Data Access Latency Research Articles

ЕФЕКТИВНІ СХОВИЩА ДАНИХ ДЛЯ РІШЕНЬ МАШИННОГО НАВЧАННЯ

Awareness of the importance of using effective data warehouses to develop machine learning solutions in today's world is becoming increasingly relevant in connection with the growth of data volumes and the growing need for accurate and productive models. The article identifies major vital requirements for data storage systems to ensure effective data management in building machine learning solutions. The study underscores the practical implications of rapid data access and minimal latency, reliable storage systems with robust backup and recovery mechanisms, high-level security, flexibility in handling different data types, and cost-effectiveness. These factors are not just theoretical considerations but directly impact the efficiency and success of machine learning solutions. Together with describing the concepts of block storage and object storage, the article also compares major cloud platforms—AWS (Amazon Web Services), Microsoft Azure, and Google Cloud Platform (GCP)—each providing a comprehensive range of flexible and customizable storage, security, and data processing services. These platforms enable users to manage their information resources flexibly and efficiently, offering unique features that cater to the specific needs of machine learning applications. Understanding the nuances of each provider's offerings empowers engineers and researchers to make informed decisions that align with their unique requirements. The article provides a thorough overview of the key considerations and available options, guiding strategic decisions to support machine learning initiatives effectively. Also the importance of staying up to date with the evolution of major platforms offerings is emphasized. In conclusion, selecting the proper data storage solution is critical for enhancing the performance, security, and cost-efficiency of machine learning models.

TALICS3: Tape library cloud storage system simulator

High performance computing data is surging fast into the exabyte-scale world, where tape libraries are the main platform for long-term durable data storage besides high-cost DNA. Tape libraries are extremely hard to model, but accurate modeling is critical for system administrators to obtain valid performance estimates for their designs. This research introduces a discrete–event tape simulation platform that realistically models tape library behavior in a networked cloud environment, by incorporating real-world phenomena and effects. The platform addresses several challenges, including precise estimation of data access latency, rates of robot exchange, data collocation, deduplication/compression ratio, and attainment of durability goals through replication or erasure coding. Using the proposed simulator, one can compare the single enterprise configuration with multiple commodity library configurations, making it a useful tool for system administrators and reliability engineers. This makes the simulator a valuable tool for system administrators and reliability engineers, enabling them to acquire practical and dependable performance estimates for their enduring, cost-efficient cold data storage architecture designs.

Hybrid fault tolerance in distributed in-memory storage systems

An in-memory storage system provides submillisecond latency and improves the concurrency of user applications by caching data into memory from external storage. Fault tolerance of in-memory storage systems is essential, as the loss of cached data requires access to data from external storage, which evidently increases the response latency. Typically, replication and erasure code (EC) are two fault-tolerant schemes that pose different trade-offs between access performance and storage usage. To help make the best performance and space trade-off, we design ElasticMem, a hybrid fault-tolerant distributed in-memory storage system that supports elastic redundancy transition to dynamically change the fault-tolerant scheme. ElasticMem exploits a novel EC-oriented replication (EOR) that carefully designs the data placement of replication according to the future data layout of EC to enhance the I/O efficiency of redundancy transition. ElasticMem solves the consistency problem caused by concurrent data accesses via a lightweight table-based scheme combined with data bypassing. It detects corelated read and write requests and serves subsequent read requests with local data. We implement a prototype that realizes ElasticMem based on Memcached. Experiments show that ElasticMem remarkably reduces the time of redundancy transition, the overall latency of corelated concurrent data accesses, and the latency of single data access among them.

Predicting Resource Utilization Trends with Southern California Petabyte Scale Cache

Large community of high-energy physicists share their data all around world making it necessary to ship a large number of files over wide- area networks. Regional disk caches such as the Southern California Petabyte Scale Cache have been deployed to reduce the data access latency. We observe that about 94% of the requested data volume were served from this cache, without remote transfers, between Sep. 2022 and July 2023. In this paper, we show the predictability of the resource utilization by exploring the trends of recent cache usage. The time series based prediction is made with a machine learning approach and the prediction errors are small relative to the variation in the input data. This work would help understanding the characteristics of the resource utilization and plan for additional deployments of caches in the future.

