Make Caching Decisions Research Articles

DAG-aware harmonizing job scheduling and data caching for disaggregated analytics frameworks

Modern data analytics frameworks often integrate with external storage services, which can lead to storage bottlenecks. Existing caching and prefetching solutions utilize high-level information from data analytics frameworks to forecast future data accesses. They employ these predictions to prefetch data into the cache and manage the cache contents. However, this approach overlooks a fundamental opportunity: rather than caching data given a prediction of job execution, influencing the job execution order can enable more effective caching and prefetching. With this key insight, we introduce a novel system called Tripod, designed to synchronize job scheduling and data caching for analytics frameworks.Leveraging the higher-level information provided by analytics frameworks, Tripod explores the best-suited job execution order, guided by developed heuristics, to facilitate prefetching and caching. To fully exploit the potential of Tripod, we also introduce a novel caching strategy named CAP. This strategy not only acknowledges the job scheduling order but also offers fine-grained control over object prefetching and eviction. Our evaluation, conducted using standard analytic benchmarks (TPC-H and TPC-DS), demonstrates that Tripod achieves a speedup of up to 1.7x on state-of-the-art approaches. Moreover, when employing CAP to make caching decisions, the performance can further be improved (as much as 1.5x).

Intelligent Caching with Graph Neural Network-Based Deep Reinforcement Learning on SDN-Based ICN

Information-centric networking (ICN) has gained significant attention due to its in-network caching and named-based routing capabilities. Caching plays a crucial role in managing the increasing network traffic and improving the content delivery efficiency. However, caching faces challenges as routers have limited cache space while the network hosts tens of thousands of items. This paper focuses on enhancing the cache performance by maximizing the cache hit ratio in the context of software-defined networking–ICN (SDN-ICN). We propose a statistical model that generates users’ content preferences, incorporating key elements observed in real-world scenarios. Furthermore, we introduce a graph neural network–double deep Q-network (GNN-DDQN) agent to make caching decisions for each node based on the user request history. Simulation results demonstrate that our caching strategy achieves a cache hit ratio 34.42% higher than the state-of-the-art policy. We also establish the robustness of our approach, consistently outperforming various benchmark strategies.

Caching Scalable Videos in the Edge of Wireless Cellular Networks

By pre-fetching popular videos into the local caches of edge nodes, wireless edge caching provides an effective means of reducing repeated content deliveries. To meet the various viewing quality requirements of multimedia users, scalable video coding (SVC) is integrated with edge caching, where the constituent layers of scalable videos are flexibly cached and transmitted to users. In this article, we discuss the challenges arising from the different content popularity and various viewing requirements of scalable videos, and present the diverse types of cached contents as well as the corresponding transmission schemes. We provide an overview of the existing caching schemes, and summarize the criteria of making caching decisions. A case study is then presented, where the transmission delay is quantified and used as the performance metric. Simulation results confirm that giving cognizance to the realistic requirements of end users is capable of significantly reducing the content transmission delay, compared to the existing caching schemes operating without SVC. The results also verify that the transmission delay of the proposed random caching scheme is lower than that of the caching scheme which only provides local caching gain.

AoI-Based Temporal Attention Graph Neural Network for Popularity Prediction and Content Caching

Along with the fast development of network technology and the rapid growth of network equipment, the data throughput is sharply increasing. To handle the problem of backhaul bottleneck in cellular network and satisfy people’s requirements about latency, the network architecture, like the information-centric network (ICN) intends to proactively keep limited popular content at the edge of network based on predicted results. Meanwhile, the interactions between the content (e.g., deep neural network models, Wikipedia-alike knowledge base) and users could be regarded as a dynamic bipartite graph. In this paper, to maximize the cache hit rate, we leverage an effective dynamic graph neural network (DGNN) to jointly learn the structural and temporal patterns embedded in the bipartite graph. Furthermore, in order to have deeper insights into the dynamics within the evolving graph, we propose an age of information (AoI) based attention mechanism to extract valuable historical information while avoiding the problem of message staleness. Combining this aforementioned prediction model, we also develop a cache selection algorithm to make caching decisions in accordance with the prediction results. Extensive results demonstrate that our model can obtain a higher prediction accuracy than other state-of-the-art schemes in two real-world datasets. The results of hit rate further verify the superiority of the caching policy based on our proposed model over other traditional ways.

