FPGA Compression Service for In-Memory Database Query Performance

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Database management systems face challenges in optimizing performance due to complex queries and large data volumes, particularly in in-memory databases where compression is needed to save memory resources but also consumes processing time, affecting query response times.

Innovation Solution

Implementing a hardware-based compression service (CaaS) using FPGA, GPU, or TPU accelerators to offload compression tasks from CPU, applying adaptive compression techniques like Re-Pair to optimize memory usage while minimizing processing overhead.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If software-based compression is used in database management systems, then memory footprint is reduced, but processing time increases

Engineering Contradiction:
Improvememory footprintVSAvoidprocessing time
Core Design Contradiction:
Quantity of substanceVSLoss of time

Solution Approach 1:

The patent replaces software-based compression mechanisms with hardware-based compression using FPGAs, GPUs, or dedicated compression appliances. This substitution moves the compression function from the software/CPU domain to specialized hardware, enabling parallel processing and significantly reducing the processing time penalty while maintaining compression effectiveness for reducing memory footprint.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces a hardware compression service as an intermediary component between the database management system and storage/memory resources. This intermediary handles compression and decompression operations independently, allowing the database to maintain smaller memory footprints without directly incurring the processing time costs of software-based compression.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Quantity of substance

If compression is applied to reduce memory usage, then memory resources are optimized, but query performance deteriorates

Engineering Contradiction:
Improvememory resourcesVSAvoidquery performance
Core Design Contradiction:
Quantity of substanceVSProductivity

Solution Approach 1:

By replacing software compression with hardware-based compression, the patent eliminates the performance penalty associated with CPU-bound compression operations. The hardware accelerators perform compression and decompression in parallel, maintaining high query performance while achieving memory resource optimization through compression.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent changes the operational parameters of compression by using hardware-based implementations with different compression algorithms and settings optimized for specific data types and access patterns. This allows tuning of compression ratios and speeds to balance memory optimization with query performance requirements.

Inventive Principle:
Principle #35Parameter changes

3Extent of automation

If CPU resources are used for compression tasks, then compression is performed, but transaction processing speed decreases

Engineering Contradiction:
Improvecompression capabilityVSAvoidtransaction processing speed
Core Design Contradiction:
Extent of automationVSSpeed

Solution Approach 1:

The patent extracts the compression function from the CPU's responsibility and relocates it to dedicated hardware compression services. This separation allows the CPU to focus entirely on transaction processing while hardware handles compression independently, eliminating the trade-off between having compression capability and maintaining transaction processing speed.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

A hardware compression service acts as an intermediary that handles all compression operations independently of the CPU's transaction processing pipeline. This intermediary processes compression requests in parallel, ensuring that transaction processing speed is not degraded while full compression capability is maintained.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP4304094B1Compression service using FPGA compression
Publication Date: 2026.04.15 SAP SE
  • EP4304094B1 patent drawingFigure 1A
  • EP4304094B1 patent drawingFigure 1B
  • EP4304094B1 patent drawingFigure 2

AI summary

A method, a system, and a computer program product for providing a compression service, such as a Compression as a Service (Caas). In some implementations, there is provided a method including receiving, at a compression service, metric information regarding one or more columns of a database server; in response to receiving the metric information, processing, by the compression service, the metric information to identify at least one column as a candidate for compression by the compression service; in response to identifying the at least one column, compressing by the compression service, the at least one column, wherein the compression service uses a hardware accelerator to execution a compression algorithm that compresses the at least one column; and notifying, by the compression service, the database server of the at least one column compressed by the compression service. Related systems, methods, and articles of manufacture are also disclosed.