Monthly Shaarli
December, 2023
We investigate the unusual way memory subsystem interacts with branch prediction and how this interaction shapes software performance.
In the heat of an enterprise deal moment, it’s easy to think very short-term about the long-term costs of one-off specials and “small requirements.” There’s tremendous pressure to maximize the importance of a feature tweak to close this quarter’s big deal, and similar pressure to minimize both
From a person who really shouldn't be giving others advice.
Contribute to chaseSpace/k8s-tutorial-cn development by creating an account on GitHub.
The title of this article is misleading. It’s not actually very important for an Engineering Manager to use career laddering, per se, or my process. It is,
维特比算法(Viterbi Algorithm)是一种动态规划算法,通常用于在隐马尔可夫模型(Hidden Markov Model,HMM)或类似的概率图模型中找到最可能的状态序列。这个算法的主要应用包括语音识别、自然语言处理、编码理论等领域。
在HMM中,观察到的数据序列通常被认为是由一个不可见的状态序列产生的,而维特比算法的目标是找到给定观察数据的最可能的状态序列,即具有最大似然性的状态序列。
维特比算法通过在状态转移概率和观察概率之间进行动态规划,逐步构建最可能的状态序列。它具有高效的计算性能,特别适用于处理长序列和大状态空间的问题。
將維特比演算法想像成一個人,即使有些單字拼字錯誤,也能找出正確的故事。他們透過查看附近的單字並糾正錯誤來做到這一點,以使故事更有意義。
👋 Intro to Irina I interviewed Irina Stanescu, an ex-Uber Staff Engineer and writer of The Caring Techie Newsletter . What makes Irina special? Across Google and Uber, Irina progressed from a new-grad engineer to Staff Engineer, tech lead, and manager within 8 years.
Recently, on LinkedIn, I read a post about an engineer who was surprised that his new, optimized version of a parser was slower than the original. The optimization consisted of removing the branches, which are the source of all evil according to the common knowledge in the street, right? His new version was slower, and a benchmark opened his eyes.
We'll break down the fundamentals of automated testing, software design that enhance testability, and explore testing tools and frameworks.
In his book "Drive: The Surprising Truth About What Motivates Us," Daniel Pink talks about "motivation 3.0", which comes after basic needs are covered ("motivation 1.0") and carrots and sticks ("motivation 2.0"). There are three main components in Pink's theory: • Autonomy to be in control of our destiny — how do we work? Is there a dr...
Even when writing is not in your main wheelhouse
Slides and other materials from CppCon 2023. Contribute to CppCon/CppCon2023 development by creating an account on GitHub.
蠻有趣的案例。文章提到了儘管使用託管的Kubernetes叢集,在叢集level上尋找問題時,仍然需要擁有相關知識。
nginx location match visible
A Deep Dive into Performance Tuning for the Intel® Xeon® CPU Max Series
在台灣的話,如果你問我建議,就是不要去新創公司。會去新創公司的人,其人格特質通常不需要等待他人的建議。
We've all been there: the trains you're servicing for a customer suddenly brick themselves and the manufacturer claims that's because you...
We demonstrate a high-performance vendor-agnostic method for massively parallel solving of ensembles of ordinary differential equations (ODEs) and stochastic differential equations (SDEs) on GPUs. The method is integrated with a widely used differential equation solver library in a high-level language (Julia's DifferentialEquations.jl) and enables GPU acceleration without requiring code changes by the user. Our approach achieves state-of-the-art performance compared to hand-optimized CUDA-C++ kernels while performing 20--100$\times$ faster than the vectorizing map (vmap) approach implemented in JAX and PyTorch. Performance evaluation on NVIDIA, AMD, Intel, and Apple GPUs demonstrates performance portability and vendor-agnosticism. We show composability with MPI to enable distributed multi-GPU workflows. The implemented solvers are fully featured -- supporting event handling, automatic differentiation, and incorporation of datasets via the GPU's texture memory -- allowing scientists to take advantage of GPU acceleration on all major current architectures without changing their model code and without loss of performance. We distribute the software as an open-source library https://github.com/SciML/DiffEqGPU.jl
Projects and Ramblings
CRDT
Again, you can't, but you can fake it.
My principles for shipping fast and protecting momentum developed over ~12 years of building software.
卡尔曼滤波器(Kalman Filter)是一种用于估计系统状态的数学方法。它通过将测量数据与系统动态模型相结合,提供了一种有效的方式来估计隐藏状态的值,即使测量数据包含噪声。
卡尔曼滤波器广泛用于各种应用,包括导航系统、控制系统、机器人技术等领域。它的主要优点是能够在不断收到新的测量数据时,实时地更新状态估计,以提供更准确的状态估计值。
將卡爾曼濾波器想像為一個對某物位置做出正確猜測的人。猜測時會考慮新舊資訊。
Master the art of annual planning for your engineering organization. Use the steps in this article to create a strategy that helps you hit the ground running, embrace flexibility, and involve your teams for alignment. And: Get a FREE planning template!