Design for simplicity : software options

xkcd.com

(I was highly inspired by Evan Martin post before trying to recap in this post)
Think twice before adding an option to a software; options are :

  • bad user experience : good software just works without asking to many things to the user.
  •  a maintenance cost : an option is a feature you have to maintain/test for the lifetime of the software
  • documentation cost : every option, even the most cryptical, should be documented to the user, explained and you should provide some examples of use. When you normally try to do this you end up removing the option from documentation
  • normally options are a way to delegate to user a decision that the designer failed to take : “shall we open on the last opened document or shall put an options ‘always open on blank document’ ?
  • remember the paradox of choice
  • options setting generates bad GUIs : they are hard to present, group or organize.

On the other hand consider that any change you make will probably break someone’s workflow, no doubt, so the temptation to create an option to enable the previous operational mode is high.

The old windows search wizard, awesome, read this : https://www.joelonsoftware.com/2000/04/12/choices/

WTF/Minute : code quality measure unit

 Some day your code will be reviewed, modified or even just read to understand the underlying algorithm by someone else that is not you that wrote the code….

My point here is that readability/maintainability of code is by far the biggest quality of a piece of software. Being able to read the code and understand it rapidly makes everything easier, either bugfixing or modifying/enhancing the code. Think that the next one that might be reading the code might not be as experienced as you with that language so

  • avoid using  unnecessarily complicated constructs that feed you coder ego but slow down the maintenance activity. You are a good coder, no doubt, but if your code is not readable by the average coder that will have to then you are not making a good service to your colleagues
  • Make clear what your code is doing and, if necessary use a comment to clarify what is not evident from the code (I’m not a huge fan of comments because most of them don’t tell anything more than the code). Use comments to refer to a requirement/user-story that is being implemented.
  • Stick to the language style guidelines that were set for your team/company
  • Tests are code too and often are read/reviewed more than real code : don’t fuck up in tests code just because “this is just a test, this code will not run in production”

Discussion on code quality is huge of course and these are simple rules that will impact on maintainability of your code (since a reviewer is doing the same activity of the next coder which will have to work on your code : read it). If you want to dig deeper into code quality measurement I’ll suggest this work from Idan Amit and Dror G. Feitelson really interesting : https://arxiv.org/pdf/2007.10912.pdf

Comic comes from here

Milan skies after fires in the alps, 2018

Hash functions – efficency and speed

A hash function is any function that can be used to map data of arbitrary size to data of a fixed size. The values returned by a hash function are called hash values, hash codes, digests, or simply hashes. Hash functions are often used in combination with a hash table, a common data structure used in computer software for rapid data lookup.

If you happen to need a non cryptographic hash function (and you might need it even if your are using languages that have builtin hashtables like java or c++ ) here are some references taken from various parts:

Interesting comparison of different hash tables :

http://www.eternallyconfuzzled.com/tuts/algorithms/jsw_tut_hashing.aspx

Some more hash functions here :

http://www.cse.yorku.ca/~oz/hash.html

Some really interesting benchmarks strchr.com/hash_functions by Peter Kankowsky.

And again hash benchmarks sanmayce.com/Fastest_Hash/.

Hash benchmarks in Rust by https://medium.com/@tprodanov/benchmarking-non-cryptographic-hash-functions-in-rust-2e6091077d11

And more benchmarks : https://medium.com/logos-network/benchmarking-hash-and-signature-algorithms-6079735ce05

Some (a mininum set) benchmarks ran using random keys of random len, 1.5 multiply factor on hash size, next power of 2 hash sizes :

Tests are run on random keys
--- hash_function speed on random long (100-350) keys
Keys    HashFunc         HashSize AvgTime(ns) Collisions LongestChain DimFactor
1000000 djb_hash         4194304    191.03      110620    6            1.500000
1000000 murmur64a_hash   4194304    60.87       110830    7            1.500000
1000000 wyhash_hash      4194304    36.69       110170    5            1.500000
1000000 elf_hash         4194304    444.25      110488    5            1.500000
1000000 jen_hash         4194304    143.90      110471    5            1.500000
1000000 djb2_hash        4194304    189.63      111048    6            1.500000
1000000 sdbm_hash        4194304    247.92      110087    6            1.500000
1000000 fnv_hash         4194304    247.87      110170    6            1.500000
1000000 oat_hash         4194304    309.55      110538    6            1.500000

