02 Sep 2021
Tags:
Distributed systems
Experienced performance analysts caution against taking averages of
percentiles. Gil Tene’s famous 2015 Strange Loop talk,
“How NOT to Measure Latency”,
features the tip, “You can’t average percentiles. Period.” (at
9:15). Tene offers a brief example using the 100th percentile, the
maximum, but does not provide many details. You might think this is
enough—simply never average percentiles and you’re safe, right? But
as Baron Schwartz points out,
the metric pipeline probably averages metrics before you see them,
and even displaying metrics as pixels on the screen is a form of
averaging. Why is it so bad to average percentiles and how
much information does the result retain, if any at all?
more ...
10 Jul 2021
Tags:
Distributed systems
Cloud vendors are rolling out products that place proprietary
hardware in customer’s datacentres, delivering a selection of the
vendor’s cloud services from the customer’s datacentre. These new
products require a rethinking of current guidelines regarding
tradeoffs between data residency, latency, and failure tolerance. As a
start towards this, I compare two specific products,
Amazon Outpost in 42U rack form factor
and
Oracle Dedicated Region Cloud @ Customer,
according to these criteria. What do these comparisons tell us about
the new possibilities such products create?
more ...
13 Jun 2021
Tags:
Rhetoric
I sometimes get asked whether I’m a “prescriptivist” or
“descriptivist”, or—worse still—labelled as one. This has always
struck me as beside the point. In none of my roles, not as a writer,
teacher of writing, or editor, has this distinction ever provided useful
guidance. Instead, I base my usage choices upon some combination of
assessed ambiguity to the reader, existing convention, degree of
formality, emotional impact, and other considerations. The resulting
choices might sometimes conform to prescriptive rules or existing
usage or neither. Effective writing is a balance of competing
requirements, not rote adherence to an overarching “philosophy”.
more ...
03 Jun 2021
Tags:
Rhetoric
Time for another rant about a word seeing increased use in technical
discussions (and elsewhere): superpower. It just doesn’t mean what
people imply it means.
more ...
31 May 2021
Tags:
Rhetoric
I see increasing use of the phrase “opinionated design”, frequent
enough to risk it becoming a
vogue word in our
field. What makes a design “opinionated”? If someone declares a
design to be “opinionated”, what might that tell you about it? How
would a designer justify that claim?
more ...