This is a collection of the things I believe about software development. I have worked for years building backend and data processing systems, so read the below within that context.
Agree? Disagree? Feel free to let me know at @JanStette. See also my blog at www.janvsmachine.net.
Keep it simple, stupid. You ain't gonna need it.
Here's a list of mildly interesting things about the C language that I learned mostly by consuming Clang's ASTs. Although surprises are getting sparser, I might continue to update this document over time.
There are many more mildly interesting features of C++, but the language is literally known for being weird, whereas C is usually considered smaller and simpler, so this is (almost) only about C.
struct foo {
struct bar {
int x;This is a short post that explains how to write a high-performance matrix multiplication program on modern processors. In this tutorial I will use a single core of the Skylake-client CPU with AVX2, but the principles in this post also apply to other processors with different instruction sets (such as AVX512).
Matrix multiplication is a mathematical operation that defines the product of
I was talking to a coworker recently about general techniques that almost always form the core of any effort to write very fast, down-to-the-metal hot path code on the JVM, and they pointed out that there really isn't a particularly good place to go for this information. It occurred to me that, really, I had more or less picked up all of it by word of mouth and experience, and there just aren't any good reference sources on the topic. So… here's my word of mouth.
This is by no means a comprehensive gist. It's also important to understand that the techniques that I outline in here are not 100% absolute either. Performance on the JVM is an incredibly complicated subject, and while there are rules that almost always hold true, the "almost" remains very salient. Also, for many or even most applications, there will be other techniques that I'm not mentioning which will have a greater impact. JMH, Java Flight Recorder, and a good profiler are your very best friend! Mea
According to this cloudflare blog article "Load Balancing without Load Balancers", we can build a rock-solid load balancer only using a router. All the magic comes from BGP and Equal-Cost Multi-Path routing.
In this howto, I will use bird as BGP router on linux instance (ie. servers).
I use GNS3 with this architecture :
| Latency Comparison Numbers | |
| -------------------------- | |
| L1 cache reference/hit 1.5 ns 4 cycles | |
| Floating-point add/mult/FMA operation 1.5 ns 4 cycles | |
| L2 cache reference/hit 5 ns 12 ~ 17 cycles | |
| Branch mispredict 6 ns 15 ~ 20 cycles | |
| L3 cache hit (unshared cache line) 16 ns 42 cycles | |
| L3 cache hit (shared line in another core) 25 ns 65 cycles | |
| Mutex lock/unlock 25 ns | |
| L3 cache hit (modified in another core) 29 ns 75 cycles |
| /* | |
| Parallel processing with ordered output in Go | |
| (you can use this pattern by importing https://github.com/MarianoGappa/parseq) | |
| This example implementation is useful when the following 3 conditions are true: | |
| 1) the rate of input is higher than the rate of output on the system (i.e. it queues up) | |
| 2) the processing of input can be parallelised, and overall throughput increases by doing so | |
| 3) the order of output of the system needs to respect order of input | |
| - if 1 is false, KISS! |
FWIW: I (@rondy) am not the creator of the content shared here, which is an excerpt from Edmond Lau's book. I simply copied and pasted it from another location and saved it as a personal note, before it gained popularity on news.ycombinator.com. Unfortunately, I cannot recall the exact origin of the original source, nor was I able to find the author's name, so I am can't provide the appropriate credits.
| /** | |
| * Solves the n-Queen puzzle in O(n!) | |
| * Let p[r] be the column of the queen on the rth row (must be exactly 1 queen per row) | |
| * There also must be exactly 1 queen per column and hence p must be a permuation of (0 until n) | |
| * There must be n distinct (col + diag) and n distinct (col - diag) for each queen (else bishop attacks) | |
| * @return returns a Iterator of solutions | |
| * Each solution is an array p of length n such that p[i] is the column of the queen on the ith row | |
| */ | |
| def nQueens(n: Int): Iterator[Seq[Int]] = | |
| (0 until n) |
Picking the right architecture = Picking the right battles + Managing trade-offs
