A data-driven business relied on daily CSV uploads containing nearly 150,000 pricing records from multiple suppliers. Each file included supplier names, pricing information, and other critical business data that needed to be processed accurately while maintaining data integrity and preventing duplicate records.
The existing processing workflow was slow, resource-intensive, and constrained by limited server capacity. Daily uploads took up to 30 minutes to complete, impacting operational efficiency and making it difficult to keep pricing information up to date without affecting platform performance.
To solve these challenges, SynergyTop redesigned the entire data processing workflow using intelligent batching, AWS Lambda, automated SQL processing, and a dedicated processing database, dramatically improving performance while ensuring uninterrupted platform availability.
Headquarters: Undisclosed
Industry: Data Processing
Service Type: Performance Optimisation & Utility Development
Geographies Impacted: Global
Project Duration: Ongoing
Key Technology: AWS Lambda, SQL, CSV Processing
The client wanted a solution that could:
The platform was designed for:
The client faced significant performance and infrastructure limitations that affected daily data processing.
SynergyTop redesigned the complete data processing workflow to maximise performance while maintaining uninterrupted access to the platform.
The solution included:
The optimised architecture enables high-speed processing while ensuring users continue accessing live pricing data without disruption.
| Steps | Process Details | Deliverable |
| Infrastructure Optimisation | Upgraded the server infrastructure from 1 GB to 4 GB RAM to provide sufficient resources for high-volume data processing. | Result: Improved processing capacity and enhanced system stability. |
| Intelligent Batch Processing | Divided uploaded CSV files into batches of 1,000 records and grouped them into sets of ten for efficient parallel execution while preventing duplicate entries through unique mapping. | Result: Faster and more reliable processing of large datasets. |
| AWS Lambda Processing | Leveraged AWS Lambda functions to process multiple record batches simultaneously, accelerating data conversion, validation, and updates. | Result: Significant reduction in overall processing time. |
| Automated Data Validation | Implemented SQL-based bulk insertion, deletion, supplier ID conversion, and validation processes while maintaining data integrity throughout the workflow. | Result: Accurate, automated data management with minimal manual intervention. |
| Background Processing Architecture | Performed all processing within a dedicated pricing_upload database while users continued accessing existing pricing information. Once processing completed, updated records replaced the previous dataset seamlessly. | Result: Continuous platform availability with zero disruption to users. |
Large CSV files are divided into manageable batches for efficient parallel execution.
Serverless processing accelerates data imports while reducing system workload.
Automated SQL workflows prevent duplicate records and preserve existing supplier information.
Background processing allows users to continue working while data updates are completed.
The optimised framework supports high-volume daily imports while maintaining system performance.
Daily CSV upload times were reduced from 20–30 minutes to just 30–40 seconds.
Users experienced uninterrupted access to the platform while data processing occurred in the background.
Automated validation and duplicate prevention maintained data integrity throughout every upload.
Batch processing and AWS Lambda significantly lowered resource consumption while improving overall efficiency.
The new architecture enables reliable processing of large daily datasets while supporting future growth.
| Aspect | Before Solution | After Solution |
| CSV Processing Time | 20–30 minutes | 30–40 seconds |
| Infrastructure | 1 GB RAM | 4 GB RAM with optimised processing |
| Data Processing | Sequential workflow | Parallel batch processing with AWS Lambda |
| Duplicate Prevention | Complex manual validation | Automated unique mapping and validation |
| Platform Availability | Processing impacted users | Zero-downtime background processing |
| System Efficiency | High server load | Optimised, high-performance workflow |
AWS Lambda
SQL
CSV Batch Processing, Automated Validation