The global data center fabric architecture market is projected to reach USD XX million by 2033, exhibiting a CAGR of XX% during the forecast period. This growth is attributed to the increasing demand for high-speed and reliable connectivity in data centers, driven by the proliferation of cloud computing, big data analytics, and artificial intelligence. Additionally, the growing adoption of virtualization and software-defined networking (SDN) technologies is further fueling the market growth. Key market trends include the increasing deployment of cloud-based data center fabric architectures, the adoption of open source platforms, and the integration of artificial intelligence (AI) and machine learning (ML) technologies. Cloud-based architectures offer scalability, flexibility, and cost-effectiveness, making them a preferred choice for enterprises. Open source platforms provide greater customization and control, enabling organizations to tailor their data center fabric architectures to specific requirements. AI and ML technologies enhance network management and optimization, improving efficiency and reducing operational costs. Major players in the market include IBM Corporation, TIBCO Software, Cisco Systems, Hewlett Packard Enterprise, Unisys, Avaya, Atos, CA Technologies, Oracle Corporation, Microsoft Corporation, and Dell.

The global data center fabric software market is projected to grow from $5.7 billion in 2025 to $15.6 billion by 2033, at a CAGR of 12.5%. This growth is driven by the increasing adoption of cloud computing, enterprise network architecture, and artificial intelligence. Additionally, the rise of big data and the Internet of Things (IoT) is fueling the demand for data center fabric software that can manage and optimize the flow of data within data centers. The market for data center fabric software is expected to face challenges from the complexity of data center networks and the need for interoperability between different vendors' solutions. However, the growing adoption of open source and software-defined networking (SDN) technologies is expected to mitigate these challenges and drive further growth in the market. This report provides a comprehensive analysis of the global data center fabric software market, focusing on concentration, trends, key segments, and growth drivers.

The Fabric Data Center market is experiencing robust growth, projected to reach $2.31 billion in 2025 and maintain a Compound Annual Growth Rate (CAGR) of 28.90% from 2025 to 2033. This expansion is driven by the increasing demand for high-bandwidth, low-latency network infrastructure to support cloud computing, big data analytics, and the Internet of Things (IoT). Businesses across various sectors, including IT & Communications, Banking & Financial Services, and Healthcare, are adopting fabric data centers to improve agility, scalability, and operational efficiency. The shift towards software-defined networking (SDN) and network function virtualization (NFV) is further fueling market growth, enabling greater automation and centralized management of network resources. Key players like Cisco, Huawei, and Arista Networks are leading the innovation, offering advanced solutions that cater to evolving enterprise needs. The market segmentation reveals strong growth across solutions (routers, switches, storage area networking), applications (IT & Communications, Banking & Financial Services), and end-users (cloud service providers, telecom service providers). The Asia-Pacific region is expected to witness particularly rapid expansion due to increasing digitalization and infrastructure investments. However, despite the significant growth trajectory, the market faces certain challenges. High initial investment costs associated with implementing fabric data center solutions can be a barrier for some organizations, particularly smaller businesses. Furthermore, the complexity of managing and integrating these advanced systems requires skilled personnel, creating a potential talent shortage. Despite these restraints, the long-term benefits of enhanced performance, scalability, and reduced operational costs are expected to outweigh these challenges, ensuring continued market expansion throughout the forecast period. The competitive landscape is marked by both established players and emerging vendors, leading to ongoing innovation and price competition which benefits end-users. Recent developments include: July 2023: Huawei's announced three innovative data center facility solutions as unveiled the next-generation indirect evaporative cooling solution EHU and the mobile intelligent management solution iManager-M. These scenario-based data center solutions promise optimal reliability throughout the lifecycle and aim to drive the high-quality development of the data center industry., December 2022: The Nokia 7220 IXR D2/D3 interconnect routers will be used as core switching datacentre leaf platforms for North's datacentre fabric, running the Nokia SR Linux network operating system (NOS). The data center is built on bare metal servers running OpenStack Ironic, which interfaces with NOS using open-source upstream code., October 2022: Cloudera, the hybrid data startup announced new hybrid data capabilities that will allow enterprises to more easily migrate data, metadata, data workloads, and data applications between clouds and on-premises in order to optimize for performance, cost, and security.. Key drivers for this market are: Increasing Demand for Data Storage and Adoption of Cloud Computing, Need for High Speed Data Transfer; Increasing Demand of Fabric Switches. Potential restraints include: Increasing Demand for Data Storage and Adoption of Cloud Computing, Need for High Speed Data Transfer; Increasing Demand of Fabric Switches. Notable trends are: Increasing Demand of Fabric Switches is Driving the Market.