ICache: An Intelligent Caching Scheme for Dynamic Network Environments in ICN-Based IoT Networks

Advanced network technologies and ubiquitous connected devices are boosting the development of the Internet of Things (IoT) at an unprecedented pace. However, as most of the connected IoT devices are battery powered, the energy consumption issue has become the bottleneck of the IoT’s development. Caching is a promising approach to reducing the energy consumption of the battery-powered devices since the requested data packets can be retrieved from intermediate nodes in the network, e.g., routers, instead of from the remote battery-powered IoT devices, which allows the IoT devices to spend more time in the sleep mode. To realize in-network caching and overcome the IP-based networks’ inefficiency support for IoT, building IoT over information-centric networking (ICN) is a promising approach advocated by researchers. However, existing works in this area assume the network environments are static, which hinders the development of existing approaches in the real dynamic network environments. In this article, we leverage the deep <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula> -networks (DQNs) to propose an intelligent caching scheme (named as iCache) that can automatically adjust the caching nodes’ caching parameters to make caching decisions for the dynamic network environments. Extensive evaluations were conducted and the results show that the proposed iCache outperforms the existing approaches in terms of the total energy consumption (e.g., more than 29% reduction compared to the caching transient data (CTD) caching scheme) and the average number of hops (e.g., more than 20% reduction compared to the CTD caching scheme).

Time Segmentation-Based Hybrid Caching in 5G-ICN Bearer Network

The fifth-generation communication technology (5G) and information-centric networks (ICNs) are acquiring more and more attention. Cache plays a significant part in the 5G-ICN architecture that the industry has suggested. 5G mobile terminals switch between different base stations quickly, creating a significant amount of traffic and a significant amount of network latency. This brings great challenges to 5G-ICN mobile cache. It appears urgent to improve the cache placement strategy. This paper suggests a hybrid caching strategy called time segmentation-based hybrid caching (TSBC) strategy, based on the 5G-ICN bearer network infrastructure. A base station’s access frequency can change throughout the course of the day due to the “tidal phenomena” of mobile networks. To distinguish the access frequency, we split each day into periods of high and low liquidity. To maintain the diversity of cache copies during periods of high liquidity, we replace the path’s least-used cache copy. We determine the cache value of each node in the path and make caching decisions during periods of low liquidity to make sure users can access the content they are most interested in quickly. The simulation results demonstrate that the proposed strategy has a positive impact on both latency and the cache hit ratio.

Heterogeneous Edge Caching Based on Actor-Critic Learning With Attention Mechanism Aiding

In recent years, the explosive growth of network traffic has placed significant strain on backbone networks. To alleviate the content access delay and reduce additional network resource consumption resulting from large-scale requests, edge caching has emerged as a promising technology. Despite capturing substantial attention from both academia and industry, most existing studies overlook the heterogeneity of the environment and the spatial-temporal characteristics of content popularity. As a result, the potential for edge caching remains largely unexploited. To address these challenges, we propose a neighborhood-aware caching (NAC) framework in this paper. The framework leverages the perimeter information from neighboring base stations (BSs) to model the edge caching problem in heterogeneous scenarios as a Markov Decision Process (MDP). To fully exploit the environmental information, we introduce an improved actor-critic method that integrates an attention mechanism into the neural network. The actor-network in our framework is responsible for making caching decisions based on local information, while the critic network evaluates and enhances the actor's performance. The multi-head attention layer in the critic network enables integration of environmental features into the model, reducing the limitations associated with local investigation. To facilitate comparison from an engineering perspective, we also propose a heuristic algorithm, Neighbor-Influence-Least-Frequently-Use (NILFU). Our extensive experiments demonstrate that the proposed NAC framework outperforms other baseline methods in terms of average delay, hit rate, and traffic offload ratio in heterogeneous scenarios. This highlights the effectiveness of the neighborhood-aware caching approach in enhancing the performance of edge caching systems in such scenarios.

