Data Center Architecture Research Articles

AlveoliNet: An Incrementally Scalable, Cost-Effective, and High-Performance Two-Layer Based Architecture for Data Centers

AlveoliNet: An Incrementally Scalable, Cost-Effective, and High-Performance Two-Layer Based Architecture for Data Centers

Data architectures and the prevalent cyberattacks encountered by West African higher educational institutions in the COVID-19 era

The growing interconnectedness of the world has led to a rise in cybersecurity risks. Although it is increasingly conventional to use technology to assist business transactions, exposure to these risks must be minimised to allow business owners to do transactions in a secure manner. While a wide range of studies have been undertaken regarding the effects of cyberattacks on several industries and sectors, However, very few studies have focused on the effects of cyberattacks on the educational sector, specifically higher educational institutions (HEIs) in West Africa. Consequently, this study developed a survey and distributed it to HEIs particularly universities in West Africa to examine the data architectures they employed, the cyberattacks they encountered during the COVID-19 pandemic period, and the role of data analysis in decision-making, as well as the countermeasures employed in identifying and preventing cyberattacks. A total of one thousand, one hundred and sixty-four (1164) responses were received from ninety-three (93) HEIs and analysed. According to the study’s findings, data-informed architecture was adopted by 71.8% of HEIs, data-driven architecture by 24.1%, and data-centric architecture by 4.1%, all of which were vulnerable to cyberattacks. In addition, there are further concerns around data analysis techniques, staff training gaps, and countermeasures for cyberattacks. The study’s conclusion includes suggestions for future research topics and recommendations for repelling cyberattacks in HEIs.

Rcmp: Reconstructing RDMA-Based Memory Disaggregation via CXL

Memory disaggregation is a promising architecture for modern datacenters that separates compute and memory resources into independent pools connected by ultra-fast networks, which can improve memory utilization, reduce cost, and enable elastic scaling of compute and memory resources. However, existing memory disaggregation solutions based on remote direct memory access (RDMA) suffer from high latency and additional overheads including page faults and code refactoring. Emerging cache-coherent interconnects such as CXL offer opportunities to reconstruct high-performance memory disaggregation. However, existing CXL-based approaches have physical distance limitation and cannot be deployed across racks. In this article, we propose Rcmp, a novel low-latency and highly scalable memory disaggregation system based on RDMA and CXL. The significant feature is that Rcmp improves the performance of RDMA-based systems via CXL, and leverages RDMA to overcome CXL’s distance limitation. To address the challenges of the mismatch between RDMA and CXL in terms of granularity, communication, and performance, Rcmp (1) provides a global page-based memory space management and enables fine-grained data access, (2) designs an efficient communication mechanism to avoid communication blocking issues, (3) proposes a hot-page identification and swapping strategy to reduce RDMA communications, and (4) designs an RDMA-optimized RPC framework to accelerate RDMA transfers. We implement a prototype of Rcmp and evaluate its performance by using micro-benchmarks and running a key-value store with YCSB benchmarks. The results show that Rcmp can achieve 5.2× lower latency and 3.8× higher throughput than RDMA-based systems. We also demonstrate that Rcmp can scale well with the increasing number of nodes without compromising performance.

An energy-efficient cloud data center architectural model for the national data center of Bangladesh

This study presents an energy-efficient cloud data center architecture concept for Bangladesh’s National Data Center. To reduce energy consumption and operational expenses, the model employs virtualization and load-balancing techniques, as well as renewable energy sources. The simulation results demon- strate that the model may drastically cut energy usage while maintaining high levels of performance and dependability. This research contributes to the worldwide challenge of climate change by offering a long-term solution for reducing the environmental effect of cloud data centers while maintaining the quality of cloud computing services.

