In-Depth Research Report on XSKY's Strategic Transformation in the AI Era

Based on the collected information, I now provide you with an

.

---

XSKY is a leading Chinese provider of Software-Defined Storage (SDS) solutions. Founded in 2015 with its headquarters in Beijing, the company was established by an industry veteran team focusing on the data infrastructure sector, and has received multiple rounds of financing from well-known investment institutions such as Qiming Venture Partners [1]. After a decade of development, XSKY has grown from an early-stage software-defined storage vendor to a comprehensive data infrastructure enterprise covering enterprise-level storage, data management, and cloud services.

Previously positioned as an infrastructure provider helping enterprises build a “digital library”, the company’s main products included IT infrastructure products such as distributed storage systems and software-defined storage platforms. With the rapid development of artificial intelligence technology, the company realized that traditional storage architectures could no longer meet the new data processing requirements of the AI era, thus initiating its strategic transformation.

In December 2025, XSKY held the AIMesh Product Strategy Launch Conference, officially announcing the company’s strategic leap from “Information Technology (IT)” to “Data Intelligence” [2]. This strategic upgrade marks a fundamental shift in the company’s positioning:

• Position Upgrade
  : From a traditional “storage vendor” to an “AI data infrastructure builder”
• Value Proposition Shift
  : From providing storage hardware/software to delivering “data factory” operation capabilities for the AI era
• Strategic Vision
  : To be the “second brain” for AI and become a core component provider in the era of data intelligence

At the strategy launch conference, the company’s CEO emphasized that in the face of the uncertain cycle of rapid iteration of algorithms and chips, XSKY chose data itself as the unchanging anchor point, to protect data value in a long-term, secure, and efficient manner, and practice the concept of “Everlasting Data” [2].

---

AIMesh is a full-stack AI data solution launched by XSKY, consisting of three core products [2]:

Product Name	Functional Positioning	Core Value
MeshFS	Training Data Network	File system-level optimization, 50% higher write bandwidth than mainstream solutions
MeshSpace	Global Object Network	Supports EB-level global namespace, breaks through hybrid cloud data silos
MeshFusion	Inference Memory Network	Uses local SSD to simulate L3 persistent memory, expands context window at 1% of the cost

The AIMesh solution has achieved breakthrough innovations addressing three major technical bottlenecks in the AI field:

Traditional AI inference faces the bottleneck of video memory capacity limitations, and the cost of expanding the context window for large models is extremely high. Through innovative persistent memory technology, MeshFusion uses local SSDs to simulate L3 persistent memory, achieving:

• Cost Advantage
  : Expands AI context window at 1% of the cost of traditional solutions
• Performance Improvement
  : Enables AI to have “ultra-long short-term memory” capability
• Technical Principle
  : Through data tiering and intelligent caching strategies, cold data is offloaded to SSDs while hot data is retained in memory

Enterprises commonly face the problem of data silos in AI applications, where data across public clouds, private clouds, and edge nodes is difficult to interconnect. MeshSpace provides:

• EB-Level Global Namespace
  : Supports unified management of massive data volumes
• Hybrid Cloud Data Mobility
  : Data can flow freely between different cloud environments without migration
• Breaking Silos
  : Unified management of data resources scattered in different locations

AI training requires continuous, high-speed data supply, and traditional storage architectures often become a bottleneck for GPU computing. MeshFS achieves this through:

• Write Bandwidth Optimization
  : 50% higher write bandwidth than mainstream solutions
• Reducing GPU Idle Time
  : Ensures ultra-fast data supply to maximize GPU utilization
• File System-Level Optimization
  : Optimized specifically for AI workload characteristics

The AIMesh solution has the following core differentiation features [2]:

1. Decoupled Architecture
   : Separates computing and storage to adapt to the elastic demands of AI workloads
2. Open Ecosystem
   : Connects to various training/inference platforms via standardized interfaces
3. Chip Compatibility
   : Compatible with NVIDIA, Huawei Ascend, and self-developed chips, maintaining absolute neutrality
4. Data Sovereignty
   : Enables enterprises to retain data sovereignty, eliminating the need to reconstruct data pipelines when switching computing power

---

According to data from authoritative market research institutions, the global AI-driven storage market is in a period of high growth [3][4]:

Indicator	2024	2032 Forecast	Compound Annual Growth Rate (CAGR)
AI-Driven Storage Market Size	USD 28.69 Billion	USD 162.65 Billion	24.22%
Network-Attached Storage (NAS) Market	USD 46.97 Billion (2025)	USD 173.12 Billion (2034)	15.50%
Enterprise Data Storage Market	USD 88.33 Billion (2026)	USD 119.53 Billion (2035)	3.4%

:

• The AI-driven storage market is growing significantly faster than the traditional storage market, reflecting the rigid demand for storage driven by AI
• Network-Attached Storage (NAS) accounts for the largest share of 46% in the AI-driven storage market
• The Asia-Pacific region is expected to become the fastest-growing regional market

The core factors driving the growth of the AI data infrastructure market include:

1. Explosion of Large Model Parameters
   : AI large model parameters are moving from the 100-billion level to the trillion level, and training data volume has jumped from the TB level to the PB level [5]
2. Storage “Super Cycle”
   : Unlike previous cyclical fluctuations, this round of storage price increases is substantially driven by rigid AI demand [5]
3. Accelerated Cloudification Trend
   : According to Gartner forecasts, by 2025, over 95% of new digital data workloads will be migrated to cloud platforms
4. Rise of Edge AI
   : The edge-side AI market is developing rapidly, leading to a surge in demand for dedicated storage chips

• Demand Characteristics
  : Ultra-large scale, high throughput, continuous writing
• Market Size
  : With the growth of large model parameters, demand for training data storage is growing exponentially
• Technical Requirements
  : PB/EB-level scalability, 10-million-level IOPS

• Demand Characteristics
  : Low latency, high concurrency, real-time response
• Market Size
  : Data access volume in the inference phase far exceeds that in the training phase
• Technical Requirements
  : Millisecond-level response, memory-level access speed

• Demand Characteristics
  : Multi-source data integration, cross-cloud data mobility
• Market Size
  : The problem of enterprise data silos has spawned demand for unified data management
• Technical Requirements
  : Global namespace, intelligent data tiering

---

The main players in the global AI-driven storage market include [3][4]:

Enterprise	Type	Core Advantages	AI Strategic Focus
Dell Technologies	International Giant	Enterprise-level IT infrastructure	Jointly developing AI data platforms with NVIDIA
NetApp	International Giant	Data management, cloud integration	Launched NetApp AFX and AI Data Engine (AIDE)
Pure Storage	International Innovator	All-flash arrays	FlashBlade//EXA high-performance storage platform
HPE	International Giant	Enterprise-level servers and storage	AI-optimized storage solutions
Huawei	Chinese Leader	Local market, chip integration	OceanStor A800 (EB-level NAS)

The Chinese AI data infrastructure market presents the following competitive characteristics:

1. Leading Enterprise Dominance
   : Huawei, Alibaba Cloud, Tencent Cloud, etc., occupy the major market share
2. Segmented Market Opportunities
   : There is room for differentiated competition in segmented fields such as software-defined storage and distributed storage
3. Domestic Substitution Trend
   : Driven by independent and controllable policies, domestic storage vendors have gained more opportunities
4. Ecosystem Integration Capability
   : Enterprises capable of integrating computing power, storage, and optical communication have more competitive advantages

:

Dimension	Competitive Advantage
Local Market Understanding	Deeply rooted in the Chinese enterprise market, with better understanding of local customer needs
Technological Architecture Innovation	Innovative architectural design of AIMesh targeting AI pain points
Ecosystem Neutrality	Maintains neutrality in chips and platforms, not bound to specific vendors
Cost-Effectiveness Advantage	Cost advantages brought by localized operations
Focused Specialization	Focused on the segmented field of data infrastructure

---

• Large model training and inference pose new requirements for data storage
• Traditional storage architectures cannot meet the characteristics of AI workloads
• There is a huge market gap for specialized AI storage solutions

• Enterprises’ demand for data security and independent controllability has increased
• Chip and platform neutrality have become important considerations
• Domestic storage vendors have gained more trust and opportunities

• Enterprise IT architectures are evolving towards hybrid cloud and multi-cloud
• Demand for cross-cloud data management and mobility has increased
• A unified data infrastructure has become a rigid demand

• The “Digital China” strategy promotes the construction of data infrastructure
• Independent and controllable policies benefit domestic vendors
• The development of the data factor market brings new opportunities

• AI technology is evolving rapidly, requiring continuous adjustment of product roadmaps
• Changes in chip architectures may affect storage architecture design
• There is uncertainty in technology roadmap selection

• International giants have profound technological accumulation and strong brand influence
• Domestic leading enterprises have abundant resources, leading to fierce price competition
• New entrants continue to increase

• Transformation from product sales to solution services
• Need to establish new sales and service capabilities
• Customer education and market cultivation take time

• Strategic transformation requires large R&D investment
• Talent competition is fierce, leading to increased costs for core talents
• Market expansion requires continuous capital support

The key factors for the success of XSKY’s strategic transformation include:

Key Factor	Specific Requirements
Technological Innovation	Sustained R&D investment to maintain technological leadership
Ecosystem Construction	Establish an open partner ecosystem
Customer Expansion	In-depth understanding of AI customer needs and provision of customized solutions
Brand Building	Establish a professional brand in the field of AI data infrastructure
Talent Reserve	Attract and cultivate talents in the interdisciplinary field of AI and storage

---

The financial characteristics of the AI data infrastructure industry include:

1. High R&D Investment
   : Driven by technological innovation, requiring continuous R&D investment
2. Economies of Scale
   : Software-defined storage has favorable marginal cost characteristics
3. Customer Stickiness
   : Enterprise-level storage customers have high switching costs, resulting in strong customer stickiness
4. Cash Flow Characteristics
   : Subscription models and long-term contracts improve cash flow

For transforming AI data infrastructure enterprises like XSKY, valuation should consider:

Valuation Dimension	Analysis Points
Revenue Growth	Growth rate of AI product revenue as a percentage of total revenue
Gross Profit Margin	Improvement in gross profit margin driven by increased software revenue share
Customer Acquisition	Number of AI customers and customer lifetime value
Technological Barriers	Uniqueness and sustainability of core technologies
Market Position	Market share in segmented fields

For investors, key areas of focus include:

1. Commercialization Progress of AIMesh
   : Order volume, number of customers, revenue contribution
2. R&D Investment Efficiency
   : Efficiency of converting technological achievements into revenue
3. Customer Structure Changes
   : Growth in the proportion of AI customers
4. Competitive Landscape Changes
   : Market share changes, consolidation of competitive advantages
5. Capital Operations
   : Listing plans, financing progress

---

XSKY’s strategic transformation reflects its profound understanding of data infrastructure requirements in the AI era. The launch of the AIMesh product marks the company’s upgrade from a traditional storage vendor to an AI data infrastructure builder, which aligns with industry development trends. The company’s technological accumulation in the software-defined storage field and its local market advantages have laid a solid foundation for its AI transformation.

:

1. High-Quality Track
   : The AI data infrastructure market is growing rapidly, with an expected CAGR of 24.22% from 2024 to 2032
2. Timely Transformation
   : The company launched targeted products during the period of explosive AI demand to seize market opportunities
3. Differentiated Competition
   : Neutral positioning and local advantages form differentiated competitiveness
4. Policy Benefits
   : The domestic substitution trend and independent and controllable policies provide development opportunities

Given that XSKY is not yet publicly listed, it is recommended to pay attention to its listing progress on the Hong Kong Stock Exchange or A-share market. For primary market investment, a reasonable valuation range should consider:

• Revenue Multiple
  : Refer to the revenue multiples of similar AI infrastructure companies (5-10x PS)
• Technology Premium
  : A 20-30% technology premium can be given for the AI concept
• Growth Discount
  : A certain growth discount can be given to unprofitable companies

Risk Type	Risk Description	Impact Level
Market Risk	AI infrastructure construction progress falls short of expectations	Medium-High
Competition Risk	Leading vendors increase investment to squeeze market space	Medium-High
Technological Risk	Wrong technology roadmap selection leads to reduced product competitiveness	Medium
Execution Risk	Strategic transformation execution falls short of expectations	Medium
Financing Risk	Changes in primary market financing environment affect development	Medium-Low

1. AIMesh receives orders from key customers
2. Announcement of listing plans on the Hong Kong Stock Exchange or A-share market
3. Introduction of favorable AI-related policies
4. Securing major strategic investment
5. Quarterly/annual performance exceeds expected growth

---

XSKY’s strategic transformation reflects the proactive adaptation of data infrastructure enterprises in the AI era. The company’s upgrade from a traditional “storage vendor” to an “AI data infrastructure builder” aligns with the new requirements for data infrastructure in the AI era.

:

1. Correct Strategic Direction
   : The AIMesh product proposes solutions to three major pain points in the AI field (memory wall, data wall, IO wall), which aligns with market demand
2. Obvious Product Differentiation
   : Decoupled, open, and durable technological architecture, as well as neutral positioning, form differentiated advantages
3. Huge Market Opportunities
   : The global AI-driven storage market is expected to maintain a high growth rate of over 24%
4. Coexistence of Challenges and Opportunities
   : Facing competition from leading vendors and technology iteration risks, but the domestic substitution trend provides a development window

:

• Achieve commercial breakthroughs for the AIMesh product
• Establish benchmark cases with AI customers
• Complete listing preparation work

:

• Significantly increase the proportion of AI-related revenue
• Establish a leading position in the AI data infrastructure field
• Achieve scale and profit improvement

:

• Become a core data infrastructure provider in the AI era
• Achieve international expansion
• Build a complete data intelligence ecosystem

For investors focusing on the AI data infrastructure track, XSKY is a target worth tracking. It is recommended to pay attention to:

1. The company’s listing progress and valuation level
2. Commercialization data of the AIMesh product
3. Changes in customer structure and acquisition of key customers
4. R&D investment and technological innovation dynamics
5. Changes in the industry competitive landscape

---

[1] Qiming Venture Partners - XSKY: SDS Industry Accelerates Data Infrastructure Construction (https://www.qimingvc.com/sites/default/files/news/)

[2] NetEase News - XSKY Launches AIMesh Full-Stack AI Data Solution: To Be the “Working Memory” Infrastructure for AI (https://www.163.com/dy/article/KJAG7J210514R9OJ.html)

[3] Credence Research - AI-Driven Storage Market Size, Share, Growth and 2032 Forecast (https://www.credenceresearch.com/zh/report/ai-aowere-storage-market-zh)

[4] Fortune Business Insights - Network-Attached Storage Market Size, Share and Analysis (https://www.fortunebusinessinsights.com/zh/industry-reports/network-attached-storage-market-100505)

[5] East Money - CC Data: Collaboration of Computing Power, Storage and Optical Communication to Reshape the New Pattern of AI Infrastructure (https://emcreative.eastmoney.com/app_fortune/article/index.html?artCode=20260111204655726166970)

---

Report Compiled by: Jinling AI Investment Research Team
Date: January 15, 2026