Multimodal semantic communication accelerated bidirectional caching for 6G MEC

Mobile Edge Computing (MEC) enables immersive XR with multimodal data by coordinating communication, computation, and caching (3C) resources upcoming 6G. The traditional communication constrained by Shannon’s theorem cannot accommodate the user demands for ultra-reliability and high throughput. In contrast, the semantic communication improves the quality of service and user experience by exploiting the semantic features. This paper constructs a multi-user MEC structure based on multimodal semantic communication for interactive AR/VR games. We construct a bidirectional caching task model to achieve cache-enhanced computing. To minimize the system cost, including the user latency, energy consumption, and storage size, we propose a content popularity-based DQN (CP-DQN) algorithm to make caching decisions. Then the CP-DQN is extended to the cache–computation coordination optimization algorithm (CCCA) to achieve the 3C resources tradeoff. Simulation results demonstrate that the proposed algorithm outperforms existing algorithms in terms of caching hit ratio, computation cost and edge resource utilization.

A Privacy-Protected Popularity Prediction Scheme for Content Caching Based on Federated Learning

With the explosive growth of mobile data in wireless communication, popular contents need to be brought closer to the edge for rapid data access. Recent studies have proposed methods of content caching based on machine learning but neglected the leakage of private data. To solve this problem, we propose a Private Federated Learning-based Caching (PFLC) scheme. Benefiting from federated learning framework and the pseudo rating matrix (PRM), the scheme can collect the statistical characteristics of a user group to make caching decisions by predicting the popularity of contents without revealing the privacy of users, which means that it can protect the privacy of individual users from being accessed by the server and other users. The performance evaluation demonstrates that the proposed scheme effectively protects individual users’ privacy without sacrificing the cache hit ratio.

Canada jays (Perisoreus canadensis) identify and exploit coniferous cache locations using visual cues

AbstractFood caching is a foraging strategy used by many vertebrates, involving the storage and subsequent recovery of food items for later consumption, when other food sources are scarce. Once cached, stored food, particularly highly perishable items, can degrade over time. Evidence suggests that for birds, some conifers may aid in cache preservation through protective properties in resin. However, due to the challenges involved with following birds to their caching locations, cache‐site preferences are not easily studied in the wild. We investigated eight captive Canada jays’ (Perisoreus canadensis) ability to both identify and exploit conifer tree species. Further, we examined potential cues that birds may use to identify and select these potentially beneficial sites. We found strong evidence to suggest that birds can quickly identify conifer tree species and subsequently exploit those cache locations preferentially. Furthermore, our evidence suggests that although birds do not appear to use olfactory cues when making caching decisions, they potentially to attend to structural cues. We suggest that visual information is essential to both the identification of conifer trees and to cache‐site selection decisions. These findings indicate that jays make rapid, fine scale assessments of their environments, discriminating amongst trees of different species and use this information to select cache‐sites.

Caching scheme for information‐centric networks with balanced content distribution

SummaryInformation‐centric network (ICN) emphasizes on content retrieval without much bothering about the location of its actual producer. This novel networking paradigm makes content retrieval faster and less expensive by shifting data provisioning into content holder rather than content owner. Caching is the feature of ICN that makes content serving possible from any intermediate device. An efficient caching is one of the primary requirements for effective deployment of ICN. In this paper, a caching approach with balanced content distribution among network devices is proposed. The selection of contents to be cached is determined through universal and computed using Zipf's law. The dynamic change in popularity of contents is also considered to take make caching decisions. For balancing the cached content across the network, every router keeps track of its neighbor's cache status. Three parameters, the proportionate distance of the router from the client (pd), the router congestion (rc), and the cache status (cs), are contemplated to select a router for caching contents. The new caching approach is evaluated in the simulated environment using ndnSIM‐2.0. Three state‐of‐the‐art approaches, Leave Copy Everywhere (LCE), centrality measures‐based algorithm (CMBA), and a probability‐based caching (probCache), are considered for comparison. The proposed method of caching shows the better performance compared to the other three protocols used in the comparison.

An Energy Reward-Based Caching Mechanism for Information-Centric Internet of Things.

