Storage Cluster Research Articles

TPRepair: Tree-Based Pipelined Repair in Clustered Storage Systems

Erasure coding is an effective technique for guaranteeing data reliability for storage systems, yet it incurs a high repair penalty with amplified repair traffic. The repair becomes more intricate in clustered storage systems with the bandwidth diversity property. We present TPRepair , a T ree-based P ipelined Repair approach, aiming to expedite the overall repair process with the tailored pipelined repair procedure. TPRepair first prioritizes selecting racks with the current minimum load to participate in the repair process. It subsequently formulates tree-based links, tailored to align seamlessly with the pipelined repair procedure. TPRepair further designs an optimization algorithm to reduce the bottleneck load when repairing multiple chunks. Large-scale simulations demonstrate that TPRepair can increase 13.8%-41.3% of the balance ratio without amplifying cross-rack traffic. Meanwhile, Alibaba Cloud ECS experiments indicate that TPRepair can increase repair throughput by 11.3% to 72.9%.

Capturing Geological Uncertainty in Salt Cavern Developments for Hydrogen Storage

Future energy systems with a greater share of renewable energy will require long-duration energy storage (LDES) to optimise the integration of renewable sources and hydrogen is one energy vector that could be utilised for this. Grid-scale underground storage of natural gas (methane) is already in operation in solution-mined salt caverns, where individual cavern capacities are ∼0.025–0.275 TWh. While salt caverns have traditionally been restricted to being developed onshore, in some offshore locations, such as the UK Continental Shelf, there are extensive evaporites that have the potential for storage development. Capacity estimates for offshore areas typically rely on generalised regional geological interpretations; they frequently do not incorporate site-specific structural and lithological heterogeneities, they use static cavern geometries and may use methodologies that are deterministic and not repeatable. We have developed a stochastic method for identifying potential salt cavern locations and estimating conceptual cluster storage capacity. The workflow incorporates principle geomechanical constraints on cavern development, captures limitations from internal evaporite heterogeneities, and uses the ideal gas law to calculate the volumetric capacity. The workflow accommodates either fixed cavern geometries or geometries that vary depending on the thickness of the salt. By using a stochastic method, we quantify the uncertainties in storage capacity estimates and cavern placement over defined regions of interest. The workflow is easily adaptable allowing users to consider multiple geological models or to evaluate the impact of interpretations at varying resolutions. In this work, we illustrate the workflow for four areas and geological models in the UK’s Southern North Sea: 1) Basin Scale (58,900 km2) - >48,800 TWh of hydrogen storage with >199,000 cavern locations. 2) Sub-Regional Scale (24,800 km2) - >9,600 TWh of hydrogen storage with >36,000 cavern locations. 3) Block Specific–Salt Wall (79.8 km2) - >580 TWh of hydrogen storage with >400 cavern locations. 4) Block Specific–Layered Evaporite (225 km2) - >263 TWh of hydrogen storage with >500 cavern locations. Our workflow enables reproducible and replicable assessments of site screening and storage capacity estimates. A workflow built around these ideals allows for fully transparent results. We compared our results against other similar studies in the literature and found that often highly cited papers have inappropriate methodologies and hence capacities.

Efficient security interface for high-performance Ceph storage systems

Ceph portrays a resilient clustered storage solution with supporting object, block, and file storage capabilities with no single point of failure. Despite these qualifications, data confidentiality defines a concern in the system, as authentication and access control are the only data protection security services in Ceph. CephArmor was proposed as a third-party security interface to protect data confidentiality by adding an extra protection layer to data at rest. Despite the added layer, the initial design of the API needed to be more efficient in addressing security and performance simultaneously. In this study, we propose a new architectural design to address the associated issues with the preliminary prototype. Comprehensive performance and security analysis verify the improvement of the proposed method compared to the initial approach. The benchmark result has indicated a 37% improvement on average in IOPS, elapsed time, and bandwidth for the write benchmark compared to the initial model.

