Databases 10 min read

Why Teradata Is Leaving China and Which Domestic Data Warehouses Can Fill the Gap

Teradata announced its withdrawal from China due to geopolitical uncertainty and rising competition from mature domestic data‑warehouse solutions, prompting a detailed analysis of its architecture, the main Chinese warehouse designs, global market positioning, and migration tools for replacing Teradata.

dbaplus Community
dbaplus Community
dbaplus Community
Why Teradata Is Leaving China and Which Domestic Data Warehouses Can Fill the Gap

Reasons for Teradata's Exit

Teradata announced withdrawal from China due to uncertainty in the current and future business environment, international relations, and incidents such as alleged surveillance. A major factor is the rapid maturation of domestic data‑warehouse products that are eroding its market share.

Teradata Technical Architecture

Teradata uses a shared‑nothing massively parallel processing (MPP) architecture. The main components are:

Parser Engine – parses SQL statements and generates execution plans.

BYNET – high‑speed interconnect that routes messages between nodes.

Access Module Processor (AMP) – each AMP stores a slice of data and executes the plan locally; a node may contain multiple AMPs, making the node the basic system unit.

Teradata architecture
Teradata architecture

Domestic Data‑Warehouse Architectures

Chinese vendors mainly adopt three architectural patterns:

Master‑Based – a dedicated master node coordinates metadata and query planning. Example: PostgreSQL‑derived products such as Greenplum.

Master‑less Integrated Compute‑Storage‑Management – compute, storage, and metadata are tightly coupled in each node, eliminating a single master.

Multi‑Master (Federated) Architecture – multiple master nodes operate in a cluster for high availability. GBase 8a MPP V9 is the only domestic product supporting this model; HDP 2.0 is a comparable foreign solution.

Master‑Based architecture
Master‑Based architecture
Master‑less architecture
Master‑less architecture
Multi‑Master architecture
Multi‑Master architecture

Global Data‑Warehouse Landscape

Gartner’s Magic Quadrant lists Huawei GaussDB and Nanda‑General GBase alongside Oracle and Teradata. In the 2017 Asia‑Pacific quadrant, three Chinese companies (GBase, Alibaba Cloud, Huawei) were the only entrants, indicating strong international recognition.

Gartner quadrant
Gartner quadrant

Representative Domestic Products

GBase 8a – analytical MPP database from Nanda‑General, version V953, in production since 2010, optimized for OLAP workloads.

GaussDB DWS – Huawei’s analytical data‑warehouse (GaussDB 200). GaussDB 100 (openGauss) targets OLTP workloads.

Migration Capabilities

Both products provide tools to migrate schemas, views, SQL, and data from Teradata:

GaussDB – shell‑based migration scripts ending with .sh that convert DDL and SQL; data migration is not supported.

GBase 8a – GBase MTK utility can convert DDL from Oracle, SQL Server, PostgreSQL, and Teradata and perform limited data movement. The tool is implemented in Python, allowing custom conversion rules.

GaussDB migration tool
GaussDB migration tool
GBase MTK tool
GBase MTK tool

Implications of Teradata’s Withdrawal

The exit highlights the risk of over‑reliance on foreign database vendors. Mature domestic solutions such as GBase 8a and GaussDB DWS are positioned to capture the displaced market and to further strengthen China’s data‑management ecosystem.

Original Source

Signed-in readers can open the original source through BestHub's protected redirect.

Sign in to view source
Republication Notice

This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactadmin@besthub.devand we will review it promptly.

Big DataData Warehousedatabase migrationGaussDBGBaseTeradata
dbaplus Community
Written by

dbaplus Community

Enterprise-level professional community for Database, BigData, and AIOps. Daily original articles, weekly online tech talks, monthly offline salons, and quarterly XCOPS&DAMS conferences—delivered by industry experts.

0 followers
Reader feedback

How this landed with the community

Sign in to like

Rate this article

Was this worth your time?

Sign in to rate
Discussion

0 Comments

Thoughtful readers leave field notes, pushback, and hard-won operational detail here.