STO Express replaced Oracle databases with Lindorm for its core logistics workloads — eliminating Oracle licensing costs, handling hundreds of millions of scan events per day, and delivering query performance five times higher than Oracle. The migration required no application code changes.
| Metric | Value |
|---|---|
| Query performance vs. Oracle | 5x higher |
| LindormTable throughput | 100,000 requests/second |
| Daily data volume | 100+ TB |
| Barcode scan events per day | Hundreds of millions |
| Code changes required | None |
Customer profile
STO Express is a courier company founded in 1993. As of January 2021, the company operates more than 4,500 branches and independent outlets, more than 25,000 logistics centers, and employs more than 300,000 people. STO Express handles both traditional courier services and e-commerce fulfillment.
Challenges
STO Express ran into three compounding problems as order volumes grew:
Data volume: Barcode scanners generate hundreds of millions of records per day. Order placement and distribution operations add tens of millions more. Total daily data exceeds 100 TB.
Query diversity: Operations require both exact lookups by order number and other dimensions, and fuzzy searches across large datasets — workloads that pull in opposite directions for a single database engine.
Peak traffic: Order volumes spike sharply during events like Double 11 Shopping Festival.
The company had originally used Oracle for its reputation for performance and stability. After learning that Cainiao Logistics used Lindorm for similar logistics workloads, STO Express evaluated the LindormTable and LindormSearch architecture.
Solution
STO Express replaced Oracle with an architecture built on three Lindorm components:
LindormTable: wide-table engine for high-throughput reads and writes on order, waybill, and sub-order data
LindormSearch: search engine for multi-dimensional queries and fuzzy searches across the same dataset
Lindorm Tunnel Service (LTS): real-time data synchronization from LindormTable to LindormSearch, so the business system writes to one engine and both stay consistent
The architecture is shown below.
Benefits
Performance: LindormTable sustains 100,000 requests per second. Query performance is five times higher than Oracle. LindormSearch handles random multi-dimensional queries from tens of thousands of service outlets and returns result sets for export and download in real time.
Scalability: Lindorm scales vertically and horizontally to absorb peak traffic during Double 11 Shopping Festival. Because computing is decoupled from storage, STO Express can expand either layer independently, without application downtime.
Operational simplicity: LTS synchronizes data from LindormTable to LindormSearch in real time, eliminating the need for the business system to write to both engines separately. STO Express did not modify any application code to complete the migration.
Cost: Migrating all data from Oracle to Alibaba Cloud reduced licensing and infrastructure costs. Lindorm's multiple storage media types and high compression ratio lower the total cost of storing 100+ TB per day.
Stability: STO Express has run Lindorm in production for nearly two years. During that time, the system has maintained the high availability and stability required for logistics operations.
Customer quote: "We found that Lindorm provides linear scalability. This makes Lindorm more suitable for logistics business. We do not need to modify the code or migrate data to new instances to handle data during rapid business growth and promotions such as Double 11 Shopping Festival. Lindorm databases are also more cost-effective than Oracle databases."