Existing information-centric networking (ICN) designs for Internet of Things (IoT) mostly make caching decisions based on probability or content popularity. From the energy-efficient perspective, those strategies may not always be energy efficient in resource-constrained IoT because without considering the energy reward of caching decisions, inappropriate routers and content objects may be selected for caching, which may lead to negative energy rewards. In this paper, we analyze the energy consumption of content caching and content retrieval in resource-constrained IoT and calculate caching energy reward as a key metric to measure the energy efficiency of a caching decision. We then propose an efficient cache placement and cache replacement mechanism based on the caching energy reward to improve the energy efficiency of caching decisions. Through analysis and experimental results, we show that the proposed mechanism achieves a significant improvement in terms of energy efficiency, stretch ratio, and cache hit ratio compared to state-of-the-art caching schemes.

Adapting Deep Learning for Content Caching Frameworks in Device-to-Device Environments

Recently, we have witnessed an expeditious growth in the intelligence and processing ability of user equipment accompanied by the explosive increase of wireless data and traffic. With such a huge demand, and a shortage of resources to fulfill that need, device-to-device (D2D) communication has surfaced as a propitious solution to reduce costs associated with backhaul links by collaboratively caching files. This paper explores the existing content caching frameworks in D2D communication environments. Further, it proposes two novel frameworks -one leveraging a combination of recurrent and deep neural networks and another based on the latest deep language models that use attention mechanisms known as transformers. The developed frameworks require minimum apriori knowledge about the environment and utilize the growth of user data to achieve performance enhancement. Our experiments show that the proposed frameworks are adaptive in nature and learn from the historical data of the environment. Further, we achieve an overall 25% increase in the D2D cache hit rate over a recently proposed framework that uses neural networks for collaborative filtering (NCF) to make caching decisions.

Providing Cache Consistency Guarantee for ICN-Based IoT Based on Push Mechanism

This paper develops POPCache (POpularity-aware Push-based Cache), a cache management strategy to support cache consistency in ICN-based IoT. Based on the proposed push mechanism, POPCache can proactively update the slated content to impose strong consistency while achieving low overhead, because POPCache relieves the heavy burden of nodes (which can be resource-constrained IoT devices) of maintaining extra information. POPCache makes caching decisions by jointly considering several determining factors. We use the theoretical study to analyze POPCache’s overhead. Moreover, we conduct simulations to compare POPCache with existing works; results show that POPCache improves cache consistency and achieves good cache and network performance.

On-path caching based on content relevance in Information-Centric Networking

The Internet of things (IoT) is a new network concept. Based on the Internet, the idea of connecting everything is proposed, which is likely to become the development direction of computer and communication in the future. Meanwhile, as a disruptive new communication network model, Information-Centric Network (ICN) has become a hot spot in the field of future network architecture research in recent years. ICN is information-centric, and uses the name of the information to implement data identification, retrieval, and routing and forwarding. The information centric network uses the cache as a built-in structure, and nodes store all the data that flows by default, so that subsequent requests can be responded to as soon as possible. In the actual network, the spread and trend of an information constantly change with time, and the popularity of the information is different in different time periods. However, the existing ICN native caching mechanism ignores the correlation between contents, and the existing relevant research makes insufficient use of content relevance. In this paper may exist a certain correlation between nodes have access to the content of the facts, we design a path cache method based on the correlation content, by discovering target content and the correlation between nodes store content, at the same time considering the node position in the path, make caching decisions, make the cache memory more efficient. The feasibility and effectiveness of the proposed method are verified by simulation experiments under different parameters.

A Privacy-Preserving Caching Scheme for Device-to-Device Communications

With device-to-device (D2D) communication, user equipment can share data with each other without the involvement of network infrastructures. In order to maintain the Quality of Service (QoS) and Quality of Experience (QoE) for user applications in D2D communications, most existing schemes use proactive content caching that needs to predict content popularity before making caching decisions which may result in privacy leakage, since the information of users is collected to train a deep learning-based model to predict content popularity. Therefore, it is crucial to guarantee secure data collection in machine learning-based framework. In this paper, we propose a privacy-preserving D2D caching scheme with a passive content caching strategy based on node importance, which can deliver more efficient caching and prevent the potential leakage of user privacy. The scheme is based on softwaredefined networking (SDN), in which the controller is responsible for calculating node importance of devices according to the information of requests and encounters collected by SDN switches. Base station will decide which device can establish reliable and secure communication with content requester based on historical information. The simulation results show that the proposed strategy can outperform other D2D caching strategies in terms of cache hit rate and data rate.