GPU-Over-IP Fabric Software Market Outlook

According to our latest research, the GPU-Over-IP Fabric Software market size reached USD 1.82 billion in 2024 globally, demonstrating robust momentum driven by surging demand for GPU virtualization across industries. The market is projected to grow at a CAGR of 26.4% from 2025 to 2033, reaching an estimated USD 13.25 billion by 2033. This remarkable growth is primarily attributed to the escalating adoption of high-performance computing (HPC), artificial intelligence (AI), and cloud-based services, all of which require highly efficient and scalable GPU resource management solutions.

One of the primary growth factors propelling the GPU-Over-IP Fabric Software market is the exponential rise in demand for high-performance computing and AI-driven workloads across enterprise and research sectors. As organizations increasingly leverage AI and machine learning for data analytics, simulation, and automation, the need to dynamically allocate GPU resources becomes critical. GPU-Over-IP fabric software enables seamless sharing and pooling of GPU resources over networks, maximizing hardware utilization and reducing capital expenditures. This technology empowers organizations to deliver superior computational performance without the constraints of physical GPU location, thereby supporting remote work, distributed teams, and global operations.

Another significant driver is the rapid expansion of cloud computing and virtualization trends. Enterprises and cloud service providers are shifting towards flexible, scalable infrastructure solutions that can efficiently support diverse workloads. GPU-Over-IP fabric software is instrumental in virtual desktop infrastructure (VDI), gaming, and data center environments, where dynamic allocation and remote access to GPU resources are paramount. This shift is further amplified by the proliferation of cloud-native applications and the growing adoption of hybrid and multi-cloud strategies, which demand robust GPU virtualization and orchestration capabilities to ensure optimal performance and cost efficiency.

The growing emphasis on digital transformation and edge computing is also fueling market growth. As industries such as healthcare, automotive, financial services, and media adopt data-intensive applications, the demand for distributed GPU resources increases. GPU-Over-IP fabric software enables organizations to extend GPU capabilities to edge devices and remote locations, supporting real-time analytics, visualization, and AI inference at the network’s edge. This capability is particularly valuable for industries requiring low-latency processing and scalability, further solidifying the market’s upward trajectory.

From a regional perspective, North America continues to dominate the GPU-Over-IP Fabric Software market, driven by early technology adoption, a strong presence of cloud service providers, and significant investments in AI and HPC infrastructure. Europe and Asia Pacific are witnessing accelerated growth due to expanding digital economies, government initiatives supporting advanced computing, and increased R&D activities. Latin America and the Middle East & Africa are emerging markets with rising adoption, particularly in sectors such as education, research, and digital services. As global enterprises prioritize agility, scalability, and operational efficiency, the regional dynamics of the market are expected to evolve rapidly.

Component Analysis

The GPU-Over-IP Fabric Software market by component is broadly segmented into software and services. The software segment encompasses the core GPU virtualization and orchestration platforms that facilitate the allocation, pooling, and management of GPU resources over IP-based networks. This segment has witnessed substantial growth due to continuous advancements in software capabilities, such as support for heterogeneous GPU environments, enhanced security, and seamless integration with existing IT infrastructure. Leading vendors are investing heavily in R&D to deliver solutions with improved scalability, fault tolerance, and automation, catering to the evolving needs of enterprises, cloud providers, and research institutions. The proliferation of open-source frameworks and APIs has further accelerated innovation, enabling organizations to customize and optimize their GPU resource management strategies.

Meanwhile, the services seg

Anisotropy of magnetic susceptibility (AMS) and anisotropy of remanence magnetization (AARM and AIRM) are efficient and versatile techniques to indirectly determine rock fabrics. Yet, deciphering the source of a magnetic fabric remains a crucial and challenging step, notably in the presence of ferrimagnetic phases. Here we use X-ray micro-computed tomography to directly compare magnetite and amphibole Shape-Preferred Orientation and spatial distribution data to AMS, AARM and AIRM data from five hypabyssal trachyandesite samples. This study thus reports quantitative petrofabric data on magnetite shape and distribution anisotropy on magnetic fabrics in igneous rocks. Our results have first-order implications for the interpretation of petrofabrics using magnetic methods.