PLPose: An efficient framework for detecting power lines via key points-based pose estimation

The inspection and maintenance of electrical power lines (PL) via unmanned aerial vehicles (UAV) require fast and accurate PL detection to ensure smooth and secure electrical operations. However, the detection of PLs from aerial images is a highly challenging task due to the thin nature of PLs and the inherent noisy image backgrounds. Traditional line and edge detection methods do not detect the PLs accurately due to the cluttered backgrounds while the more recent deep learning (DL) CNNs are also not feasible for efficient PL detection due to the coarse bounding boxes and the computationally expensive pixel-based segmentations. Hence, in this study we propose PLPose, a novel framework for detecting the PLs via key points-based pose estimation technique and adapt the MobileNetV3 CNN for this task (kMobileNetV3), by adding a simple key point detection head to predict the PL key points. We also introduce a novel data-centric architecture (kMobileNetV3 + UDP), by adding the unbiased data processing (UDP) module to our kMobileNetV3, for faster and more accurate key point-based PL detection along with novel methods for data annotations and performance evaluation. Evaluations of PLPose on three benchmark PL datasets (PLDM, PLDU and the Mendeley Powerline Dataset) reveal that our proposed framework outperforms the state-of-the-art top-down pose estimation networks (HRNet-w32, HRNet-w32 + UDP and Resnet-50 Simple Baselines) in processing speed (∼29 FPS) and model size (5.23 M) for PL detection. Thus, the comprehensive experimental results demonstrate the effectiveness of our proposed framework. Our code is available from Github (https://www.github.com/rubeea/pl_mmpose).

VERCO: A privacy and data-centric architecture for verifiable cooperative maneuvers

Cooperative, Connected and Automated Mobility (CCAM) applications, enabled by vehicular communications and vehicle automation technologies, are set to increase traffic safety and efficiency. An important feature of CCAM is the potential to decide and coordinate maneuvers among automated vehicles in a more efficient and secure manner, when compared to human drivers. However, maneuver decision and coordination is a rather complex topic due to the multitude of possible maneuvers and the occasional cooperation failures, as caused by issues derived from the use of wireless communications in the vehicular ecosystem, like packet loss or even misbehaving entities. This work focuses on the latter problem by introducing a secure communication design for the decision of general cooperative maneuvers and the distributed storage of related data for accountability purposes using mechanisms based on distributed ledger technologies, while meeting user and data privacy requirements. We present the VERCO (VERifiable COoperation), a scalable geographical-based multi-layered blockchain architecture able to support the high volume of shared vehicular data in order to enhance the security and accountability of cooperative maneuvers, as well as potentially being able to support other vehicular services. To support this architecture, we also provide a performant communication protocol for the decision and negotiation of cooperative maneuvers, based on a new message: the Verifiable Cooperation Message (VCM). The architecture and communication protocol are analyzed and tested employing hardware-in-the-loop (HiL) simulations using two ETSI ITS standard compliant on-board units (OBUs) while deciding maneuvers in a lane-merging scenario. To test the scalability of the architecture a roadside unit (RSU) is put under different stress loads using a variable number of simulated vehicles in the same lane-merging scenario. Experimental results show the feasibility of the communication protocol, with an average delay of 34.68 ms, and the lightweightness of the proposed architecture, with an average overhead of only 2.62 ms for two vehicles and less than 6 ms for dozens of vehicles.

Architecture and Data Knowledge of the Regional Data Center for Intelligent Agriculture

The main task of the National Research Program “Smart crop production”, supported by the Ministry of Education and Science of Bulgaria and approved by the Council of Ministers, is the development of a regional data center to facilitate the work of farmers. The regional data center is part of the implementation of a smart crop production environment called ZEMEL which provides personal assistants supporting the work of farmers. The environment provides intelligent services for crop analysis and prevention and assists farmers in performing basic tasks related to crop production. The objective of the proposed article is to present the implementation of the architecture, infrastructure, and data architecture of a regional data center in the Plovdiv region. In order to clearly present the results of this work, which are the architectural and physical implementations of a regional data center and the storage of dynamic data and background knowledge, a methodology consisting of several steps is followed: the system infrastructure of the data center and the data architecture are discussed; one of the local pieces of infrastructure, implemented in the Institute of Plant Genetic Resources (IPGR) in the town of Sadovo in the Plovdiv region, is presented in detail, including the different types of sensors and their connection to the data center in wheat cultivation; the data repositories are discussed where dynamic data and background knowledge are stored. The paper pays special attention to background knowledge developed as ontologies for winter wheat cultivation. The results are summarized by drawing some conclusions and recommendations for the design of the local infrastructure of the center and the stored data to improve its performance.