Optimal operation of multi-energy microgrids with shared energy storage clusters by master-slave games

Abstract As shared energy storage power plants have become a research hotspot, the cooperation between them and microgrids has attracted attention. In this paper, we propose a master-slave game optimization operation strategy for multi-energy microgrids, taking into account shared energy storage (ESS) clusters. Firstly, the inclusion of ESS clusters is considered in a combined heat and power (CHP) type multi-energy microgrid. Then, the optimal operation model is established and solved in two layers. The master-slave game model in the context of ESS clusters is proposed by analyzing the game relationship between the microgrid side and the user side, the arithmetic simulation with MATLAB, and the modeling and solving are carried out by the Yalmip tool and the CPLEX solver, and the combination of heuristic algorithms and the solver is used to optimize the strategies of the microgrid side and the user side. The optimization of the system after the optimization of this paper’s methodology can enhance the system economy.

Frequency emergency control strategy in power systems considering the participation of energy storage clusters

Recently, the power systems with a high penetration of renewables and power electronics have come into being. In these power systems, complex system dynamics, emergency faults, and insufficient frequency regulation reserve pose threats to system frequency stability. Based on the clustering development of energy storage, to ensure the system frequency stability when emergency faults occur, this paper proposes a decentralized frequency emergency control (FEC) strategy considering the participation of energy storage clusters (ESCs). First, the overall framework of the optimal-droop-based FEC strategy is introduced, which achieves the coordination between FEC and conventional frequency regulation strategies. Second, to appropriately allocate the unbalanced power among the generators and ESCs, a general design method for optimal droop coefficients is proposed, which is applicable to various control objectives. The optimality of the droop coefficients is rigorously proven. This case study is carried out on an electromagnetic transient simulation platform, and the simulation results verify the effectiveness and optimality of the proposed FEC strategy.

The Chinese Clinical Sleep Database: An Innovative Database System Includes Large-Scale Clinical Data of Chinese Population.

In this study, we established the Chinese Clinical Sleep Database (CCSD), aiming to provide a safe, scalable, and user-friendly database that includes high-quality clinical data from Chinese population to facilitate sleep research. We collect individual's demographic data, scales, anthropometric measurements, clinical diagnosis, and polysomnography (PSG) recordings from the routine medical process of sleep medicine centers using standardized procedures. The distributed cluster storage technology are utilized to store these data. The structured data are stored in a high-performance MySQL database, while the unstructured data are stored in an object storage service. And we have developed an online data platform to share and manage our data. The data collection has been conducted in three hospitals. In the preliminary stage of data collection (from October 18, 2022 to September 4, 2023), our database included a total of 1183 patients. Among them, 56.8% were male and their ages ranged from 3 to 88 years. These patients were diagnosed with various types of sleep disorders. Since the CCSD's inception, it has demonstrated good stability, security, and scalability. As an public database, the CCSD also exhibits user-friendliness. The CCSD contains comprehensive clinical data, which can contribute to the advancement of the diagnosis and treatment strategies for sleep disorders, ultimately promoting sleep health.

Enhancing Security in Database Grid Infrastructure for Storage Clusters

This research project primarily focused on improving security by addressing vulnerabilities and creating a cost-scalable storage cluster solution, for applications. The Oracle Grid Infrastructure for Storage Clusters (OGISC) represents cutting-edge advancements in this field. The research extensively explores the philosophy of OGISC. Explains its core methodology. Recognizing the weaknesses of existing security mechanisms, we propose a solution to enhance security measures. A key aspect of our investigation involves integrating Glusterfs, a system known for its ability to scale linearly. We delve into the architecture of the solution demystifying the storage scale-out processes to its operation. Our study goes beyond integration by developing an approach and metadata model specifically tailored for Glusterfs ensuring optimal performance and robustness. One noteworthy aspect of our research is the application of Glusterfs compression with OpenVPN. This exploration highlights the benefits derived from this integration emphasizing how OpenVPN enhances Glusterfs capabilities. Rigorous analysis stages serve as the foundation for our findings resulting in a forward-thinking solution. Finally, we conclude this research with an examination of avenues for future exploration in this dynamic field.