Data generated during the study Mattsson et al. 'Decrypting magnetic fabrics (AMS, AARM, AIRM) through the analysis of mineral shape fabrics and distribution anisotropy'

-MicroXCT mineral data extracted from the software Blob3D are given in spreadsheet named after the samples. The data files named TT_XX.xls can be opened with the software Tomofab (Petri et al. 2020). The data include extracted grain volume and XYZ position in mm, XYZ radius length in mm, and XYZ orientation in directional cosines of best-fit ellipsoid of the extracted mineral grain. Text files with the same data are included. -MicroXCT mineral data extracted with Avizo are given in a spreadsheet named 'Avizo_XCT_data'. The data are given as trend and plunge of the extracted grain long axes in geological specimen position. -AMS data are given in Agico (https://www.agico.com/) AMS data files and in csv files. The files are named after sample name. The AMS data files can be opened with the software Anisoft5 (https://www.agico.com/text/software/anisoft/anisoft.php). -Magnetic remanance data are given as Agico (https://www.agico.com/) JR6 data files and the 'JR6_data' csv file. The Jr6 files are named after sample name. Analysis parameters are given in an spreadsheet with the name 'Jr6 Analysis Parameters'. Data can be accessed and analysed with the software Rema6. (https://www.agico.com/text/software/rema6/rema6.php). -Files named Samplename_distance give the distance to the nearest neighbor of magnetite grains. -The 'SampleOrientation' Spreadsheet gives the reorientation parameters for XCT data to geological specimen position. -Files with 'T-X' in its name include temperature vs. magnetic susceptibility data and experiment log. The field 'corrected susceptibility' is corrected for the susceptibility of the sample holder. -Descriptions of headings/variables in the data files are given in the 'variable_codebook' .txt file.

Magnetic and microXCT data were collected on five 21 × 24 mm cores from the trachyandesite samples (CB-15C2, CB-19A1, CB-46A1, CB-55A1 and CB-61B2).

AMS measurements were performed in the Laboratory for Experimental Palaeomagnetism at the Department of Earth Sciences, Uppsala University with an Agico Kappabridge MFK1-FA in semi-automatic spinning mode. A field of 200 A/m and frequency of 976 Hz were used for the measurements.

The thermomagnetic properties (T-X) of the samples were determined by measuring the bulk magnetic susceptibility on powders of the five samples in three steps using the CS-4 attachment of the KLY-5 kappabridge at the University of St. Andrews M3Ore Lab. The samples were first cooled to -194 °C and the susceptibility of the samples was measured until the temperature reached 0 °C. The samples were then heated at a rate of 12 °C/min in an argon atmosphere from room temperature to 700°C and then cooled back to room temperature. The heated sample was again cooled to -194 °C and the bulk magnetic susceptibility was measured until the sample reached 0 °C.

AARM and AIRM measurements were performed with the JR6-a spinner magnetometer at the University of St. Andrews M3Ore Lab. Samples were demagnetized with AGICO’s LDA5 and PAM1 instruments and remanence measured in the JR6-a spinner magnetometer in a near zero field space. Analysis parameters are given in the Excel spreadsheet with the same name. During IRM acquisition samples were magnetized in fields above 20 mT with a MMPM 10 pulse magnetizer.

In order to assess the rock fabric, five 21 × 24 mm cores from the five samples (CB-15C2, CB-19A1, CB-46A1, CB-55A1 and CB-61B2) were imaged by X-ray computed microtomography (µXCT or µCT). The cores were scanned with a Nikon Metrology XT H 225 ST X-ray microtomograph at the Natural History Museum, University of Oslo. µXCT analyses was conducted using a 140 kV acceleration voltage, a current of 300 µA, 1 s exposure time and 3016 rotational projections and using a 0.25 mm copper filter. The X-rays transmitted through the sample were collected on a planar 1920×1536 pixels detector.