Labeled graph partitioning scheme for distributed edge caching

Distributed edge caching could address latency and congestion problems in large-scale data access effectively, improving system throughput and performance. However, the lack of specialized edge caching solutions for graph data resulted in low cache hit rates and high data access latency. This limitation stemmed from existing graph partitioning schemes that overlooked the specific requirements of edge caching, such as communication overhead between edge servers, data query frequency, and query patterns. To overcome these challenges, we proposed LGPE, a labeled graph partitioning scheme for distributed edge caching based on frequent query patterns. LGPE generated frequent query patterns according to the user’s historical query subgraphs and then performed labeled graph partitioning based on the frequent pattern, ensuring that the labeled graph divided into edge servers had a high hit rate for frequent user queries while satisfying a minimum edge cut. Evaluation on real datasets from various application domains demonstrated that LGPE achieved approximately 40% higher cache hit rates compared to benchmarks like METIS and SCOTCH, and it also performed well in terms of edge cuts in special graph cases.

QoS-Aware and Energy Data Management in Industrial IoT

Two crucial challenges in Industry 4.0 involve maintaining critical latency requirements for data access and ensuring efficient power consumption by field devices. Traditional centralized industrial networks that provide rudimentary data distribution capabilities may not be able to meet such stringent requirements. These requirements cannot be met later due to connection or node failures or extreme performance decadence. To address this problem, this paper focuses on resource-constrained networks of Internet of Things (IoT) systems, exploiting the presence of several more powerful nodes acting as distributed local data storage proxies for every IoT set. To increase the battery lifetime of the network, a number of nodes that are not included in data transmission or data storage are turned off. In this paper, we investigate the issue of maximizing network lifetime, and consider the restrictions on data access latency. For this purpose, data are cached distributively in proxy nodes, leading to a reduction in energy consumption and ultimately maximizing network lifetime. To address this problem, we introduce an energy-aware data management method (EDMM); with the goal of extending network lifetime, select IoT nodes are designated to save data distributively. Our proposed approach (1) makes sure that data access latency is underneath a specified threshold and (2) performs well with respect to network lifetime compared to an offline centralized heuristic algorithm.

Latency-Aware Task Scheduling in Software-Defined Edge and Cloud Computing With Erasure-Coded Storage Systems

The collaborative edge and cloud computing system has emerged as a promising solution to fulfill the unprecedented high requirements of 5G application scenarios. Due to vendor variations, it is often difficult to manage hardware facilities in such a collaborative system. Moreover, the amount of data generated and tasks requested by end devices are increasing exponentially, which introduces storage and computation bottlenecks. To address these issues, a novel systematic framework called software-defined edge and cloud computing (SD-ECC) is designed to manage the underlying physical resources of edge and cloud layers via software.SD-ECC is combined with an erasure-coded storage system, for which a task scheduling problem is formulated by considering data access and task processing steps. Then, a joint data access and task processing (JDATP) algorithm is proposed to minimize the task response time including data access latency and task processing latency. A practical SD-ECC platform is developed on OpenStack, OpenDaylight, and Kubernetes. Experiments are conducted with real-world datasets. The experimental results demonstrated that our proposed JDATP algorithm can reduce 20.87% of the task response time and increase 14.16% of the remaining storage space on average by comparing it with alternative schemes.

Adaptive and Scalable Caching With Erasure Codes in Distributed Cloud-Edge Storage Systems

Erasure codes have been widely used to enhance data resiliency with low storage overheads. However, in geo-distributed cloud storage systems, erasure codes may incur high service latency as they require end users to access remote storage nodes to retrieve data. An elegant solution to achieving low latency is to deploy caching services at the edge servers close to end users. In this paper, we propose adaptive and scalable caching schemes to achieve low latency in the cloud-edge storage system. Based on the measured data popularity and network latencies in real time, an adaptive content replacement scheme is proposed to update caching decisions upon the arrival of requests. Theoretical analysis shows that the reduced data access latency of the replacement scheme is at least 50% of the maximum reducible latency. With the low computation complexity of our design, nearly no extra overheads will be introduced when handling intensive data flows. For further performance improvements without sacrificing its efficiency, an adaptive content adjustment scheme is presented to replace the subset of cached contents that incur the aforementioned performance loss. Driven by real-world data traces, extensive experiments based on Amazon Simple Storage Service demonstrate the effectiveness and efficiency of our design.

