Errata

Implementing Service Level Objectives

Errata for Implementing Service Level Objectives

Submit your own errata for this product.

The errata list is a list of errors and their corrections that were found after the product was released.

The following errata were submitted by our customers and have not yet been approved or disproved by the author or editor. They solely represent the opinion of the customer.

Color Key: Serious technical mistake Minor technical mistake Language or formatting error Typo Question Note Update

Version Location Description Submitted by Date submitted
50
4th paragraph

Not sure of the exact page, but the calculation under Dependency Math, the 4th paragraph:
0.9999^40 = 0.9960077
vs
0.999^40 = 0.96077

 Anonymous  Jul 19, 2021 
Printed Page 117
last line

I can't figure out how to go from the values in table 7-3 to the calculation on bottom of page 117, and I believe that by mistake, the calculation from page 83 has been repeated here, and instead it should be 2250/2150. But I still don't understand how the value 95,56% on the next page (118)

 Anders Bruvik  Dec 08, 2020 
117
last line

Page 117
last line
referencing Anders Bruvik's post - I think the text should have something to the effect of
300 + 400 + 800 + 500 + 100 + (100/2) = 2150. So there are roughly 2150 observations at or below 75ms. Then this value, 2150 is 95.56% of the overall count (2150/2250 = .9556)

 Anonymous  Jul 20, 2021 
Printed Page 188
figure 9-21

the printed figure 9-21 is a copy of figure 9-20, and does not show the distribution with 10 data point as the description indicates

 Anders Bruvik  Dec 13, 2020 
Printed, Page 231
middle of the page

In the below paragraph, right under the heading "Data and Data Application Reliability", the first past of the sentence has been erroneously cut.

------

Data and Data Application Reliability
live the heat death of the universe, despite all of the nines in the vast armada of servers its engineers construct for it, out in the black vastnesses between the stars.

Polina Giralt  Dec 08, 2020 
Printed, Page 231
Data and Data Application Reliability

As Polina mentioned, the first sentence is missing and the secind sentence is mangled.

should read:

"We cannot stop every type of fault, no matter what we do. Google will not outlive the heat death of the universe, no matter how numberless are the nines in the astral armada of servers its engineers construct for it, out in the black vastnesses between the stars."

 Anonymous  Jan 12, 2021 
240
4th paragraph

I had to guess on the page number. It's the 4th paragraph in the section:
OPERATIONAL INSTRUMENTATION FOR NONRECOVERABLE SLOS
Very resilience! Much reliable! Wow! Now we have a better idea of the risk we’re carrying, and an idea of where our failures outstrip our ability to recover from them. Pretty great, eh?
Should read something like:
Very high resilience! Much more reliable! Wow! Now we have a better idea of the risk we’re carrying, and an idea of where our failures outstrip our ability to recover from them. Pretty great, eh?

 Anonymous  Jul 29, 2021 
PDF Page 275
3rd paragraph, 2nd item in the indented area

The SLOs for smaller services are described in opposite way. I assume the SLOs should be "no more than 2" for 10-19 pods and "no more than 1" for -10 pods services.

> For services with 10–19 pods, no fewer than 2 of their configured number will be unavailable 99.9% of the time.
> For services with fewer than 10 pods, no fewer than 1 of their configured number will be unavailable 99.9% of the time.

 Yoshi Yamaguchi  Apr 24, 2022