Gadgets

A server with 1.5TB of RAM can open 4,500 Google Chrome tabs before crashing

A server equipped with 1.5TB of DDR5 RAM and dual AMD EPYC 9684X processors can support 4,500 open Google Chrome tabs before the browser crashes. The experiment, conducted on both Windows 11 and Ubuntu, utilized a custom Python script to automate tab creation and monitor system resources. While the system remained responsive with 500 tabs, performance degraded as the tab count increased. On Windows 11, the browser crashed after reaching approximately 4,500 tabs. The test revealed that modern websites are increasingly resource-intensive, with the GPU's VRAM often becoming a significant bottleneck before system RAM capacity is exhausted. The study also highlighted differences in how operating systems manage hardware information, with Linux treating metrics as accessible files, whereas Windows relies on formal system API queries. Additionally, the experiment demonstrated the utility of zRAM in Linux for compressing data in RAM, which can help optimize memory usage on systems with limited capacity, although it was not the primary limiting factor in this high-memory configuration.

Tech4K StaffVideo report

A server equipped with 1.5TB of DDR5 RAM and dual AMD EPYC 9684X processors can support 4,500 open Google Chrome tabs before the browser crashes. The experiment, conducted on both Windows 11 and Ubuntu, utilized a custom Python script to automate tab creation and monitor system resources. While the system remained responsive with 500 tabs, performance degraded as the tab count increased. On Windows 11, the browser crashed after reaching approximately 4,500 tabs. The test revealed that modern websites are increasingly resource-intensive, with the GPU's VRAM often becoming a significant bottleneck before system RAM capacity is exhausted. The study also highlighted differences in how operating systems manage hardware information, with Linux treating metrics as accessible files, whereas Windows relies on formal system API queries. Additionally, the experiment demonstrated the utility of zRAM in Linux for compressing data in RAM, which can help optimize memory usage on systems with limited capacity, although it was not the primary limiting factor in this high-memory configuration.

A server with 1.5TB of DDR5 RAM can maintain 4,500 open Google Chrome tabs before the browser crashes. The system's GPU VRAM often becomes a limiting factor for browser performance before the total system RAM is exhausted.

Windows 11 and Ubuntu handle hardware information differently, with Linux treating metrics as files and Windows using system API queries. The experiment demonstrated that modern websites are significantly more resource-intensive than in previous years.

zRAM in Linux can compress data in RAM to optimize memory usage, though it was not the primary bottleneck in this test.

Chapter guide

IntroductionThe host introduces the experiment to determine how many Chrome tabs can be opened before a system crashes.Hardware SetupThe server configuration is detailed, featuring dual AMD EPYC 9684X processors and 1.5TB of DDR5 RAM.Testing MethodologyThe host explains the use of a custom Python script to automate the opening of Chrome tabs and monitor system metrics.Initial TestingThe team begins opening tabs and monitoring the system's response on Windows 11.Linux TestingThe experiment is repeated on an Ubuntu system to compare performance and resource management.Apple's Resource ManagementThe host discusses observations regarding how macOS manages memory and pre-caches data.Final ResultsThe team reaches the limit of 4,500 tabs and analyzes the final performance data.

Worth noting

  • The video includes a sponsored segment for the game War Thunder.
  • The test methodology is limited to a single server configuration and may not reflect performance on consumer-grade hardware.
  • The results are specific to the version of Chrome and the operating systems used at the time of testing.

Watch the original video ↗