EA2-IMDG: Efficient Approach of Using an In-Memory Data Grid to Improve the Performance of Replication and Scheduling in Grid Environment Systems

This paper proposes a novel approach, EA2-IMDG (Efficient Approach of Using an In-Memory Data Grid) to improve the performance of replication and scheduling in grid environment systems. Grid environments are widely used for distributed computing, but they are often faced with the challenge of high data access latency and poor scalability. By utilizing an in-memory data grid (IMDG), the aim is to significantly reduce the data access latency and improve the resource utilization of the system. The approach uses the IMDG to store data in RAM, instead of on disk, allowing for faster data retrieval and processing. The IMDG is used to distribute data across multiple nodes, which helps to reduce the risk of data bottlenecks and improve the scalability of the system. To evaluate the proposed approach, a series of experiments were conducted, and its performance was compared with two baseline approaches: a centralized database and a centralized file system. The results of the experiments show that the EA2-IMDG approach improves the performance of replication and scheduling tasks by up to 90% in terms of data access latency and 50% in terms of resource utilization, respectively. These results suggest that the EA2-IMDG approach is a promising solution for improving the performance of grid environment systems.

NCDE: In-Network Caching for Directory Entries to Expedite Data Access in Tiled-Chip Multiprocessors

The processing of data-intensive applications, followed by an unprecedented amount of data traffic, drives explosive accesses to the memory subsystem. The overloaded memory subsystem experiences increased data access latency. To expedite data access, a network caching technique that leverages network-on-chip (NoC) virtual channels (VCs) as an expanded memory subsystem has emerged. Previous network caching studies focused on utilizing VCs on the NoC’s local input port as a victim cache to reduce local data access latency. In contrast to previous studies, we explore the opportunity of mitigating problems associated with shared data access via in-network caching for directory entries (NCDE), which can utilize every input port’s VCs to hold directory entries. NCDE exploits VCs as the victim and prefetch buffers of the directory entries, each reducing directory eviction-induced invalidations and simplifying the cache-to-cache (C2C) data transfer. The effectiveness of NCDE was evaluated using a gem5 full-system simulator, and the results show that the average memory access time (AMAT) and workload execution time were reduced by 7.69% and 5.82%, respectively. As a cost for accelerating the data access latency, implementing NCDE incurs a negligible router area overhead of 1.56%.

Popularity-Based Data Placement With Load Balancing in Edge Computing

In recent years, edge computing has become an increasingly popular computing paradigm to enable real-time data processing and mobile intelligence. Edge computing allows computing at the edge of the network, where data is generated and distributed at the nearby edge servers to reduce the data access latency and improve data processing efficiency. One of the key challenges in data-intensive edge computing is how to place the data at the edge clouds effectively such that the access latency to the data is minimized. In this paper, we study such a data placement problem in edge computing while different data items have diverse popularity. We propose a popularity based placement method which maps both data items and edge servers to a virtual plane and places or retrieves data based on its virtual coordinate in the plane. We then further propose additional placement strategies to handle load balancing among edge servers via either offloading or data duplication. Simulation results show that our proposed strategies efficiently reduce the average path length of data access and the load-balancing strategies indeed provide an effective relief of storage pressures at certain overloaded servers.

Evaluation of Hash Join Operations Performance Executing on SDN Switches: A Cost Model Approach

Distributed database systems store and manipulate data on multiple machines. In these systems, the processing cost of query operations is mainly impacted by the data access latency between machines over the network. With recent technology advances in programmable network devices, the network switches provide new opportunities for dynamically managing the network topology, enabling the data processing on these devices with the same network throughput. In this paper, we explore the programmable network switches in query processing, evaluating the processing performance of a cost model in executing the hash join operation. We assume the storage of the hash table built from outer relation and the materialization of the join probing are made in switches using advanced matching techniques similar to package inspections enabled by Ternary Content-Addressable Memories (TCAM) or SRAM via hashing. Our results show that processing the hash join operation using network switches achieved the best results compared to traditional servers, with an average time reduction of 91.82% (Query-10 from TPC-H) and 96.52% (Query-11 from TPC-H).

PPOAccel: A High-Throughput Acceleration Framework for Proximal Policy Optimization

Reinforcement Learning (RL) is a major branch of AI that enables agents to learn optimal decision making via interaction with the environment. Proximal Policy Optimization (PPO) is the state-of-the-art policy optimization based RL algorithm which achieves superior overall performance on various benchmarks. A PPO agent iteratively optimizes its policy - a function which chooses optimal actions approximated by a DNN, with each iteration consisting of two computationally intensive phases: Sample Generation - where agents inference on its policy and interact with the environment to collect data, and Model Update - where the policy is trained using the collected data. In this paper, we develop the first high-throughput PPO accelerator on CPU-FPGA heterogeneous platform. Our unified systolic-array based design accelerates both the inference and the training of the deep neural network used in a RL algorithm, and is generalizable to various MLP and CNN models across a wide range of RL applications. We develop novel optimizations to simultaneously reduce data access and computation latencies, specifically: (a) optimal data flow mapping to systolic array, (b) novel memory-blocked data layout to enable streaming stall-free data access in both forward and backward propagations, and, (c) a systolic array compute sharing technique to mitigate load imbalance in the training of two networks. We evaluate our design on widely used robotics and gaming benchmarks, achieving 1.4×–26× and 1.3×–2.7× improvements in throughput, respectively, when compared with state-of-the-art CPU/CPU-GPU implementations.