A case study of stand-alone hybrid power systems for a data center using HOMER and DIgSILENT

The number of mega data centers is steadily expanding as the cloud business expands. The data center consumes heavy electrical power because of their structural and equipment characteristics (e.g., workstations and cooling equipment). The mega data center becomes a microgrid (e.g., islanding operation). So, a lot of researchers have proposed the architecture design of a green energy data center to solve the heavy energy consumption problem and to reduce the emission release. For example, the goal of a green energy data center is to maximize energy efficiency while reducing environmental impacts (e.g., the reduction of greenhouse gases). So, the goal of this study is to design the most effective power system for a stand-alone hybrid green data center at the aspects of energy savings and the reliable operation. The ideal architecture of a stand-alone data center powered by renewable energy is initially designed by the HOMER program. Then, the steady-state and transient responses of a data center that functions as an island microgrid using DIgSILENT© Power Factory are examined.

Do accrual models effectively detect earnings management? A study in the Indian context

The cloud's Flexibility and scalability, has significantly altered the architecture, implementation, and processing power of legacy data centers. Microservices are the atom network services that cloud architectures provide. Microservices, in contrast to virtual machines, can be implemented as tiny resource footprint programmer such as container or even lower as unkennels. Field Programmable Arrays (FPGA) boards were first used in servers to effectively offload the execution of computation-intensive applications. FPGAs are increasingly regarded as processing resources that meet cloud standards. FPGAs, however, cannot be used to run several microservices simultaneously in today's cloud data centers. This greatly restricts how effectively microservices can be deployed. We also Used AWS based Architecture to Explore the concept of FPGA in Cloud. Key Words: Microservices, AWS, FPGA, Cloud Computing.

Multi-Objective Optimization Strategy Applied in Data Centers Electrical Subsystems

Data center infrastructures must have high availability, low cost, and high energy efficiency. However, these objectives are often conflicting. For instance, an additional Uninterruptible Power Supply (UPS) improves the system availability but may jeopardize the total cost and the energy consumption. This paper presents a strategy based on the Non-dominated Sorting Genetic Algorithm II (NSGA-II), a multi-objective evolutionary optimization algorithm, to improve the design of electrical data center architectures. To show the applicability of the proposed strategy, we present a comparative study between the brute force algorithm and proposed strategy based on NSGA-II, as it has shown promising results in multi-objective problems 1, 2 and 3. When setting complex electrical infrastructure data center models, the results showed that applying the proposed strategy reduces the runtime 961 times and achieves Pareto optimal curvature with a difference of approximately 1%.

Advanced photonic integrated circuit building blocks for reconfigurability in hyperscale data centres - INVITED

In this talk we introduce two new European projects DYNAMOS and ADOPTION, which will develop advanced integrated photonic technologies for WDM hyperscale data centre architectures including a novel stamped metallic micro-mirror array for advanced PIC-to-fibre coupling with the potential to dramatically reduce PIC design and assembly costs. We also discuss international standardization efforts for these corresponding optical interconnect technologies for future hyperscale data centre, HPC and 6G and Quantum environments.

