article thumbnail

LISA2019 Linux Systems Performance

Brendan Gregg

Systems performance is an effective discipline for performance analysis and tuning, and can help you find performance wins for your applications and the kernel. So far my workshops have not been video recorded, but in the interests of supporting WFH I hope to get a workshop recorded sometime and put online.

Systems 144
article thumbnail

Impact of Data locality on DB workloads.

n0derunner

In this video I migrate a Postgres DB running PGbench benchmark. The variation in the transaction rate is due to the benchmark itself, the transaction rate is not expected to be uniform. The Postgres DB is totally un-tuned and contains purely default settings. The DB is running on a Host which is CPU constrained.

Insiders

Sign Up for our Newsletter

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

Trending Sources

article thumbnail

New web performance insights with additional metrics and enhanced Visually complete for synthetic monitors

Dynatrace

This is definitely a great starting benchmark against which to optimize your application. Dynatrace made the decision when we first introduced Visually complete three years ago to base our calculations on changes to the DOM rather than measure Visually complete based on a video capture of the page load.

Metrics 215
article thumbnail

LISA2019 Linux Systems Performance

Brendan Gregg

Systems performance is an effective discipline for performance analysis and tuning, and can help you find performance wins for your applications and the kernel. So far my workshops have not been video recorded, but in the interests of supporting WFH I hope to get a workshop recorded sometime and put online.

Systems 52
article thumbnail

LISA2019 Linux Systems Performance

Brendan Gregg

Systems performance is an effective discipline for performance analysis and tuning, and can help you find performance wins for your applications and the kernel. The video is on youtube : The slides are on slideshare (or PDF ): The different background colors I used for the screenshots have no meaning: I just got bored of using gray.

Systems 52
article thumbnail

Building Netflix’s Distributed Tracing Infrastructure

The Netflix TechBlog

Investigating a video streaming failure consists of inspecting all aspects of a member account. Our engineering teams tuned their services for performance after factoring in increased resource utilization due to tracing. which is difficult when troubleshooting distributed systems.

article thumbnail

Performance gains for postgres on Linux with hugepages

n0derunner

I use the “built in” benchmark pgbench to run a simple set of queries. Then reboot the kernel I used these excellent resources How to modify the kernel command line How to enable hugepages and this great video on Linux virtual memory. For this experiment I am using Postgres v11 on Linux 3.10