Intelligent Replica Selection in Edge and IoT Environments Using Artificial Neural Networks

Cloud, edge and Internet of Things (IoT) technologies have emerged to overcome the challenges involved in sharing computational resources and information services. Within generic cloud systems, two models have been identified as having widespread applicability: computation clouds and data clouds. A data cloud is cloud computing that aims to manage, unify and operate multiple data workloads. Many current applications generate datasets consisting of petabytes (PB) of information. Managing large datasets is a complex issuel; in particular, datasets associated with many applications can be distributed widely in geographical terms, particularly in IoT systems. Edge and IoT systems are facing new challenges with increased complexity, making scalability an important issue that will affect the performance of the system. Data replication services are widely accepted techniques to improve availability and fault tolerance, and to improve the data access time. Current replication services, however, often exhibit an increase in response time, reflecting the problems associated with the ever-increasing size of databases. This paper proposes a prediction model to predict replica locations using the files’ access profile, which feeds the neural networks with the access and location behavior (file profile) to minimize the overhead of transferring large volumes of data, which slows down the system and requires careful management. This new model has shown high accuracy and low overheads. The result shows a significant improvement in total task execution time using the proposed model for locating files by 16.34% and 30.45%; in addition, the results show bandwidth improvement by 24.7% and 49.4% compared to the user profile prediction model and replica service model without prediction, respectively. Consequently, the proposed algorithm can improve data access speed, reduce data access latency and decrease bandwidth consumption.

HTIDB: Hierarchical Time-Indexed Database for Efficient Storage and Access to Irregular Time-series Health Sensor Data.

With the enormous amount of data collected by unobtrusive sensors, the potential of utilizing these data and applying various multi-modal advanced analytics on them is numerous and promising. However, taking advantage of the ever-growing data requires high-performance data-handling systems to enable high data scalability and easy data accessibility. This paper demonstrates robust design, developments, and techniques of a hierarchical time-indexed database for decision support systems leveraging irregular and sporadic time series data from sensor systems, e.g., wearables or environmental. We propose a technique that leverages the flexibility of general purpose, high-scalability database systems, while integrating data analytics focused column stores that leverage hierarchical time indexing, compression, and dense raw numeric data storage. We have evaluated the performance characteristics and tradeoffs of each to understand the data access latencies and storage requirements, which are key elements for capacity planning for scalable systems.

Optimal Caching for Low Latency in Distributed Coded Storage Systems

Erasure codes have been widely considered as a promising solution to enhance data reliability at low storage costs. However, in modern geo-distributed storage systems, erasure codes may incur high data access latency as they require data retrieval from multiple remote storage nodes. This hinders the extensive application of erasure codes to data-intensive applications. This paper proposes novel caching schemes to achieve low latency in distributed coded storage systems. Assuming that future data popularity and network latency information are available, an offline caching scheme is proposed to explore the optimal caching solution for low latency. The proposed scheme categorizes all feasible caching decisions into a set of cache partitions, and then obtains the optimal caching decision through market clearing price for each cache partition. Furthermore, guided by the optimal scheme, an online caching scheme is proposed according to the measured data popularity and network latency information in real time, without the need to completely override the existing caching decisions. Both theoretical analysis and experiment results demonstrate that the online scheme can approximate the offline optimal scheme well with dramatically reduced computation complexity.

A tail-tolerant cloud storage scheduling based on precise periodicity detection

Cloud storage is a fundamental component of the cloud computing system, which significantly affects the overall performance and quality of service of the cloud. Cloud storage servers face the challenge of imbalanced workloads. According to our observations on the time series generated by cloud storage, we found that the imbalance workloads will dramatically increase the tail latency of data access in the multi-tenant scenario. The intuitive solution is to periodicity detect the imbalance storage nodes and re-balance the loads. However, there are four challenges to accurately detect load of storage in the cloud with multiple tenants since the load may change frequently in cloud. This paper proposes PrecisePeriod, a precise periodicity detection algorithm customized for multi-tenant cloud storage. It removes outliers through data preprocessing, employs the discrete wavelet transform to remove high-frequency noise while keeping frequency domain information, computes the candidate periodicity queue using the autocorrelation function, and determines precise period through periodicity verification. Then, we design a cloud storage load balancing scheduling strategy based on PrecisePeriod, and the evaluation shows that the PrecisePeriod scheduling significantly reduces tail latency while only bringing 1-2% overhead.