On the Complexity of Configuration and Orchestration for Enabling Disaggregated Server Provisioning in Optical Composable Data Centres

Due to the limitations of traditional data center (DC) architectures, the concept of infrastructure disaggregation has been proposed. DC resources are separated into multiple blades to be exploited independently. As a result, composable DC (CDC) infrastructures are achieved, enhancing the modularity of resource provisioning. However, disaggregation introduces additional challenges that need to be carefully analyzed. One relates to the potential complexity increase on the orchestration and infrastructure configuration that need to be performed when provisioning resources to support services. This aspect is highly influenced by the distribution of resources at the physical infrastructure. As such, when analyzing the performance of a CDC, it becomes essential to also study the related operational complexity of the resource orchestration and configuration phases. Furthermore, the requirements of several tenant services may impose heterogeneous deployments over the shared physical infrastructure in the form of either disaggregated single-server or multi-server distributions. The associated orchestration/configuration cost is again highly influenced by the data plane architecture of the CDC. With these aspects in mind, in this paper, we provide a methodology for analysis of the complexity of resource orchestration for a service deployment and the associated configuration cost in optical CDCs, considering various service deployment setups. A selected set of CDC architectures found in the literature is employed to quantitatively illustrate how the data plane design and service deployment strategies affect the complexity of infrastructure configuration and resource orchestration.

An Adaptable and Agnostic Flow Scheduling Approach for Data Center Networks

Cloud applications have reshaped the model of services and infrastructure of the Internet. Search engines, social networks, content delivery and retail and e-commerce sites belong to this group of applications. An important element in the architecture of data centers where these applications run is the communication infrastructure, commonly known as data center networks (DCNs). A critical challenge DCNs have to address is the processing of the traffic of cloud applications, which due to its properties is essentially different to the traffic of other Internet applications. In order to improve the responsiveness and throughput of applications, DCNs should be able to prioritize short flows (a few kB) over long flows (several MB). However, given the time and space variations the traffic presents, the information about flow sizes is not available in advance in order to plan the flow scheduling. There has been a wealth of solutions developed in this space, and prior work includes flow scheduling mechanisms optimizing for a specific workload but fall short when workloads are not known in advance, or comprise a collection of applications changing dynamically. In this paper, we present an adaptable mechanism called Adaptable Workload-Agnostic Flow Scheduling (AWAFS). It is an adaptable approach that can adjust in an agnostic way the scheduling configuration of DCN forwarding devices. This agnostic adjustment contributes to reduce the Flow Completion Time (FCT) of those short flows, representing around 85% of the traffic handled by cloud applications. AWAFS operates by observing the traffic and detecting statistical properties that provide a hint to adapt the scheduling parameters. Our evaluation results based on simulation show that AWAFS can reduce the average FCT of short flows between 16.9% and 45.2% when compared to the best existing agnostic non-adaptable solution, without inducing starvation on long flows. Indeed, it can provide improvements as high as 39% for long flows. Additionally, AWAFS can improve the FCT for short flows in scenarios with high heterogeneity in the traffic present in the network, with a reduction up to 5% for the average FCT and 15% for the tail FCT.

An Efficient Location-Based Forwarding Strategy for Named Data Networking and LEO Satellite Communications

Low Earth orbit (LEO) satellite constellations are increasingly gaining attention as future global Internet providers. At the same time, named data networking (NDN) is a new data-centric architecture that has been recently proposed to replace the classic TCP/IP architecture since it is particularly well suited to the most common usage of the Internet nowadays as a content delivery network. Certainly, the use of NDN is especially convenient in highly dynamic network environments, such as those of next LEO constellations incorporating inter-satellite links (ISL). Among other native facilities, such as inbuilt security, NDN readily supports the mobility of clients, thus helping to overcome one of the main problems raised in LEO satellite networks. Moreover, thanks to a stateful forwarding plane with support for multicast transmission and inbuilt data caches, NDN is also able to provide a more efficient usage of the installed transmission capacity. In this paper, we propose a new location-based forwarding strategy for LEO satellite networks that takes advantage of the knowledge of the relative position of the satellites and the grid structure formed by the ISLs to perform the forwarding of NDN packets. So, forwarding at each node is done using only local information (node and destination locations), without the need of interchanging information between nodes, as is the case with conventional routing protocols. Using simulation, we show that the proposed forwarding strategy is a good candidate to promote the efficient and effective future use of the NDN architecture in LEO satellite networks.

