Remove 2015 Remove Latency Remove Speed
article thumbnail

The Speed of Time

Brendan Gregg

A Cassandra database cluster had switched to Ubuntu and noticed write latency increased by over 30%. Measuring the speed of time Is there already a microbenchmark for os::javaTimeMillis()? I also shared setting the clocksource in my talks and in my 2015 [Linux tunables] post. Microbenchmark os::javaTimeMillis() on both systems.

Speed 103
article thumbnail

USENIX LISA2021 Computing Performance: On the Horizon

Brendan Gregg

## References I've reproduced the talk references below, so you can click on links: - [Gregg 08] Brendan Gregg, “ZFS L2ARC,” [link] Jul 2008 - [Gregg 10] Brendan Gregg, “Visualizations for Performance Analysis (and More),” [link] 2010 - [Greenberg 11] Marc Greenberg, “DDR4: Double the speed, double the latency? Ford, et al., “TCP

Insiders

Sign Up for our Newsletter

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

article thumbnail

USENIX SREcon APAC 2022: Computing Performance: What's on the Horizon

Brendan Gregg

My personal opinion is that I don't see a widespread need for more capacity given horizontal scaling and servers that can already exceed 1 Tbyte of DRAM; bandwidth is also helpful, but I'd be concerned about the increased latency for adding a hop to more memory. Ford, et al., “TCP

article thumbnail

The Speed of Time

Brendan Gregg

A Cassandra database cluster had switched to Ubuntu and noticed write latency increased by over 30%. Measuring the speed of time Is there already a microbenchmark for os::javaTimeMillis()? I also shared setting the clocksource in my talks and in my 2015 Linux tunables post. Microbenchmark os::javaTimeMillis() on both systems.

Speed 52
article thumbnail

The Speed of Time

Brendan Gregg

A Cassandra database cluster had switched to Ubuntu and noticed write latency increased by over 30%. Measuring the speed of time Is there already a microbenchmark for os::javaTimeMillis()? I also shared setting the clocksource in my talks and in my 2015 [Linux tunables] post. Microbenchmark os::javaTimeMillis() on both systems.

Speed 40
article thumbnail

Linux Load Averages: Solving the Mystery

Brendan Gregg

Anyway, the following patch seems to make the load average much more consistent WRT the subjective speed of the system. Some people have found values that seem to work for their systems and workloads: they know that when load goes over X, application latency is high and customers start complaining. One system with a ratio of 1.5

Latency 111
article thumbnail

SpaceX Spending $10 Billion to Make the Internet 20ms Faster

MachMetrics

Elon Musk’s need for speed. However, there is excitement around Starlink for other reasons – namely, the implications it might have for internet speed and latency – even by just a small amount (20 milliseconds on average). This ultimately guides us to one conclusion: Speed is incredibly important.