The sample volume was reconstructed using the software Blob3D (Ketcham, 2005); the resulting voxel (volume pixel) size was about 16 µm3 (see Table S1). The obtained stack of 1534 grayscale images on each core represent the attenuation of the X-rays in the scanned volume, i.e., phases of higher densities have lighter grayscale voxels and lower density phases have darker voxels. Magnetite and amphiboles have comparatively higher densities than plagioclase phenocrysts and the groundmass and can therefore easily be distinguished in the scan slices. Beam-hardening effects are visible on the scan slices and is most distinct 1.5 to 2 mm from the edge of the core. Beam-hardening had no effect on magnetite segmentation due to its high attenuation of X-rays, however amphibole segmentation was strongly affected. As a consequence, a 1.5 to 2 mm rim of the scanned core was segmented as a single blob and removed from the sample set during the manual separation of the data.

In order to assess the rock fabric, five 21 × 24 mm cores from the five samples (CB-15C2, CB-19A1, CB-46A1, CB-55A1 and CB-61B2) were imaged by X-ray computed microtomography (µXCT or µCT). The cores were scanned with a Nikon Metrology XT H 225 ST X-ray microtomograph at the Natural History Museum, University of Oslo. µXCT analyses was conducted using a 140 kV acceleration voltage, a current of 300 µA, 1 s exposure time and 3016 rotational projections and using a 0.25 mm copper filter. The X-rays transmitted through the sample were collected on a planar 1920×1536 pixels detector.

The sample volume was reconstructed using the software Blob3D (Ketcham, 2005); the resulting voxel (volume pixel) size was about 16 µm3 . The obtained stack of 1534 grayscale images on each core represent the attenuation of the X-rays in the scanned volume, i.e., phases of higher densities have lighter grayscale voxels and lower density phases have darker voxels. Beam-hardening is visible on the scan slices and is most distinct 1.5 to 2 mm from the edge of the core. Beam-hardening had no effect on magnetite segmentation due to its high attenuation of X-rays, however amphibole segmentation was strongly affected. As a consequence, a 1.5 to 2 mm rim of the scanned core was segmented as a single blob and removed from the sample set during the manual separation of the data.

MicroXCT data was processed with Blob3D from X-ray serial sections. For each grain, the volume, center x, y, z coordinates and the length and orientation of the three orthogonal principal axes (long axis; intermediate axis; short axis) were determined using Blob3D by fitting a best-fit ellipsoid to the separated grain volume surfaces.

We validated the accuracy of the phase separation between Blob3D and the commercial software Avizo Fire edition. In Avizo, magnetite and amphibole were separated from the scanned cores using grayscale thresholding in the labelling module. Magnetite grains smaller than 1000 to 2000 voxels and amphibole grains smaller than 3000 to 4000 voxels were removed to limit noise and the erosion tool was employed to remove breakdown rims on crystals and imaging artefacts. No individual review of separated crystal volumes was performed in Avizo, which resulted in fast extraction of data, but the crystal volumes extracted largely represent several crystals in aggregates.

References

Ketcham, R.A., 2005. Computational methods for quantitative analysis of three-dimensional features in geological specimens. Geosphere 1, 32–41. https://doi.org/10.1130/GES00001.1

Petri, B., Almqvist, B.S.G., Pistone, M., 2020. 3D rock fabric analysis using micro-tomography: An introduction to the open-source TomoFab MATLAB code. Computers & Geosciences. https://doi.org/10.1016/j.cageo.2020.104444

Summary
This submission contains three tomographic datasets of a fragment of fabric woven in tapestry weave. The data is collected at three different zoom levels to achieve different reconstructed image resolution.
The data is made available as part of [Bossema 2020].

Apparatus
The dataset is acquired using the custom-built and highly flexible CT scanner, FleX-ray Laboratory, developed by TESCAN-XRE, located at CWI in Amsterdam. This apparatus consists of a cone-beam microfocus X-ray point source that projects polychromatic X-rays onto a 1944-by-1536 pixels, 14-bit, flat detector panel. Full details can be found in [Coban 2020].

Sample Information
The sample is a fragment of woven fabric approximately 7cm x 15cm in size. The fabric was hung vertically on a piece of foam, held on the top by a wooden stick through one of the holes and on the bottom by a piece of plastic tape. See Figure 5 in [Bossema 2020] for a picture of the object and examples of the reconstruction.

