Huawei Cloud Software / SaaS China Translated from Chinese

Suwen Technology

Enterprise Knowledge Search

Suwen Technology needed to process large volumes of multilingual market and technology data, keep its AI research platform stable, and make large-model answers fast enough for enterprise users without maintaining a fragmented database stack.

AI model deployment and execution efficiency improved by more than 15x
average response time improved from tens of seconds to seconds
database speed improved by more than 7x

Reported impact

Business outcome

TechAgent simplified its data architecture, moved response times from tens of seconds to seconds, and greatly improved AI execution efficiency for enterprise research and strategy workflows.

  • Productivity Improvement
  • Speed to Market

Solution

Suwen Technology built TechAgent as an enterprise research and strategy platform that combines large-model reasoning with structured market and technology data. The platform continuously collects multi-source data, stores it on GaussDB and OBS instead of a patchwork of self-managed databases, and uses Huawei GPU resources to run LLM methods and custom AI models for NLP, knowledge-graph processing, search, and recommendation. On top of that data and model layer, Pangu model capabilities and NL2SQL let enterprise users ask natural-language questions and receive summarized analysis directly against the underlying data. HSS, Anti-DDoS, WAF, and CBR protect the web-delivered agent service, while the simplified architecture reduces operational burden. The end-to-end flow moves from data ingestion and storage, to model inference and knowledge processing, to interactive enterprise answers that help clients track market changes faster and turn raw industry signals into planning insight.

Data flow

TechAgent gathers multilingual market and technology data into GaussDB and OBS, then sends that data through GPU-backed NLP, knowledge-graph, search, and recommendation models. Pangu and NL2SQL convert user prompts into structured queries and summarized outputs that are returned through the web-delivered agent interface.

Solution architecture

5 components · 3 layers
  1. Compute
    • Huawei Cloud GPU compute resources Run large-language-model methods and custom AI models for NLP, knowledge graphs, search, and recommendation.
  2. Storage
    • GaussDB Stores high-volume multi-source platform data in a unified managed database layer.
    • OBS Holds platform data and backups alongside the core database stack.
  3. Serving
    • TechAgent Provides the enterprise-facing market-research and strategy-analysis interface powered by large models.
    • Pangu model with NL2SQL Turns natural-language enterprise questions into domain-specific analysis and database interactions.

Architecture clues

  • Pangu NL2SQL is used to let users interact with data through natural language.
  • The SaaS service is protected by HSS, Anti-DDoS, WAF, and CBR.
  • The platform is designed for multilingual market and technology analysis.
  • The platform runs NLP, knowledge-graph, search, and recommendation workloads on Huawei GPUs.

Evidence from the source

"Huawei GPU resources improved deployment and runtime efficiency for LLM methods and custom AI models by more than 15 times. (original: TechAgent每天需要运行大量大语言模型方法和不同自研AI算法模型... 使用华为云GPU计算资源以及相应配套服务,可以大幅提升AI算法模型部署上线和运行效率,平均执行效率提升15倍以上。)"
"Pangu NL2SQL lets users interact with databases in natural language and greatly improves data-processing efficiency and convenience. (original: 素问科技结合盘古大模型NL2SQL能力,使得用户能够通过自然语言直接与数据库交互,极大地提高了数据处理的效率和便捷性。)"
"TechAgent provides personalized market and technology data acquisition and analysis, and uses large-model capability to summarize and interact with that information. (original: TechAgent(素问科技)专注于为企业打造智能市场调研和战略分析平台。TechAgent平台可实现企业个性化的市场和技术数据的获取及分析,通过大模型能力实现信息的总结和交互。)"
"The optimized architecture improved average response time from tens of seconds to seconds. (original: 通过优化,系统的平均响应时间从数十秒大幅缩短至秒级。)"
"Using Huawei GaussDB and OBS simplified database management and raised performance more than sevenfold versus the self-built database stack. (original: 使用华为GuassDB数据库和对象存储OBS后,可以统一数据库管理备份,降低研发人员业务操作复杂度,面对各种复杂查询场景性能很高,而且相比自建数据库速度提升7倍以上。)"
Transparency Evidence & methodology 100/100 reference value · 81% confidence · 1 source

Review how this reference was scored and trace each normalized claim back to its documented source.

01

Scoring methodology

How scores are calculated

Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.

Customer reference value100/100
  • Problem clarity20/20
    Measures whether the customer's operational problem is specific enough to compare.
  • Implementation depth25/25
    Rewards named components, data flow, integrations, and architecture detail.
  • Outcome specificity20/20
    Measures whether the reported benefit is concrete rather than a generic claim.
  • Quantified impact25/25
    2 metrics describe measurable impact.
  • Deployment scale10/10
    Rewards evidence of a live or scaled deployment and meaningful integration scope.
Evidence confidence81%
  • Source authority20/25
    The main evidence is an official vendor-published customer story.
  • Claim traceability21/25
    10 of 12 key fields link to a named source.
  • Direct evidence20/20
    5 direct source quotes retained.
  • Date provenance0/10
    The publication date has no documented provenance.
  • Source transparency10/10
    1 source reference document publisher, URL, authority, retrieval, and supported fields.
  • Implementation specificity10/10
    Rewards source-backed components, data flow, quantified impact, and a specific outcome.
02

Traceability

Sources & provenance

Field coverage shows which facts each source supports. Dates are shown only when documented by the source.

  1. Huawei Cloud

    素问科技_客户案例_初创生态-华为云

    First-party
    https://www.huaweicloud.com/cases/suwenkeji.html
    Source type
    Vendor-published customer story
    Supports
    • Customer
    • Business problem
    • Solution
    • Products
    • Solution components
    • Data flow
    • Integrations
    • technical area
    • use case category
    • Business outcome
    • outcome category
    • Metrics
    • Architecture
    • evidence quotes

Date evidence

  • First archived Jul 22, 2024 · Internet Archive · Independent First Wayback Machine capture; confirms the page existed by this date, not when it was published.
    Open date source
Not separately sourced Industry, Region, Company size. Treat these normalized attributes as lower-confidence until a dedicated source is added.