Practical Design Considerations for Wide Locally Recoverable Codes (LRCs)

Most of the data in large-scale storage clusters is erasure coded. At exascale, optimizing erasure codes for low storage overhead, efficient reconstruction, and easy deployment is of critical importance. Locally recoverable codes (LRCs) have deservedly gained central importance in this field, because they can balance many of these requirements. In our work, we study wide LRCs; LRCs with large number of blocks per stripe and low storage overhead. These codes are a natural next step for practitioners to unlock higher storage savings, but they come with their own challenges. Of particular interest is their reliability , since wider stripes are prone to more simultaneous failures. We conduct a practically minded analysis of several popular and novel LRCs. We find that wide LRC reliability is a subtle phenomenon that is sensitive to several design choices, some of which are overlooked by theoreticians, and others by practitioners. Based on these insights, we construct novel LRCs called Uniform Cauchy LRCs , which show excellent performance in simulations and a 33% improvement in reliability on unavailability events observed by a wide LRC deployed in a Google storage cluster. We also show that these codes are easy to deploy in a manner that improves their robustness to common maintenance events. Along the way, we also give a remarkably simple and novel construction of distance-optimal LRCs (other constructions are also known), which may be of interest to theory-minded readers.

Evaluating and aggregating the grid-support capability of energy storage clusters by considering the peak regulation requirements

With the rapid progression of Energy Storage Systems (ESSs), the capability of extensively distributed and heterogeneous ESSs to support the power grid remains largely underexplored. To better exploit the potential of these numerous ESSs and enhance their service to the power grid, this paper proposes a model for evaluating and aggregating the grid-support capability of energy storage clusters by considering the peak regulation requirements. To begin with, the proposed model employs subjective and objective combination weighting methods to establish a grid-support capability matrix between ESSs indicators and grid demand scenarios, thereby facilitating the identification of the ESSs with a strong ability to regulate peak power. Next, based on the dual-peak pattern of grid load and diverse characteristics of ESSs, the ESSs in the peak regulation cluster are evaluated by clustering again. In addition, taking into account the operational constraints of the ESSs and the peak regulation requirements, a grid-support capability aggregation model for energy storage clusters based on the revised Chino polytope is proposed. The case study results demonstrate that the proposed model not only balances computational efficiency and aggregation accuracy to a certain extent but also enhances the capability of energy storage clusters to participate in peak regulation of the power grid.

Hybrid COOT-Reverse Cognitive Fruit Fly Optimization-Based Big Data Services and Virtual Machine Allocation for Cloud Storage System

In recent years, cloud computing technologies have been developed rapidly in this computing world to provide suitable on-demand network access all over the world. A cloud service provider offers numerous types of cloud services to the user. But the most significant issue is how to attain optimal virtual machine (VM) allocation for the user and design an efficient big data storage platform thereby satisfying the requirement of both the cloud service provider and the user. Therefore, this paper presents two novel strategies for optimizing VM resource allocation and cloud storage. An optimized cloud cluster storage service is introduced in this paper using a binarization based on modified fuzzy c-means clustering (BMFCM) algorithm to overcome the negative issues caused by the repetitive nature of the big data traffic. The BMFCM algorithm utilized can be implemented transparently and can also address problems associated with massive data storage. The VM selection is optimized in the proposed work using a hybrid COOT-reverse cognitive fruit fly (RCFF) optimization algorithm. The main aim of this algorithm is to improve the massive big data traffic and storage locality. The CPU utilization, VM power, memory dimension and network bandwidth are taken as the fitness function of the hybrid COOT-RCFF algorithm. When implemented in CloudSim and Hadoop, the proposed methodology offers improvements in terms of completion time, overall energy consumption, makespan, user provider satisfaction and load ratio. The results show that the proposed methodology improves the execution time and data retrieval efficiency by up to 32% and 6.3% more than the existing techniques.