Experimental Plan
The data in this submission was collected to illustrate the use of zooming for the investigation of cultural heritage objects. Three region-of-interest scans of the lower part, containing a hole, were collected at different zoom levels. For each scan, the sample was rotated 360° in circular and continuous motion, with a dark-field (closed-shutter), and flat-field (open-shutter) image taken before the acquisition. Each dataset consists of 1200 projections. The source-detector distance was kept at 1098mm. At the first level of zooming, the object was placed at 963mm from the source yielding a magnification of 1.14 and 131 micron resolution. For the second level the object was moved closer to the source so that the source-object distance was 603mm, yielding a magnification of 1.82 and 82 micron resolution. At the third level, the source-object distance was reduced to 243, yielding a magnification of 4.5 and 33 micron resolution.
All raw data (i.e. no corrections) is made available in .tif format.

List of Contents

The content of the submission is given below.

• level1: lowest resolution scan.
• level2: higher resolution scan.
• level3: highest resolution scan.

Each data folder contains:

• dark-field (or closed-shutter) image, di000000.tif,
• flat-field (or open-shutter before acquisition) image, io000000.tif,
• raw (unprocessed or uncorrected) projections, scan_*.tif,
• data settings XRE.txt, a text file with scanner metadata,
• scan settings.txt, a text file with scanner metadata in a human readable format, and
• data settings XRE.ini, a snapshot text file of basic geometry information at the start of a scan.
• script_executed.txt, the text file containing the list of commands the apparatus has executed.

Additional Links

These datasets are produced by the Computational Imaging group at Centrum Wiskunde & Informatica (CI-CWI). For any relevant Python/MATLAB scripts for the FleX-ray datasets, we refer the reader to our group's GitHub page.

Contact Details
For more information or guidance in using these datasets, please get in touch with

• bossema [at] cwi.nl

Acknowledgments
We thank Suzan Meijer of the Rijksmuseum for providing this sample.

The Geospatial Fabric for National Hydrologic Modeling (Viger and Bock, 2014; Bock and others, 2020) is a dataset of hydrographic features and spatial data designed for use within the National Hydrologic Model that covers the conterminous United States (CONUS), Hawaii, and most major river basins that flow in from Canada. This U.S. Geological Survey (USGS) data release consists of the geospatial fabric features and other related spatial datasets created to expand the National Hydrologic Model to Alaska. This child item contains data and information related to the GIS features of the Geospatial Fabric for National Hydrologic Model, Alaska domain. Two Open Geospatial Consortium geopackages are provided: one containing source layers that have had some pre-processing done from their native data formats (Reference_19.gpkg), and one (NHM_19.gpkg) containing 4 final feature layers for the NHM: points of interest (pois), a stream network (nsegment), aggregated catchments (catchments), and hydrologic response units (nhru). Features were derived from the MERRIT Hydro Global Hydrography Dataset.

The Telecom Cloud market size was valued at USD 18.81 USD Billion in 2023 and is projected to reach USD 89.19 USD Billion by 2032, exhibiting a CAGR of 24.9 % during the forecast period. Telecom cloud is a software-defined, highly resilient cloud infrastructure that allows telecommunications service providers (telcos) to add services more quickly, respond faster to changes in network demand, and manage central and decentralized resources more efficiently. It is one of the key foundational components in a successful digital transformation. More specifically, a telco cloud focuses on the creation of a common virtualized infrastructure to manage various network functions required to deliver communications services. Each function now is disaggregated from the hardware, to be operated from a horizontal platform as a virtual network function or cloud-native network function. This function is designed to execute a specific network function such as load balancing or firewalls. Early telco clouds applied virtualization concepts from the datacenter into the networks. With 5G at first, the scope of telco clouds expanded beyond the use of network function virtualization (NFV) and began using newer technologies such as containers and microservices, as well as hybrid cloud architectures. As they are deployed into telco clouds, containers and microservices still have to coexist with older virtualized network functions (VNFs). Recent developments include: February 2024: Dell Technologies announced updated solutions designed to help communications service providers (CSPs) accelerate network cloud and achieve improved economics and agility while keeping network reliability. Dell leverages its years of experience in digital transformation and deep industry partnerships to help CSPs simplify deployment, automate operations, and simplify support and the lifecycle management of distributed network and cloud infrastructure. , June 2023: Nokia Corporation strategically announced a collaboration with Red Hat, Inc. to integrate open-source solutions such as Red Hat OpenShift and Red Hat OpenStack Platform into its network application. The collaboration supports Nokia testing and development, and the Red Hat cloud offers flexibility and deployment options., February 2023: A data cloud company, Snowflake Inc., launched Telecom Data Cloud to deliver industry-specific data insights that can help clients drive enhanced decisions. The new solution is helping modernize the telecommunication network, and operational efficiency helps maximize revenue. , February 2023: Dell Technologies introduced its new collaboration of Dell Telecom Infrastructure Blocks for Red Hat to help sectors with open network architecture. The new solution helps implement 5G technology and its radio access network., February 2023: Google LLC introduced three unified cloud solutions for the telecommunication industry: telecom data fabric, telco subscriber insights, and telecommunication network automation. The new telco cloud products aim to support communication service providers with hybrid cloud principals, helping collect data and improving customer experience.. Key drivers for this market are: Increasing 5G Technology to Drive Market Growth. Potential restraints include: Potential Risk in Data Security to Hamper Market Growth. Notable trends are: Rise of Cloud-native in the Telecommunication Industry to Drive Market Growth.

