Remove 2017 Remove Benchmarking Remove Latency Remove Tuning
article thumbnail

Building Netflix’s Distributed Tracing Infrastructure

The Netflix TechBlog

If we had an ID for each streaming session then distributed tracing could easily reconstruct session failure by providing service topology, retry and error tags, and latency measurements for all service calls. Our engineering teams tuned their services for performance after factoring in increased resource utilization due to tracing.

article thumbnail

The Speed of Time

Brendan Gregg

A Cassandra database cluster had switched to Ubuntu and noticed write latency increased by over 30%. I love short benchmarks like this as I can disassemble the resulting binary and ensure that the compiled instructions match my expectations, and the compiler hasen't messed with it. ## 6. include <sys/time.h>

Speed 126
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

The Speed of Time

Brendan Gregg

A Cassandra database cluster had switched to Ubuntu and noticed write latency increased by over 30%. I love short benchmarks like this as I can disassemble the resulting binary and ensure that the compiled instructions match my expectations, and the compiler hasen't messed with it. ## 7. include <sys/time.h>

Speed 40
article thumbnail

A peculiar throughput limitation on Intel’s Xeon Phi x200 (Knights Landing)

John McCalpin

Introduction: In December 2017, my colleague Damon McDougall (now at AMD) asked for help in porting the fused multiply-add example code from a Colfax report ( [link] ) to the Xeon Phi x200 (Knights Landing) processors here at TACC. of the “adjusted peak performance”, there is no longer a significant upside to performance tuning.

Latency 40