Caching Policies for Transient Data

This work focuses on designing caching policies for transient data, i.e., data which can be used to serve requests only for a finite duration of time after which it becomes redundant. We first characterize the fundamental limit on the performance of caching policies for transient data and characterize the performance of traditional caching policies like LRU for this setting. Traditional caching policies often make decisions based on the popularity of the data being cached. We propose a new caching policy which uses both the popularity and the residual life-time (time remaining before the data becomes redundant) to make caching decisions. We show that in the setting where data being cached is transient, our policy outperforms traditional caching policies.

Prediction Based Vehicular Caching: Where and What to Cache?

The explosive growth of user demands for mobile data puts forward higher requirements for existing network architecture. In this case, many novel technologies, such as edge computing, ultra-dense network (UND), are recently proposed to alleviate the network burden of cellular macro base stations (MBSs). However, their performance is limited by dynamic user requests and uncertain user mobility since they highly rely on the fixed infrastructure. Along with the advent of autonomous driving era, autonomous vehicles possess strong computing and communication ability, which provide a new idea to resolve the limitation of infrastructure based network architecture. In this paper, we exploit the performance of a vehicular caching scheme where the role of moving vehicles is changed from service consumers in traditional networks to service providers and service consumers. As such, a prediction based vehicular caching scheme is proposed. Specifically, an optimization problem is firstly formulated by exploring the relationship of caching vehicles and mobile users, in order to optimize the network energy efficiency. The nonconvex optimization problem is solved by decomposing it into a nonlinear programming problem. By applying the Lyapunov method and autoregressive neural network (ANN), an online caching decision algorithm is finally proposed to make caching decisions. Extensive simulations are conducted to evaluate the caching scheme in different scenarios. Results show that the vehicular caching scheme can obviously improve network energy efficiency with different requests, but the increment is reduced under the scenario of intensive requests. The comparison between online and offline caching also shows the necessity of online caching decision making due to its benefit in resource utilization where system gain is increased from 8.4% to 59.24%.

IBP Based Caching Strategy in D2D

Device to Device (D2D) communication is a key technology in 5th generation wireless systems to increase communication capacity and spectral efficiency. Applying caching into D2D communication networks, the device can retrieve content from other devices by establishing D2D communication links. In this way, the backhaul traffic can be significantly reduced. However, most of the existing caching schemes in D2D are proactive caching, which cannot satisfy the requirement of real-time updating. In this paper, we propose an Indian Buffet Process based D2D caching strategy (IBPSC). Firstly, we construct a geographical D2D communication network to provide high quality D2D communications according to physical closeness between devices. Then devices are divided into several social communities. Devices are ranked by their node importance to community in each community. The base station makes caching decisions for devices according to contrition degree. Experimental results show that IBPSC achieves best network performance.

NA-Caching: An Adaptive Content Management Approach Based on Deep Reinforcement Learning

Video streaming is a dominant application over today's Internet. The current mainstream video streaming solution is to utilize the services of a Content Delivery Network (CDN) provider. By replicating video content closer to the network edge, caching provides an effective mechanism for alleviating the demand for massive bandwidth for the Internet backbone. It reduces the network traffic and capital expense for streaming the video content, and in the meantime, enhance Internet's Quality of Service (QoS). In this paper, we propose a neural adaptive caching approach, named NA-Caching, for helping cache learn to make caching decisions from its own experiences rather than a specific mathematical model, in a way similar to how a human being learns a new skill (e.g. cycling, swimming). NA-Caching leverages the benefits of the Recurrent Neural Network (RNN) as well as the Deep Reinforcement Learning (DRL) to maximize the cache efficiency by jointly learning request features, caching space dynamics and making decisions. Specifically, we utilize Gated Recurrent Unit (GRU) to characterize the evolving features of the dynamic requests and caching space. Moreover, the above GRU-based representation network is integrated into a Deep Q-Network (DQN) framework for making adaptive caching decisions online. To evaluate the performance of the proposed approach, we conduct extensive experiments on anonymized real-world traces from a video provider. The results demonstrate that our algorithm significantly outperform several candidate methods.