SHENZHEN, China, April 25, 2025 /PRNewswire/ -- Under the Data+AI convergence trend, Lakehouse represents the core evolution direction for big data platforms. Tencent Cloud Big Data-Data Lake Compute (DLC) is an open, unified cloud-native Serverless Lakehouse data warehousing and analytics service. It features cloud-native storage-compute separation, lake format governance, deep optimizations for Iceberg, unified metadata management, and machine learning training support, enabling enterprises to build agile Data+AI Lakehouse platforms. DLC has helped a leading Southeast Asian retail conglomerate reduce data platform and machine learning costs by 70%.
Core Challenges
- Separated Resources: Big data and AI required isolated infrastructure investments with no resource sharing.
- Data Movement Overheads: Data silos between Data and AI systems caused latency and additional costs.
DLC Solution
Tencent Cloud's one-stop intelligent platform WeData, powered by Data Lake Compute (DLC), delivers an integrated DataOps+MLOps platform solution enabling:
- Seamless Experience: Unified development for both data analytics and ML workflows
- Cost Efficiency: 70% cost reduction through shared compute resources for Data+AI workloads
- AI Innovation Acceleration: Rapid deployment of GenAI applications via unified data governance
Key Features
- Unified Engine: Supports data engineering and ML training (Scikit-Learn, PyTorch, TensorFlow)
- Meson Engine: Proprietary high-performance engine delivering 2x faster processing than open-source Spark
- Dynamic Resource Allocation: Flexible resource groups for data engineering, interactive analytics, and ML tasks
Business Value
DLC integrates data engineering, SQL analytics, and machine learning into a unified platform with industry-leading cost-effectiveness. The solution has already enabled a Southeast Asian retail leader to achieve 70% TCO reduction across data and ML infrastructure.
#Tencent Cloud Big Data #Data Lake Compute
source: Tencent Cloud
【etnet獨家優惠】親手炮製母親節＆端午節海鮮盛宴！使用優惠碼享95折優惠！► 立即了解
