What's the Cost of a Millisecond?

Note: Presentation times are in Coordinated Universal Time (UTC).

Tuesday, 2021, October 12 - 15:0015:30

Avishai Ish-Shalom, ScyllaDB


We all want fast services, but how fast is fast? Would you work hard to shave off a millisecond off the mean latency? The 99th percentile? If aiming for 300ms latency, you might answer "probably not." However, due to various phenomena collectively known as "latency amplification," a single millisecond deep in your stack can turn into a large increase in user-visible latency—and this is very common in microservices-based systems. What is the true cost of a millisecond?

Avishai Ish-Shalom, ScyllaDB

"In a world where anything has an API, everything is a software problem," this insight has guided Avishai Ish-Shalom throughout his diverse career working on improving the complex socio-technical systems that create and operate modern software and promoting the use of Mathematics in system design and operations. Spending 15 years in various software fields and capacities, Avishai has served as Engineer in Residence in Aleph VC, engineering manager at Wix.com, co-founded Fewbytes, and consulted many other companies on software operations, reliability, design, and culture. Currently, Avishai is a Developer Advocate for ScyllaDB (the boring database ;-)

SREcon21 Open Access Sponsored by Indeed

@conference {276645,
author = {Avishai Ish-Shalom},
title = {What{\textquoteright}s the Cost of a Millisecond?},
year = {2021},
publisher = {USENIX Association},
month = oct,

Presentation Video