Integrating the BIDS Neuroimaging Data Format and Workflow Optimization for Large-Scale Medical Image Analysis.

A robust medical image computing infrastructure must host massive multimodal archives, perform extensive analysis pipelines, and execute scalable job management. An emerging data format standard, the Brain Imaging Data Structure (BIDS), introduces complexities for interfacing with XNAT archives. Moreover, workflow integration is combinatorically problematic when matching large amount of processing to large datasets. Historically, workflow engines have been focused on refining workflows themselves instead of actual job generation. However, such an approach is incompatible with data centric architecture that hosts heterogeneous medical image computing. Distributed automation for XNAT toolkit (DAX) provides large-scale image storage and analysis pipelines with an optimized job management tool. Herein, we describe developments for DAX that allows for integration of XNAT and BIDS standards. We also improve DAX's efficiencies of diverse containerized workflows in a high-performance computing (HPC) environment. Briefly, we integrate YAML configuration processor scripts to abstract workflow data inputs, data outputs, commands, and job attributes. Finally, we propose an online database-driven mechanism for DAX to efficiently identify the most recent updated sessions, thereby improving job building efficiency on large projects. We refer the proposed overall DAX development in this work as DAX-1 (DAX version 1). To validate the effectiveness of the new features, we verified (1) the efficiency of converting XNAT data to BIDS format and the correctness of the conversion using a collection of BIDS standard containerized neuroimaging workflows, (2) how YAML-based processor simplified configuration setup via a sequence of application pipelines, and (3) the productivity of DAX-1 on generating actual HPC processing jobs compared with earlier DAX baseline method. The empirical results show that (1) DAX-1 converting XNAT data to BIDS has similar speed as accessing XNAT data only; (2) YAML can integrate to the DAX-1 with shallow learning curve for users, and (3) DAX-1 reduced the job/assessor generation latency by finding recent modified sessions. Herein, we present approaches for efficiently integrating XNAT and modern image formats with a scalable workflow engine for the large-scale dataset access and processing.

Moving Target Defense for Cyber–Physical Systems Using IoT-Enabled Data Replication

This article proposes a novel moving target defense (MTD) strategy that leverages the versatility of the Internet of Things (IoT) networks to enhance the security of cyber–physical systems (CPSs) by replicating relevant sensory and control signals. The replicated data are randomly selected and transmitted to create two layers of uncertainties that reduce the ability of adversaries to launch successful cyberattacks, without affecting the performance of the system in a normal operation. The theoretical foundations of designing the IoT network and optimal allocation of replicas per signal are developed for linear-time-invariant systems, and fundamental limits of uncertainties introduced by the framework are calculated. The orchestration of the layers and applications integrated in the proposed framework is demonstrated in experimental implementation on a real-time water system over a WiFi network, adopting a data-centric architecture. The implementation results demonstrate that the proposed framework considerably limits the impact of false-data-injection attacks, while decreasing the ability of adversaries to learn details about the physical system operation.

Real-time temperature predictions via state-space model and parameters identification within rack-based cooling data centers

Current rack-based cooling architecture of data centers (DCs) is a promising method since it simplifies airflow distribution and provides fast cooling regulation. Real-time on-demand control of the cooling system is an effective way to improve its operational energy efficiency without sacrificing the thermal security of IT equipment. Accurate and fast temperature distribution prediction serves as one of the bases for ensuring the superior performance of advanced control algorithms. Thus, this study proposed a novel grey-box state-space model, to rapidly predict the dynamic temperature distribution for rack-based cooling DCs. Various heat transfer physics in the rack-based cooling DCs, including heat production caused by servers and heat movements by airflows, were modeled as a state-space structure using the zonal modeling approach. The coefficient matrices were identified through the prediction-error method (PEM), in order to avoid the extremely time-consuming process of obtaining accurate physical parameters regarding the system. This developed model was validated with an experimentally validated mechanistic model and computational fluid dynamics (CFD) simulations. Additionally, the impact of the prediction horizon's size and IT workload transient changes on the proposed model's prediction accuracy were investigated as well. Through simulation, the developed model achieves sufficient accuracy with an average root mean square error (RMSE) equal to 0.19 °C and less than 3% relative error for predicting 1-min dynamic temperature evolutions. Also, the developed model shows satisfying performances for the long-horizon prediction and transient scenario, which will facilitate advanced control techniques for DC cooling systems.

