RabbitMQ Streams are designed for high-throughput scenarios, but what happens when your ingress rate is low?| RabbitMQ Blog
Modern AI workloads often don’t utilize the full capacity of advanced GPU hardware, especially when running smaller models or during development phases. The AMD GPU partitioning feature addresses this challenge by allowing you to divide physical GPUs into multiple virtual GPUs, dramatically improving resource utilization and cost efficiency.| AMD ROCm Blogs
In this blog post, we walk through how to use Matrix Cores in HIP kernels, with a focus on low-precision data types such as FP16, FP8, and FP4, as well as the new family of Matrix Core instructions with exponent block scaling introduced in the AMD CDNA™4 architecture. Through code examples and illustrations, we provide the necessary knowledge to start programming Matrix Cores, covering modern low-precision floating-point types, the Matrix Core compiler intrinsics, and the data layouts requi...| AMD ROCm Blogs
I was recently invited to speak at the Vancouver Java User Group https://www.linkedin.com/groups/14603101/. I proposed an AMA on performance tuning as in my experience, this turns a presentation into a group discussion which allows everyone to be engaged and that makes it more fun for everyone. The discussion in this session was mostly about methodology, or lack of it. I showed a chunk of code from one of the exercises from my performance tuning workshop. This seemingly simple exercise has st...| Random Notes and Cheat Sheets
OpenZL is a new open source data compression framework that offers lossless compression for structured data. OpenZL is designed to offer the performance of a format-specific compressor with the eas…| Engineering at Meta
2021-07-20 Beating the L1 cache with value speculation| mazzo.li
Blog of Vincent Rolfs. Vincent is a freelance Software Engineer and IT-Consultant. He has more than ten years experience in developing complex applications on the web.| vincentrolfs.dev
While working on Indiana Jones and the Great Circle and later DOOM: The Dark Ages, it became apparent that both titles are particularly sensitive to anisotropic sampling cost on the Xbox Series consoles. This is despite both having very different renderers, Indiana Jones uses what I would best describe as a semi-forward system, while Doom […]| Martin Fuller's Graphics Ramblings
Let me run a thought experiment: “What would you do if your weren’t told what to do?” You know the goals you must achieve, let’s say quota-based, or similar revenue based attainment goals. If you are in a non-selling role, think about project goals/deadlines or other important goals. You have all sorts of resources to […] The post What Would You Do If You Weren’t Told What To Do? appeared first on Partners in EXCELLENCE.| Partners in EXCELLENCE
Yes! The new Visual Studio 2026 edition is available in preview (now called Insiders). I'll take some time this week to walk through some of the features I like and maybe some of the rough edges I discover along the way.| The art of simplicity
In my last article I wrote about some tooltip optimization to reduce the start time of the Unreal Editor by 2-5 seconds. Turns out people do really care about their editor start time. So much that …| Larst Of Us
Learn how to decrease or increase the icon cache size in Windows 11 and Windows 10 using the Registry Editor in simple steps.| WindowsLoop
The Arts Scholars experience Hamilton on tour, after which they join the cast and crew in the front rows for an exclusive talkback.| Arts at MIT
A walkthrough of how I built a custom profiler to measure UI responsiveness, using .NET and Silhouette.| minidump.net
My return performance to Small Batch Cellars! Extra-Sensory Performance: Mind-reading and Paranormal Entertainment by Keith Zalinger http://www.readerofminds.com Recipient of TheBash.com’s “Best of” award for 2012 through 2024 Keith Zalinger is...| RSS Feed
I quite like Nextcloud as a self-hosted alternative to Dropbox - it works well for making documents and password databases available between machines.Where photos are concerned, the functionality it i| www.bentasker.co.uk
I was going to save this announcement for later, but I’m just too excited: Harry Roberts will be speaking at Web Day Out! Goddamn, that’s one fine line-up, and it isn’t even complete yet! Get your ticket if you haven’t already. There’s a bit of a story behind the talk that Harry is going to give… Earlier this year, Harry posted a most excellent screed in which he said: The web as a platform is a safe bet. It’s un-versioned by design. That’s the commitment the web makes to you...| Adactio: Journal
Understanding query planner quirks yielded a ~35% speedup.| Evan Schwartz
Understand individual learning within a new L&D model created by Andrew Gibbons, following four decades in the industry.| TrainingZone
