Test Findings

Overview

This test plan outlines the use cases to benchmark the query response time of the nwaku storev3 protocol (WAKU2-STORE), as well as the testing environment. The store protocol enables querying of messages received through the relay protocol and stored by other nodes. Testing will mainly focus on simulating the configuration and use cases of the Status network to ensure reliability.

Use cases

• UC1: Request messages (with content) by content-topics and time range

  The client requests the retrieval of messages with their full content, using content-topics and a specified time range as the primary query criteria.

• UC2: Request messages hashes by content-topics and time range

  The client requests the retrieval of message hashes, using content-topics and a specified time range as the primary query criteria.

• UC3: Request messages (with content) by specific message hashes

  The client requests the retrieval of messages with their full content, using specific message hashes as the primary query criteria.

• UC4: Request message hashes by specific message hashes

  The client requests the retrieval of message hashes, using specific message hashes as the primary query criteria.

TL;DR

There are many variables, conditions and configurations that have varying influence on the query response time. The testing objective is to determine to what extent each one influences the query response time, hopefully identifying bottlenecks and optimal system configuration.

The store query response time will be benchmarked against:

• query type (message hash or time range and content-topics)
• query criteria (different time ranges and content-topics)
• query rate
• total store nodes connected to a single db
• db size (number of records)
• db content (message size and different content-topics)
• db config (number and size of partitions)
• db write rate (related to the rate at which the store node receives relay messages)