Ludin Nento @fncode

vxlan-mesh is an agent that runs on hosts within a single layer 2 broadcast network and allows peers to discover each other It provides a stable IP network environment for kubernetes clusters or any other application that is sensitive to changing IPs ➜ github.com/antongisli/vxl…

Salah satu penyebab slow down pada API adalah "excessive database roundtrip", atau bahasa Indonya kebanyakan bolak-balik panggil database. Saya akan tunjukin gimana dengan hanya mengurangi database calls dari API, kita bisa improve performance API kita hingga hampir 2x lipat. Anggaplah kita punya sebuah aplikasi untuk ngerjain quiz online. Ketika user selesai dan mau submit jawabannya, biasanya API akan ngelakuin hal berikut: 1. Query ke database untuk calculate berapa jawaban yang benar dan berapa total poin yang didapat user. 2. Simpan hasilnya ke database kalo ada salah satu jawaban yang benar. API developer (terutama yang menggunakan ORM) biasanya akan melakukan 2 step tersebut di level API, yang berarti API akan melakukan 2 database calls (+ 2 DB calls lagi untuk begin dan commit transaction). Di bawah ini saya contohin menggunakan Rust dan databasenya menggunakan PostgreSQL: Gimana performancenya? Saya coba benchmark menggunakan k6 load testing tool dengan 100 VUs (virtual concurrent users) dan durasi benchmarknya 10 detik. Ini hasilnya: Dengan approach seperti di atas, API kita dapat menghandle 3561 requests per second (RPS). Bisa kita improve ga? Yes. Ada banyak cara untuk minimize database calls. Salah satu cara yang saya suka adalah meng-combine kedua query tadi menjadi 1 stored procedure / user-defined function. API-nya juga jadi lebih simpel, tinggal select from function yang udah kita buat di atas. Tapi gimana performancenya sekarang? Boom, dari 3561 RPS improve drastis menjadi 5995 RPS, nyaris dua kali lipat! Well, ibarat kurir ekspedisi, mengantar 100 barang sekaligus dalam satu trip pasti lebih efisien daripada setiap jalan cuma bawa satu barang. Konsep yang sama berlaku juga di sini. API manggil DB hanya satu kali akan lebih efisien daripada bolak-balik manggil DB sampai empat kali. Untuk yang terbiasa menggunakan full ORM, yes konsekuensinya adalah harus terjun ke raw query untuk bisa achieve performance improvement di atas. Menurut saya agak susah sih kalo full ngandelin ORM. Makannya saya lebih suka pakai raw query atau stored procedures sekarang. Tapi kalo ada yang tau cara achieve behavior di atas hanya dalam 1 database call dengan menggunakan full ORM bisa kasih tau saya sih. I would love the feedback as always. Dah gitu aja, semoga bermanfaat 😉

keyboard_arrow_left Previous keyboard_arrow_right Next

How to handle missing values. The tutorial explains different types of missing data (i.e. MCAR, MAR, and MNAR) and provides example codes in the R programming language: statisticsglobe.com/missing-data/ #RStats #database #RStudio #programming #DataAnalytics #datasciencetraining

Latency controls revenue. Amazon observed that every 100ms of latency cost them 1% in sales. That's a whopping figure when you think about their volumes. If you're forced to optimize just one thing in your system, I’d recommend trying to go for low latency. Here are 8 battle-tested ways to do so: [1] Caching - Use caching to store frequently accessed data - The goal of caching is to minimize costly database lookups and heavy computations [2] CDN - Take caching one step further by using a CDN. - CDNs can help cache content closer to the users This reduces latency. [3] Pre-caching and Pre-rendering - Use predictive techniques to pre-cache and pre-render pages - Since the pages are already rendered, you can serve the incoming requests much faster [4] Database Indexing - Databases are a huge reason for high latency. - Ensure databases are well-indexed so queries can run faster - Also, optimize queries by analyzing and refactoring slow database queries [5] Load Balancing - Distribute incoming traffic evenly across multiple servers with load balancers - This prevents any single server from becoming a bottleneck [6] Async Processing - Implement async processing wherever applicable for long-running tasks. - This lets your system respond quickly to the user while performing further processing in the background. [7] Compression - Compress data before sending it over the network. - This decreases transfer time and reduces the overall latency. [8] Keep-Alive Connections - Reuse TCP connections wherever possible. - This helps avoid the overhead of re-establishing connections for new requests - Also, maintain a pool of established connections Any other tricks to achieve low latency you know of?