By Brody Weaver Holly Timpener is a non-binary performance artist, facilitator, and PhD Candidate in the Interdisciplinary Studies program at Concordia University. Their extensive body of performance…| Femme Art Review
Small Data Small data appears to be a very exciting movement that is moving the overton window away from Big Data onto much simpler and cheaper solutions ...| topicpartition.io
At Johnny’s Software Lab we’ve spent a lot of time deep-diving into advanced performance topics — vectorization, cache hierarchies, memory bandwidth, you name it. But not everyone is ready to jump straight into assembly listings and microarchitectural details. This post is for the beginners. For the fresh graduates and junior developers who are just starting… Read The post 9 Things Every Fresh Graduate Should Know About Software Performance appeared first on Johnny's Software Lab.| Johnny's Software Lab
I’m participating in an AI Conference where so much of the discussion is around agentic AI. A number of speakers have focused on the concept of “Real time AI coaching.” I’m also seeing a lot of this in many of the new AI selling platforms, it seems to be THE next thing in driving seller […] The post The “Miracle” Of Real Time AI Coaching appeared first on Partners in EXCELLENCE.| Partners in EXCELLENCE
There are books & many articles online, like this one arguing for using Postgres for everything. I thought I’d take a look at one use case - using Postgres instead of Redis for caching. I work with APIs quite a bit, so I’d build a super simple HTTP server that responds with data from that cache. I’d start from Redis as this is something I frequently encounter at work, switch it out to Postgres using unlogged tables and see if there’s a difference.| Dizzy zone
It’s been a long time since I left a theater with such happiness, filled with music and almost dancing to my car. It’s been a long time since the music […]| The Georgetowner
We’re into the full-on spirit and fun of the fall season now. Arts institutions are debuting their season openers; there are galas galore and the Nationals celebrate the last home […]| The Georgetowner
Our work in selling is tough. There are details to manage, conflicts to resolve, shifts in direction. Some of the work is tedious, even unpleasant. Much of it is exhausting. When the work gets tough, our natural reaction is avoidance. At the same time, we mask that avoidance with busyness and activity. We avoid the […]| Partners in EXCELLENCE
One of the significant improvements in PG18 is AIO. What are some basic tuning recommendations?| Tomas Vondra
A description of the performance optimizations made to the free-threaded garbage collector for Python 3.14.| labs.quansight.org
A database engineer at QuestDB discovers that an AI-suggested optimization for array shape calculation in their Parquet reader actually made the code slower, and achieves a 5x average speedup by applying a few simple optimizations.| QuestDB Blog
QuestDB's new TOLERANCE clause for ASOF JOIN ensures you only join with relevant, recent data, preventing stale joins in time-series analyses. This synergises perfectly with recent upgrades to our ASOF JOIN algorithm, helping to bring you accurate results with low latency.| QuestDB Blog
After many years of waiting, we benchmarked InfluxDB 3 OSS Core (Alpha). How does it fare? What are its strengths and weaknesses? We take an initial look, benchmark ingestion, and spot some caveats..| QuestDB Blog
Unravel a mysterious network bandwidth issue in QuestDB's primary-replica replication was identified and resolved. Learn about the tools and techniques used, including Rust for packet capture and Python with Polars for data analysis, to optimize network performance.| QuestDB Blog
Learn about AS OF JOIN, a powerful feature that lets you correlate events in time-series data. It's that rare moment where the database will just Do What You Mean, in a single keyword!| QuestDB Blog
Learn about SIMD, its mnemonics, registers and instructions. We'll demonstrate how these parallel processing techniques enhance database query speeds through an accessible walkthrough. Impress your programmer friends!| QuestDB Blog
Explore the ingenious method behind the One Billion Row Challenge's fastest temperature parsing technique, featuring a deep dive into Quân Anh Mai's (@merykitty) optimization code without loops or if statements, leveraging bitwise operations and ALU magic for unparalleled efficiency.| QuestDB Blog
I took part in the Billion Row Challenge. Enjoy a deep, step-by-step summary of how you get from a Parallel Java Streams implementation that takes 71 seconds to a super-optimized version that takes 1.7 seconds. Example code and walkthroughs included!| QuestDB Blog
Why is a fast hash table important for optimal SQL performance? We answer this question and explain how the QuestDB team designed FastMap, our hash table specialized for SQL execution.| QuestDB Blog