Memory Clustering Autoencoder Method for Human Action Anomaly Detection on Surveillance Camera Video

Unsupervised deep-learning methods with a deep clustering model are widely used to detect anomaly human actions obtained by surveillance camera due to their powerful image feature learning abilities. These methods aim to optimize a model through a clustering induction target to provide useful cluster assignment, and usually use a forward propagation process: an autoencoder reconstructs an input sequence, a clustering model provides cluster allocation, and a scoring model evaluates distribution and provides scores for each sample. However, these methods have two main problems: one is that an autoencoder is difficult to handle human posture when abnormal and normal actions occur in crowd scenes captured by surveillance camera. The other is that network updating is interrupted by feature extraction and clustering in one epoch. To solve these problems, we design a deep memory storage clustering method based on graph convolution autoencoder to implement the real-time updating of pseudo-labels and network parameters. We also design a new loss function to express similarity between sample feature and centroid feature in a memory storage module and to constrain the parameter update of a network.We evaluate the unsupervised method for three important and representative datasets mainly composed of surveillance videos and use the AUC score as an experimental evaluation indicator.

Data Distribution for Heterogeneous Storage Systems

The exponential growth of data in many science and engineering domains poses significant challenges to storage systems. Data distribution is a critical component in large-scale distributed storage systems and plays a vital role in placing petabytes of data and beyond, among tens to hundreds of thousands of storage devices. Meantime, heterogeneous storage systems, such as those having devices with hard disk drives (HDDs) and storage class memories (SCMs), have become increasingly popular for massive data storage due to their distinct and complement characteristics. This paper presents a new data distribution algorithm called SUORA (Scalable and Uniform storage via Optimally-adaptive and Random number Addressing) specifically for heterogeneous devices to maximize the benefits of them. SUORA provides a fully symmetric, highly efficient methodology to distribute data across a hybrid and tiered storage cluster. It divides heterogeneous devices into different buckets and segments, and adopts pseudo-random functions to map data onto them with the balanced consideration of capacity, performance and life-time. By analyzing hotness and access patterns, SUORA gradually moves hot data from HDDs to SCMs to optimize the throughput, and moves cold data reversely for load balance. It combines data replication with migration to significantly reduce movement overhead while making data placement more adaptive to different workloads. Extensive evaluations on simulation and Sheepdog storage system show that, with considering distinct characteristics of various devices thoroughly, SUORA improves the overall performance efficiency of heterogeneous storage systems.

Data Warehouse Failover Cluster for Analytical Queries in Banking

Introduction. The banking sector assigns high priority to data storage, as it is a critical aspect of business operations. The volume of data in this area is steadily growing. With the increasing volume of data that needs to be stored, processed and analyzed, it is critically important to select a suitable data storage solution and develop the required architecture. The presented research is aimed at filling the gap in the existing knowledge of the data base management system (DBMS) suitable for the banking sector, as well as to suggest ways for a fault-tolerant data storage cluster. The purpose of the work is to analyze the key DBMS for analytical queries, determine the priorities of the DBMS for the banking sector, and develop a fault-tolerant data storage cluster. To meet the performance and scalability requirements, a data storage solution with a fault-tolerant architecture that meets the requirements of the banking sector has been proposed.Materials and Methods. Domain analysis allowed us to create a set of characteristics that a DBMS for analytical queries (OnLine Analytical processing — OLAP) should correspond to, compare some popular DBMS OLAP, and offer a fault-tolerant cluster configuration written in xml, supported by the ClickHouse DBMS. Automation was done using Ansible Playbook. It was integrated with the Gitlab version control system and Jinja templates. Thus, rapid deployment of the configuration on all nodes of the cluster was achieved.Results. For OLAP databases, criteria were developed and several popular systems were compared. As a result, a reliable cluster configuration that met the requirements of analytical queries has been proposed for the banking industry. To increase the reliability and scalability of the DBMS, the deployment process was automated. Detailed diagrams of the cluster configuration were also provided.Discussion and Conclusions. The compiled criteria for the DBMS OLAP allowed us to determine the need for this solution in the organization. Comparison of popular DBMS can be used by organizations to minimize costs when selecting a solution. The proposed configuration of the data warehouse cluster for analytical queries in the banking sector will improve the reliability of the DBMS and meet the requirements for subsequent scalability. Automation of cluster deployment by the mechanism of templating configuration files in Ansible Playbook provides configuring a ready-made cluster on new servers in minutes.

