Serverless Function Latency Reduction via Direct Container Addressing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current serverless computing environments face inefficiencies such as low resource utilization, long function response latencies, and inefficient inter-function communication, leading to reduced overall efficiency.

Innovation Solution

The system directly identifies and communicates with called functions within containers using their addresses, bypassing platform services, and optimizes container placement and scheduling based on communication patterns and slack times to reduce latency and resource usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If function calls are routed through platform services in serverless computing environments, then service management and orchestration are simplified, but communication latency increases and resource utilization decreases

Engineering Contradiction:
Improvefunction response latencyVSAvoidcommunication infrastructure complexity
Core Design Contradiction:
Loss of timeVSDevice complexity

Solution Approach 1:

The patent extracts the direct communication capability from the platform service abstraction layer, allowing functions to communicate directly with each other through identified addresses rather than routing all communication through platform services. This reduces latency while maintaining simplified management for non-critical operations.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent introduces an address identification mechanism that acts as a lightweight intermediary, providing direct function access without full platform service mediation. This intermediary layer reduces communication overhead while maintaining service orchestration benefits for management tasks.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Productivity

If traditional serverless computing routing is used, then service orchestration is maintained, but resource utilization efficiency decreases due to indirect communication paths

Engineering Contradiction:
Improveresource utilization efficiencyVSAvoidcommunication overhead time
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent segments communication paths into two types: direct function-to-function communication for performance-critical operations, and platform service routing for management tasks. This segmentation allows optimal resource utilization by using the most efficient path for each communication type.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent changes the communication parameter from indirect platform-service-mediated routing to direct address-based routing. This parameter change reduces communication overhead time and improves resource utilization efficiency by eliminating unnecessary intermediate processing steps.

Inventive Principle:
Principle #35Parameter changes

3Speed

If direct function addressing is implemented, then communication latency is reduced, but system complexity for address management increases

Engineering Contradiction:
Improvefunction call speedVSAvoidaddress management complexity
Core Design Contradiction:
SpeedVSDevice complexity

Solution Approach 1:

The patent implements self-service address identification where functions automatically obtain and use addresses of called functions without manual configuration. This reduces address management complexity while enabling fast direct communication, as the system manages addresses automatically through the execution environment.

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS20240231898A1Serverless Computing with Latency Reduction
Publication Date: 2024.07.11 INTERNATIONAL BUSINESS MACHINE CORPORATION
  • US20240231898A1 patent drawing
  • US20240231898A1 patent drawing
  • US20240231898A1 patent drawing

AI summary

A computer implemented method manages function execution. A number of processor units identify an address of a called function in a second container called by a calling function in a first container. The number of processor units sends a request directly from the calling function in the first container to the called function in the second container using the address.