimum distance based file relocation scheme (Min_Dist), frequency-based relocation scheme (Fre), size-based relocation scheme (Size), and one that relocates files with highest value of (file size * file access number) first to the SSD (Fr*Sz). From the simulation based experiment, launch time is reduced by about 50% using only 10% of sum of all file sizes accessed during a launch of an application.

A Reliable File Protection System Based on Transparent Encryption

As data leakage moves to ever more challenging areas, improving the security level of anti-data leakage and lowering overhead to the operating system becomes increasingly important. Therefore, the paper presents a novel double cache file filter driver based on transparent encryption, called as DBFD. In order to boost security of data, that is controlled by the file filter driver. The DBFD overcomes the limitation of double cache in the file system kernel in windows operating system and used the transparent encryption method to protect the data security. To evaluate DBFD, we used Iometer as the measurement tools to measure performance. The simulation results indicate that proposed DBFD has higher security, less overhead to the windows operating system.

Distributed caching in unstructured peer-to-peer file sharing networks

Nowadays, the peer-to-peer (P2P) system is one of the largest Internet bandwidth consumers. To relieve the burden on Internet backbone and improve the query and retrieve performance of P2P file sharing networks, efficient P2P caching algorithms are of great importance. In this paper, we propose a distributed topology-aware unstructured P2P file caching infrastructure and design novel placement and replacement algorithms to achieve optimal performance. In our system, for each file, an adequate number of copies are generated and disseminated at topologically distant locations. Unlike general believes, our caching decisions are in favor of less popular files. Combined with the underlying topology-aware infrastructure, our strategy retains excellent performance for popular objects while greatly improves the caching performance for less popular files. Overall, our solution can reduce P2P traffic on Internet backbone, and relieve the over-caching problem that has not been properly addressed in unstructured P2P networks. We carry out simulation experiments to compare our approaches with several traditional caching strategies. The results show that our algorithms can achieve better query hit rates, smaller query delay, higher cache hit rates, and lower communication overhead.

A New Approach to Coding in Content-Based MANETs

In content-based mobile ad hoc networks (CB-MANETs), random linear network coding (NC) can be used to reliably disseminate large files under intermittent connectivity. Conventional NC involves random unrestricted coding at intermediate nodes. This however is vulnerable to pollution attacks. To avoid attacks, a brute force approach is to restrict the mixing at the source. However, source restricted NC generally reduces the robustness of the code in the face of errors, losses and mobility induced intermittence. CB-MANETs introduce a new option. Caching is common in CB MANETs and a fully reassembled cached file can be viewed as a new source. Thus, NC packets can be mixed at all sources (including the originator and the intermediate caches) yet still providing protection from pollution. The hypothesis we wish to test in this paper is whether in CB-MANETs with sufficient caches of a file, the performance (in terms of robustness) of the restricted coding equals that of unrestricted coding. In this paper, we examine and compare unrestricted coding to full cache coding, source only coding, and no coding. As expected, we find that full cache coding remains competitive with unrestricted coding while maintaining full protection against pollution attacks.

QoE-Driven Cache Management for HTTP Adaptive Bit Rate Streaming Over Wireless Networks

In this paper, we investigate the problem of optimal content cache management for HTTP adaptive bit rate (ABR) streaming over wireless networks. Specifically, in the media cloud, each content is transcoded into a set of media files with diverse playback rates, and appropriate files will be dynamically chosen in response to channel conditions and screen forms. Our design objective is to maximize the quality of experience (QoE) of an individual content for the end users, under a limited storage budget. Deriving a logarithmic QoE model from our experimental results, we formulate the individual content cache management for HTTP ABR streaming over wireless network as a constrained convex optimization problem. We adopt a two-step process to solve the snapshot problem. First, using the Lagrange multiplier method, we obtain the numerical solution of the set of playback rates for a fixed number of cache copies and characterize the optimal solution analytically. Our investigation reveals a fundamental phase change in the optimal solution as the number of cached files increases. Second, we develop three alternative search algorithms to find the optimal number of cached files, and compare their scalability under average and worst complexity metrics. Our numerical results suggest that, under optimal cache schemes, the maximum QoE measurement, i.e., mean-opinion-score (MOS), is a concave function of the allowable storage size. Our cache management can provide high expected QoE with low complexity, shedding light on the design of HTTP ABR streaming services over wireless networks.

Memory reservation and shared page management for real-time systems

Memory reservations are used to provide real-time tasks with guaranteed memory access to a specified amount of physical memory. However, previous work on memory reservation primarily focused on private pages, and did not pay attention to shared pages, which are widely used in current operating systems. With previous schemes, a real-time task may experience unexpected timing delays from other tasks through shared pages that are shared by another process, even though the task has enough free pages in its own reservation. In this paper, we first describe the problems that arise when real-time tasks share pages. We then propose a shared-page management framework which enhances the temporal isolation provided by memory reservations in resource kernels that use the resource reservation approach. Our proposed solution consists of two schemes, Shared-Page Conservation (SPC) and Shared-Page Eviction Lock (SPEL), each of which prevents timing penalties caused by the seemingly arbitrary eviction of shared pages. The framework can manage shared data for inter-process communication and shared libraries, as well as pages shared by the kernel's copy-on-write technique and file caches. We have implemented and evaluated our schemes on the Linux/RK platform, but it can also be applied to other operating systems with paged virtual memory.