Building APIs is fun until you run into performance issues. Support tickets loaded with complaints start flooding your team’s Slack channels or mailboxes. Your manager is breathing down your neck to fix things. Before you know it, the hike you were expecting has evaporated into thin air. There’s no option but to roll up your sleeves and fix those pesky API performance issues. Here are 4 methods that can help you dig out of the mess: [1] Caching If long-running queries are the cause of performance issues, caching might be your secret weapon. By storing frequently accessed data in the cache, you avoid computing the result for every single request. Whether it's data chunks, responses, or entire pages, caching minimizes server load and maximizes speed. But caching is an ocean of possibilities. The right configuration of cache and database relation matters in how well it might work for you. [2] Load Balancing If your application server is overwhelmed by the number of requests, load balancing can be your BFF. With load balancing, you share the load between multiple instances of your application so that no single server is overloaded. Direct benefits are performance and reliability. But load balancing works best when your application is stateless and easy to scale horizontally. Plus, you may have to pick the right load-balancing algorithm. [3] Async Processing Sometimes, you can’t solve every problem at the same time. So, why not park it for later? With async processing, you can let the clients know that their requests are registered and under process. Then, you process the requests one by one and communicate the results to the client later on. This allows your application server to take a breather and give its best performance. But of course, async processing may not be possible for every requirement. [4] Connection Pooling An API often needs to connect to the database to fetch some data. Creating a new connection for each request can degrade performance. It’s good to use connection pooling to set up a pool of database connections that can be reused across requests. This is a subtle aspect but in highly concurrent systems, connection pooling can have a dramatic impact on performance. So - we've looked at 4 reliable methods to improve an API’s performance. But I’m 141% sure there are more tricks available. Can you suggest any other API performance improvement technique you may have used or seen?

A few years old but still an excellent research work on Broadcom Wi-Fi stack exploitation by Gal Beniamini. Part 1: googleprojectzero.blogspot.com/2017/04/over-a… Part 2: googleprojectzero.blogspot.com/2017/04/over-a… Part 3: googleprojectzero.blogspot.com/2017/10/over-a… #wifi #wireless #infosec #cybersecurity #hacking #embedded

keyboard_arrow_left Previous keyboard_arrow_right Next

Years ago when my wife and I we were planning to buy our home, my dad stunned me with a quick mental calculation of loan payments. I asked him how - he said he'd learned the strange formula for compound interest from his father, who was a merchant in 19th century Iran. 🧵 1/4

Project Tracking App in Jetpack Compose (github.com/janewaitara/Pr…) Track Projects. Tech Stack = Hilt, Room, DataStore, Navigation, Material

playwright-go: Playwright for Go a browser automation library to control Chromium, Firefox and WebKit with a single API. ⭐️ 1335 Updated on: 2023-06-23 #golang github.com/playwright-com…

SVG path animation is child’s play when you’re using a no-code tool 🎈 See for yourself 👇 svgator.com/svg-path-anima…

gops: A tool to list and diagnose Go processes currently running on your system ⭐️ 6169 Author: @GoogleOSS #golang github.com/google/gops

/1 Netflix Tech Stack - Part 1 (CI/CD Pipeline) Planing: Netflix Engineering uses JIRA for planning and Confluence for documentation. Coding: Java is the primary programming language for the backend service, while other languages are used for different use cases.

#ONEPIECE1080 #ONEPIECE1080SPOILERS Chapter 1,080 first spoilers.

This is how to find sql-Injection 100% of the time For site.com/?q=HERE /?q=1 /?q=1' /?q=1" /?q=[1] /?q[]=1 /?q=1` /?q=1\ /?q=1/*'*/ /?q=1/*!1111'*/ /?q=1'||'asd'||' <== concat string /?q=1' or '1'='1 /?q=1 or 1=1 /?q='or''=' credit:@TodayCyberNews #BugBountyTips

git-lfs: Git extension for versioning large files ⭐️ 11218 #golang github.com/git-lfs/git-lfs

DevOps Guide by @tikamalma github.com/Tikam02/DevOps… #devops #devsecops #kubernetes #cicd #k8s #linux #docker #sysadmin #automation #Azure #infrastructureascode #cloudcomputing #serverless #terraform #ansible #yaml

Qix: Machine Learning、Deep Learning、PostgreSQL、Distributed System、Node.Js、Golang ⭐️ 14061 #golang github.com/ty4z2008/Qix

cronsun: A Distributed, Fault-Tolerant Cron-Style Job System. ⭐️ 2705 #golang github.com/shunfei/cronsun

Testing DNS resolution with Go buff.ly/3VfEmYU #kubernetes #k8s #docker

#AWS EKS Tutorial | What is #EKS? In this tutorial, we will explain the architecture, components of Amazon Elastic Container Service (EKS), a step-by-step procedure to provision Kubernetes Cluster on EKS. youtu.be/CukYk43agA4 #KodeKloud #Docker #Kubernetes #Microservices