An In-depth Comparative Analysis of Cloud Block Storage Workloads: Findings and Implications

Cloud block storage systems support diverse types of applications in modern cloud services. Characterizing their input/output (I/O) activities is critical for guiding better system designs and optimizations. In this article, we present an in-depth comparative analysis of production cloud block storage workloads through the block-level I/O traces of billions of I/O requests collected from two production systems, Alibaba Cloud and Tencent Cloud Block Storage. We study their characteristics of load intensities, spatial patterns, and temporal patterns. We also compare the cloud block storage workloads with the notable public block-level I/O workloads from the enterprise data centers at Microsoft Research Cambridge, and we identify the commonalities and differences of the three sources of traces. To this end, we provide 6 findings through the high-level analysis and 16 findings through the detailed analysis on load intensity, spatial patterns, and temporal patterns. We discuss the implications of our findings on load balancing, cache efficiency, and storage cluster management in cloud block storage systems.

Organically distributed sustainable storage clusters

Organically distributed sustainable storage clusters

Development of CCUS clusters in Croatia

Carbon Capture and Storage is a concept that is not yet fully implemented largely because of the high costs. Clustering of industrial stakeholders is imposed as a measure for cost reduction. All relevant emitters, possible transport routes, including existing gas pipeline corridors, and their geographic location in relation to potential storage locations are assessed in this paper. Site availability and CO2 storage capacity are examined, summarizing all study results gathered under the Strategy CCUS project. The CO2 enhanced oil recovery is being studied for CO2 storage rather than extra oil recovery. As logical choices, three clusters were recognized. Only less expensive, onshore injection was taken in consideration for assessment of early (economic) feasibility in the Adriatic, Central, and Eastern clusters. Because of the shorter distance between CO2 emitters and injection sites, the Eastern and Central clusters are being investigated in more detail, despite the fact that the largest point source emitter is in the Adriatic region. Because of small number of point source CO2 emitters and huge theoretical storage capacity, further research is needed to better assess the storage capacities as well as possibilities for development of cross-border projects. Based on previous research (particularly regarding the emitters), the number of facilities (fewer facilities, with more concentrated emissions), and the availability of storage objects, the Eastern cluster is recommended to be further studied as the next stage of Carbon Capture, Utilization and Storage cluster research and development in Croatia and nearby cross-border regions.

Oasis : Controlling Data Migration in Expansion of Object-based Storage Systems

Object-based storage systems have been widely used for various scenarios such as file storage, block storage, blob (e.g., large videos) storage, and so on, where the data is placed among a large number of object storage devices (OSDs). Data placement is critical for the scalability of decentralized object-based storage systems. The state-of-the-art CRUSH placement method is a decentralized algorithm that deterministically places object replicas onto storage devices without relying on a central directory. While enjoying the benefits of decentralization such as high scalability, robustness, and performance, CRUSH-based storage systems suffer from uncontrolled data migration when expanding the capacity of the storage clusters (i.e., adding new OSDs), which is determined by the nature of CRUSH and will cause significant performance degradation when the expansion is nontrivial. This article presents MapX , a novel extension to CRUSH that uses an extra time-dimension mapping (from object creation times to cluster expansion times) for controlling data migration after cluster expansions. Each expansion is viewed as a new layer of the CRUSH map represented by a virtual node beneath the CRUSH root. MapX controls the mapping from objects onto layers by manipulating the timestamps of the intermediate placement groups (PGs). MapX is applicable to a large variety of object-based storage scenarios where object timestamps can be maintained as higher-level metadata. We have applied MapX to the state-of-the-art Ceph-RBD (RADOS Block Device) to implement a migration-controllable, decentralized object-based block store (called Oasis ). Oasis extends the RBD metadata structure to maintain and retrieve approximate object creation times (for migration control) at the granularity of expansion layers. Experimental results show that the MapX -based Oasis block store outperforms the CRUSH-based Ceph-RBD (which is busy in migrating objects after expansions) by 3.17× ∼ 4.31× in tail latency, and 76.3% (respectively, 83.8%) in IOPS for reads (respectively, writes).