Tools Towards the Sustainability and Circularity of Data Centers.

We are living in an age when data centers are expanding, require abundant spaces, and are an integral part in the urban communities, using massive amounts of environmental resources, and remains in the foreseeable future as the primary driver of the global energy consumption. This demand is disruptive and at times of both peril and opportunity due to impacts such as the COVID-19 pandemic, which is altering the demand of digital infrastructure around the world. With the global call for zero carbon emissions, there needs to be solutions put in place for the de-carbonization of data centers. New innovations are made available, which will have an economic, social, and environmental impact on data centers. Concepts such as circular economy and fourth industrial revolution technologies are useful procedural tools that can be used to systematically analyze data centers, control their mining and critical raw materials, can be utilized in the transition towards a sustainable and circular data center, by objectively assessing the environmental and economic impacts, and evaluating alternative options. In this paper, we will look at the current research and practice, the impact on the United Nations Sustainable Development goals, and look at future strides being taken towards more sustainable and circular data centers. We had discovered that decreasing the environmental effect and energy consumption of data centers is not sufficient. When it comes to data center architecture, both embodied and operational emissions are critical. Data centers also have a vital societal role in our daily lives, enabling us to share data and freely communicate via social media, transacting on the blockchain with cryptocurrencies, free online education, and job creation. As a result, sustainability and efficiency measures have expanded in a variety of ways, including circularity and its associated tools, as well as newer technologies.

Smart Vision Enabled Low Cost Autonomous Robot

Need for security and human resources for monitoring are growing nowadays. Current security systems are static in nature, lack analysis and prone to several threats and attacks. Alternative proctoring robots are hardwired, cannot reuse and expensive to implement making it unfit for Industrial and Public uses. Therefore, we propose a low-cost, re-configurable autonomous robot system for security and monitoring purposes. The robot system architecture is inspired from cloud data centre architecture where the applications are sandboxed and virtualized for efficient utilization of resources. The AI Model, Source Code, Executable scripts, internal resources are contained as a docker container. These containers are called as modules which are connected in a loosely coupled format. Modules can be replaced, added, deleted, updated, scaled over within the robot. Modules are classified as functional and auxiliary where functional modules performs AI operations, analysis and auxiliary modules performs remote results streaming, recording data also backing up footage and data to private or public cloud. The autonomous machine’s camera is virtualized for simultaneous camera access by modules and to reduce computational overhead. As the resource utilization is optimized the power consumption is also reduced with combined efficiency of ARM and RISC-V chipsets. Thus, with this configurable, power efficiency, autonomous robot we hope to improve the quality of life and standards in public and industrial work places.

Architecture of virtual edge data center with intelligent metadata service of a geo-distributed file system

Metadata service of geo-distributed file systems plays an important role in many web-based applications, such as supply-chain management, process mining from log files, intelligent aggregation and collaboration among business processes deployed over Industrial Internet, etc. The edge computing paradigm provides an ideal infrastructure for some value-added services by using geo-distributed file systems. However, web-based applications among the geo-distributed file systems often need an intelligent metadata service mechanism for low-latency data visiting and processing. Unfortunately, it is often a challenge in practice, as the edge computing paradigm often lacks an edge data center that is indispensable for metadata storage and metadata operation. In view of this challenge, an architecture of a virtual edge data center with intelligent metadata service is investigated, in this paper, for taking full advantage of the geo-distributed file systems. Concretely, a prototype of the virtual edge data center is proposed by dynamically gathering the idle storage capability of edge servers into a virtual resource pool. Then, its intelligent metadata service mechanism is investigated The experimental results demonstrate the feasibility of the architecture and show the superior performances with comparison analysis.