Market Analysis: Data Observability Market The global data observability market is projected to reach USD 2.37 billion by 2033, exhibiting a CAGR of 12.2% during the forecast period (2025-2033). The increasing adoption of cloud-based technologies, big data analytics, and the Internet of Things (IoT) is driving the market growth. Enterprises are seeking comprehensive solutions to monitor and ensure the reliability, performance, and stability of their complex data pipelines and systems. Key Drivers and Trends The adoption of data observability solutions is driven by the increasing need for data-driven decision-making, compliance with data privacy regulations, and the desire to improve operational efficiency. The growing complexity of data ecosystems, including the use of multiple data sources and technologies, necessitates real-time and comprehensive data monitoring capabilities. Moreover, the trend towards agile data management and DevOps practices is fueling demand for data observability solutions that can provide continuous visibility and insights into the performance of data pipelines and systems. Recent developments include: In July 2024, International Business Machines Corporation, an American multinational technology company acquired StreamSets to expand its data integration capabilities, enabling real-time data pipelines and enhancing IBM Data Fabric's support for AI-driven insights. This acquisition aims to provide robust tools that empower organizations to utilize data effectively in hybrid multi-cloud environments. , In June 2024, Splunk Inc. introduced new data management innovations to provide customers with unified visibility across their enterprise. This includes preprocessing data through a centralized pipeline to optimize data economics and enhance digital resilience, addressing the complexities of managing data growth across diverse IT environments. , In April 2024, the Microsoft Azure Container Networking team introduced Retina, an open-source container networking observability platform designed to provide comprehensive visualization and analysis of workload traffic across diverse environments. Retina utilizes eBPF technology to achieve kernel-level monitoring without requiring container agents, ensuring efficient and scalable network monitoring across Linux and Windows environments. , In April 2024, Tech Mahindra Limited, an Indian multinational information technology services and consulting company, collaborated with Microsoft to launch a unified workbench on Microsoft Fabric. The collaboration focuses on streamlining data workflows, enhancing analytics capabilities, and accelerating business transformation. , In November 2023, Monte Carlo introduced integrations with vector databases and Apache Kafka, enhancing their data observability capabilities to ensure data quality and reliability across complex pipelines, including those supporting generative AI models. These advancements, showcased at Impact 2023, include Performance Monitoring and Data Product Dashboard tools to optimize data pipeline efficiency and monitor data product reliability. .

Hand exoskeleton robots have been developed as rehabilitation robots and assistive devices. Based on the material used, they can be soft or right exoskeleton. Soft materials such as fabric can be used as a component of the wearable robot to increase comfortability. In this paper, we proposed an affordable soft hand exoskeleton based on fabric and motor-tendon actuation for hand flexion/extension motion assistance in daily activities. On-Off control and PI compensator were implemented to regulate the finger flexion and extension of the soft exoskeleton. The controllers were embedded into a microcontroller using Simulink software. The input signal command comes from the potentiometer and electromyography (EMG) sensor to drive the flexion/extension movement. Based on the experiments, the proposed controller successfully controlled the exoskeleton hand to facilitate a user in grasping various objects. The proposed soft hand exoskeleton is lightweight, comfortable, portable and affordable, making it easily manufactured using available hardware and open-source code. The developed soft exoskeleton is a potential assistive device for a person who lost the ability to grasp objects.