A Configurable Model-Based Reinforcement Learning Framework for Disaggregated Storage Systems

With the rapid growth of data-intensive jobs and the use of different hardware in storage, disaggregated storage architecture systems are being used to improve the operational cost efficiency of data centers. The hardware heterogeneity and mixed configurations of disaggregated storage systems, along with the diversity of workloads, often make it difficult for administrators to operate them optimally. In this work, we investigate model-based reinforcement learning (RL) schemes to develop automated system operations and maintain the storage performance across various system settings and workloads in self-managed storage systems. Specifically, we propose a novel configurable model structure in which a system environment is abstracted with a two-level hierarchy of storage devices and a platform and thus the environment can be reconfigured according to a given system specification. Using that novel model structure, we implement a configurable model-based RL framework CoMoRL by which RL agents are trained through model variants that represent a variety of storage system specifications; thus, their learned management policy can be highly robust to the diverse operation conditions of real-world storage systems. We evaluate our CoMoRL framework using a storage cluster that relies on NVMe-oF devices and demonstrate that the framework can be adapted to different scenarios such as volume placement scenarios with Kubernetes and primary affinity control scenarios with Ceph. The learned management policy outperforms an IOPS-based heuristic method and a model-based method by 0.7%~5.1% and 11.8%~29.7%, respectively, for various Kubernetes system specifications, and by 1.6%~5.6% and 8.2%~16.5%, respectively, for various Ceph system specifications, without requiring model and policy retraining. This zero-shot adaptation superiority of our framework makes it possible to realize RL-based self-managing storage systems in data centers with frequent system changes.

Fast Proactive Repair in Erasure-Coded Storage: Analysis, Design, and Implementation

Erasure coding offers a storage-efficient redundancy mechanism for maintaining data availability guarantees in large-scale storage clusters, yet it also incurs high performance overhead in failure repair. Recent developments in accurate disk failure prediction allow soon-to-fail (STF) nodes to be repaired in advance, thereby opening new opportunities for accelerating failure repair in erasure-coded storage. To this end, we present a fast proactive repair solution called FastPR, which carefully couples two repair methods, namely migration (i.e., relocating the chunks of an STF node) and reconstruction (i.e., decoding the chunks of an STF node through erasure coding), so as to fully parallelize the repair operation across the storage cluster. FastPR solves a bipartite maximum matching problem and schedules both migration and reconstruction in a parallel fashion. We show that FastPR significantly reduces the repair time over the baseline repair approaches for both Reed-Solomon codes and Azures Local Reconstruction Codes via mathematical analysis, large-scale simulation, and Amazon EC2 experiments.

Bandwidth-Aware Scheduling Repair Techniques in Erasure-Coded Clusters: Design and Analysis

Erasure codes offer a storage-efficient redundancy mechanism for maintaining data availability guarantees in storage clusters, yet also incur high network traffic consumption and recovery time in failure repair. Extensive research has been carried out to reduce the recovery time. However, previous works either target specific erasure code constructions which are not commonly used in today’s distributed storage clusters or neglect the heterogeneous bandwidth property in real network environments. Since erasure-coded clusters are typically composed of multi-node with heterogeneous bandwidth and accessed in parallel, the whole recovery time is mainly restricted by the low-bandwidth links. In this article, we propose SMFRepair, a single-node multi-level forwarding repair technique that is designed to improve the performance in heterogeneous networks based on Reed-Solomon codes for general fault tolerance. SMFRepair carefully selects the helper nodes and uses idle nodes to bypass low-bandwidth links. Idle nodes have sufficient and unused network bandwidth. It also pipelines the repair links that are optimized by idle nodes. Furthermore, a multi-node scheduling repair technique, called MSRepair, is proposed. MSRepair carefully schedules the multi-node repair link to saturate the most unoccupied bandwidth and transfers data from as large-bandwidth links as possible, with the primary objective of minimizing the recovery time. Large-scale simulation and Amazon EC2 real experiments show that compared to state-of-the-art repair techniques, SMFRepair can accelerate the single-node recovery by up to 47.69%, and MSRepair can reduce the multi-node recovery time by 33.78% <inline-formula><tex-math notation="LaTeX">$\sim$</tex-math></inline-formula> 67.53%.