An energy‐efficient data management scheme for industrial IoT

SummaryIndustry 4.0 is the emerging trend of manufacturing technologies towards smart factories. The industrial internet of things (IIoT) plays a primordial role to achieve the objectives of Industry 4.0, by enabling the cyber‐physical production systems to communicate and cooperate with each other and with humans both internally and across the participants of the supply chain. The Publish/Subscribe is the communication model used commonly in IIoT networks, in which the data generated by sensors are cached in a central controller to be subsequently consumed by actuators. However, the traditional centralized data management is inappropriate for real applications because of its communication overhead and inability to cope with strict delay requirements of IIoT applications and energy constraints of IIoT networks. To address these issues, we propose an energy‐efficient data management scheme (EDMS), in which a set of IoT nodes are selected to cache data distributedly, so that the network energy consumption is optimized while the constraints on data access latency and nodes' cache capacity are met. In this measure, the extra IoT nodes not involved in data transmission nor in data caching are switched off to prolong the network lifetime. We model the aforementioned optimization problem as an integer linear programming (ILP) considering the realistic constraints and characteristics of IIoT networks and solve it using the CPLEX tool. Moreover, we propose a metaheuristic algorithm based on ant colony optimization to solve the aforementioned problem in a timely manner. Simulation results show that EDMS outperforms significantly alternative solutions in terms of energy consumption, data latency, and cache utilization.

Joint Content-Prefetching, Transmission Scheduling, and Rate Adaptation in Vehicular Networks

Content prefetching at road side units (RSUs) in vehicular networks can improve users’ experience by reducing the average data access latency. Prefetching algorithms proposed in the literature usually rely on a fixed modulation scheme, that corresponds to the lowest possible data rate for RSU-to-vehicles transmission. As a consequence, the network bandwidth is often inefficiently used. In this article, we propose to study the problem of efficient <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">content broadcasting</i> in vehicular networks and express it as a joint <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">data-prefetching</i> and broadcast transmission scheduling optimization problem, subject to cache size constraints and transmission rate adaptation constraints. In our study we consider a highway scenario covered by several RSUs, and take into account the deadline requirement of interactive applications. We express the problem as an integer linear programming problem, and solve it using a heuristic algorithm that determines: the content prefetching schedule, the broadcast transmission schedule and the rate adaptation schedule sequentially. Simulation results show that, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i)</i> it is beneficial to use adaptive rates to broadcast data instead of relying on single-rate with unicast transmissions; <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">ii)</i> using vehicular trajectories and content popularity in conjunction yields better results than simply relying on content popularity for content prefetching and caching; and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">iii)</i> this holds even with inaccurately predicted vehicle locations and with delay-sensitive contents.

Toward a Detailed Evaluation of Wireless Industrial Data Distribution Approaches.

Data distribution is a cornerstone of efficient automation for intelligent machines in Industry 4.0. Although in the recent literature there have been several comparisons of relevant methods, we identify that most of those comparisons are either theoretical or based on abstract simulation tools, unable to uncover the specific, detailed impacts of the methods to the underlying networking infrastructure. In this respect, as a first contribution of this paper, we develop more detailed and fine-tuned solutions for robust data distribution in smart factories on stationary and mobile scenarios of wireless industrial networking. Using the technological enablers of WirelessHART, RPL and the methodological enabler of proxy selection as building blocks, we compose the protocol stacks of four different methods (both centralized and decentralized) for data distribution in wireless industrial networks over the IEEE 802.15.4 physical layer. We implement the presented methods in the highly detailed OMNeT++ simulation environment and we evaluate their performance via an extensive simulation analysis. Interestingly enough, we demonstrate that the careful selection of a limited set of proxies for data caching in the network can lead to an increased data delivery success rate and low data access latency. Next, we describe two test cases demonstrated in an industrial smart factory environment. First, we show the collaboration between robotic elements and wireless data services. Second, we show the integration with an industrial fog node which controls the shop-floor devices. We report selected results in much larger scales, obtained via simulations.