Fast Frozen Fun

As some of you may know I’m retired from performance work and my only “job” is working one day a week as a tour guide at Ben & Jerry’s.  I love ice cream and I love hands-on experimenting and lately I’ve been experimenting with instant fruit ice cream and it’s great. Here is the recipe.

Put 10 -12 ounces (280 – 340grams) of hard frozen fruit in a food processor. I’ve tried strawberries, blueberries, mangos, and peaches and they all taste great. Add to that 1/2 cup (100grams) of sugar. Now run it until the fruit begins to break down into small bits and then add about 1 cup (250ml) of half-n-half. What you end up with is soft-serve ice cream that you should plan to eat immediately.  Total time: less than 2 minutes.yum

Some notes from my experiments:

  • If you want a stiffer product, pre-freeze the work bowl and blade of the food processor as well as the sugar.
  • Try dropping a chocolate bar into the food processor as it is spinning. Strawberry and chocolate go so well together.
  • Try adding a few drops of vanilla, almond, or cherry extract. Yum!
  • Heavy cream did not work for me, as the food processor acts like a churn and the ice cream had a distinctive greasy mouth feel that food scientists refer to as “buttering”.

 

Interactive Computer Latency Numbers Through Time

Go here: http://www.eecs.berkeley.edu/~rcs/research/interactive_latency.html

Grab the slider at the top of the screen and see how latency values for common computer tasks have changed starting in 1991 and projected out to 2020. To me, the precise values aren’t as interesting as seeing how the performance battles programmers fight change over time.

Sherman, set the wayback machine to…wayback

Tuning Chocolate Chip Cookies

In most computer performance work, big changes (both good and bad) can result from small adjustments. In the blog posts below, you’ll find someone who has done some excellent experimental science as she adjusts the tuning on a chocolate chip cookie recipe. She starts with a control recipe (Nestle Tollhouse Chocolate Chip Cookie) and she varies one thing at a time and makes a batch. She shows her work, details the techniques, and keeps excellent notes.

So why is a blog on computer performance linking to a cooking blog? Because the things I write about here are the things that will always be true on any computer system. Good experimental technique will always be useful and I’m pretty sure that a huge fraction of people will always gratefully accept a well made chocolate chip cookie. Her work is a delicious example of good science, which is an essential ingredient of all computer performance work.

The Ultimate Guide to Chocolate Chip Cookies – Part 1
The Ultimate Guide to Chocolate Chip Cookies – Part 2
The Ultimate Guide to Chocolate Chip Cookies – Part 3
The Ultimate Guide to Chocolate Chip Cookies – Part 4

Cookie-Testing02TEXT

 

Yogi Berra on Computer Performance Work

berra_yogi1Yogi Berra is one of the best catchers in baseball history. He is also known for his gift for creating memorable, and humorously mangled, quotes…

“There are some people who, if they don’t already know, you can’t tell ’em.”

He was good at what he did, served his country honorably in WWII, and was married to Carmen Berra for 65 years. He clearly has played the game of life well.

Although I doubt Yogi ever worked with computers, some of his “yogisms” seem to apply to the field. Here are a few of Yogi’s thoughts that resonate with key disciplines of computer performance work.

PERFORMANCE MONITORING
    “You can observe a lot by just watching.”

CAPACITY PLANNING
    “Prediction is difficult, especially about the future.”

LOAD TESTING
“In theory there is no difference between theory and practice.
In practice there is.”

PERFORMANCE MODELING
“The future ain’t what it used to be.”

PRESENTING YOUR RESULTS
“I never said most of the things I said.”

Douglas Adams on Our Reactions to Technology

I’ve come up with a set of rules that describe our reactions to technologies:adams

  1. Anything that is in the world when you’re born is normal and ordinary and is just a natural part of the way the world works.
  2. Anything that’s invented between when you’re fifteen and thirty-five is new and exciting and revolutionary and you can probably get a career in it.
  3. Anything invented after you’re thirty-five is against the natural order of things.

Source: The Salmon of Doubt

Correlation Is Not Causality

researchcausation

When gathering performance data it is important to note that correlation does not prove causality. A strong correlation shows the compared values move up/down together, but does not prove (or disprove) a cause and effect relationship. If two performance meters (X and Y) are correlated it could be the case that X causes Y, or that Y causes X, or that changes in both X and Y are caused by some external force.

Notice below how well the Divorce rate in Maine correlates with Per capita consumption of margarine in the US. If you can figure out, and clearly show, how these two variables are causally linked, then a Nobel might be in your future.

correlation

Noticing correlation between the activity you see in various meters is an interesting clue to many performance questions. It’s a good place to start wondering, but before you make up your mind about a causal link between X and Y, gather more data and do it under differing circumstances. Figure out the mechanism for how X causes Y. Look for evidence that supports (and/or destroys) your X causes Y hypothesis.