--- hash_function speed on short keys (10 to 45 char len)
Keys    HashFunc        HashSize AvgTime(ns) Collisions LongestChain DimFactor
1000000 djb_hash         4194304    31.14       110363    6            1.500000
1000000 murmur64a_hash   4194304    28.24       110264    6            1.500000
1000000 wyhash_hash      4194304    21.84       109798    5            1.500000
1000000 elf_hash         4194304    49.88       110845    5            1.500000
1000000 jen_hash         4194304    33.28       110541    5            1.500000
1000000 djb2_hash        4194304    31.21       110664    6            1.500000
1000000 sdbm_hash        4194304    35.60       109906    6            1.500000
1000000 fnv_hash         4194304    35.59       109940    6            1.500000
1000000 oat_hash         4194304    43.28       110123    6            1.500000

Tests done with the hash table code you can find github.com/paulborile/clibs

And last a research study for SIMD optimized hash functions : arxiv.org/pdf/1612.06257.pdf  some theory and code github.com/google/highwayhash/blob/master/c/highwayhash.c. The all use the vector extensions present in gcc.

Photo by Luca Campioni on Unsplash

Code benchmarks : how can I measure how fast my software is (and make it faster) ?

You probably heard many times a coder say “this way is faster” : nothing can be more wrong nowadays unless you prove your statement by measuring the same piece of code running with algorithm A and algorithm B (the faster one). And there were times when you could measure the speed of the execution time of a program 1, 2 or 10 times without noticeable variability : those times are gone. Measuring how fast a piece of software runs is about getting a set of results with the lowest possible variability.

Variability in computing time in modern computer architectures is just unavoidable; while we can guarantee the results of a computation we cannot guarantee how fast this computation will be :

“Computer can reproduce anwsers, not performance” : Bryce Adelstein Lellback, https://youtu.be/zWxSZcpeS8Q?t=6m45s

Reasons for variance in computation time can be recap in :

  • Hardware jitter : instruction pipelines, cpu frequency scaling and power management, shared caches and many other things
  • OS activities : a huge list of things the kernel can do to screw up your benchmark performance
  • Observer effect : every time we instruments code to measure performance we introduce variance.

Also warming up the cpu seems to have become necessary to get meaningful results. Running hot instead of cold on a single piece of code is well described here :

https://youtu.be/zWxSZcpeS8Q?t=18m51s

This said the process for benchmarking an piece of code is :

  • Write some benchmark code : define a subset of execution (let’s call it op1) in your code and measure how many op1/sec. Do the same for other parts (op2, op3) but be sure that all parts can be run indipendently from each other   
  • Assure that samples population that we get is “normal” distributed

In details I use C code to measure and :

  • Use GNU Scientific Library (gsl) for mean, median, variance, stdev, skew running stats
  • Use CLOCK_MONOTONE clock_gettime() mode. An example code here.
  • Check if the set returned has ‘normal’ distribution : find the algorithm you like here (I use mean-median/max(mean,median) < 1%, not really correct but ok for my purposes).
  • Narrow down as much as possible the size of the code you are measuring
  • Warm up the cpu before taking samples to avoid having too much variability. You do this by running the code you want to measure many times before actually taking samples. Ideally you should run it until you have normal distributed results. If you don’t then probably the piece of code you are measuring is too big and you are experiencing the effects of other perturbations (os ? io ?).
  • Consider using a benchmark framework : google benchmark (c++) is great for this but I’m afraid you will have to measure C++ code.
  • Lock the cpu clock (if you can) on the host you are using for benchmarks.

Prepare to see strange results 🙂

gcc/clang -fsanitize is saving lifes

 There where times when you had to write C/C++ code and find all the bounds errors and memory leaks by hand or with ancient tools (who remembers Purify?). Valgrind introduced a lot of features but sanitize features added in gcc/clang are just awesome (thanks google guys for this). Just add -fsanitize=address both to compilation and link of your unit tests (we use google test framework), execute the test and get valuable info on memory leaks and out of bounds access to data. In detail :

Testing and inclusion of sanitize is under way in our lab : more info on this in a future post. Work from this team led to the inclusion in Golang of the Data Race detector.