A guide to adding Rust to a Java codebase with JNI and the rust-maven-plugin.| QuestDB Blog
The story of how QuestDB optimizes the Time-Series Benchmark Suite.| QuestDB Blog
Key Performance Indicators are meant to achieve one thing, sharpen our focus. They help translate our strategies into focused execution. The KPIs help us understand the things that are most important in achieving our goals. They serve to align the organization around these critical goals and activities. They give everyone involved in executing the strategies […] The post KPIs, When You Measure Everything, You Measure Nothing! appeared first on Partners in EXCELLENCE.| Partners in EXCELLENCE
We are trained to ask questions. But our orientation is to ask questions that demand answers. As a result, conversations often resemble a ping pong game. The server starts with a question, then the response, back and forth. Then the other person serves, asks a question, expects a response, back and forth. A lot of […] The post Ask Questions That Require Thinking appeared first on Partners in EXCELLENCE.| Partners in EXCELLENCE
Emmanuel Houcou, a promising French sprinter from Martinique, joins Q36.5 Pro Cycling Team on a three-year deal to develop in sprints, classics and puncheur races. The post Emmanuel Houcou joins Q36.5 Pro Cycling Team appeared first on Q36.5 Pro Cycling Team.| Q36.5 Pro Cycling Team
British rider Thomas Gloag signs with Q36.5 Pro Cycling Team for 2026–2027. A versatile climber with untapped potential, aiming for consistency and growth. The post Thomas Gloag Joins Q36.5 Pro Cycling Team appeared first on Q36.5 Pro Cycling Team.| Q36.5 Pro Cycling Team
Tom Pidcock finishes 3rd overall at La Vuelta 2025, giving Q36.5 Pro Cycling Team a historic Grand Tour podium as a ProTeam underdog. The post Tom Pidcock Secures Vuelta Podium appeared first on Q36.5 Pro Cycling Team.| Q36.5 Pro Cycling Team
Tom Pidcock defends third place overall at La Vuelta 2025 after a dramatic Stage 20 finish on Bola del Mundo. Historic podium in sight. The post Tom Pidcock Secures Third Place Overall appeared first on Q36.5 Pro Cycling Team.| Q36.5 Pro Cycling Team
Tom Pidcock consolidates third place overall after time trial. He took time on all riders behind him on the overall classification, on those fighting for the podium with him. The post Tom Pidcock consolidates third place overall after time trial appeared first on Q36.5 Pro Cycling Team.| Q36.5 Pro Cycling Team
om Pidcock finishes 2nd on Stage 17 of La Vuelta atop El Morredero, strengthening his GC podium position with Q36.5 Pro Cycling Team. The post La Vuelta Stage 17: Pidcock 2nd on El Morredero appeared first on Q36.5 Pro Cycling Team.| Q36.5 Pro Cycling Team
Inside the hidden 1%: how Q36.5, Scott, Amacx and our performance partners fuel marginal gains that keep Tom Pidcock in podium contention at La Vuelta. The post The Last One Percent | Q36.5 Pro Cycling Team at La Vuelta appeared first on Q36.5 Pro Cycling Team.| Q36.5 Pro Cycling Team
Learn what progressive enhancement is based on a real world example within this blog post.| Lusitos Tech Blog
David Lattimore - 2025-09-23| davidlattimore.github.io
The Saudi Arabian artist displays performance work alongside Iranian Shirin Neshat in Cartographies of Presence in London| 1854 Photography
At my job, I’m refactoring a 30k LOC codebase that simulates learning in the mammal brain. The emergent behavior of these large brain models is hard to test for, so we opted to take the safe route and preserve bit-equality in the weights of the trained model to guarantee that we are not breaking anything.The main downside of hash-based regression testing during refactoring is that it doesn’t check numerical stability. The most common change we applied was inlining many small functions to ...| siboehm
In this post, I’ll iteratively optimize an implementation of matrix multiplication written in CUDA.My goal is not to build a cuBLAS replacement, but to deepl...| siboehm.com
Storing Unwise Amounts of Data in JavaScript Bigints This is a short note to document storing data in JS’s bigint type. I tried this out in a project recently, and it’s not obviously a terrible idea, as long as you take into account all the caveats that make it only useful for very specific cases. Why This is a performance1 thing. In the project where I was exploring this, we had a large amount of objects representing configuration values. The possible configuration keys were the same, bu...| jonathan-frere.com
A tale of how elegant SSE code passed code review, worked perfectly locally and in staging, but nearly brought down our production servers when thousands of users connected simultaneously during a real DDoS attack.| cekrem.github.io
At Johnny’s Software Lab we’ve spent a lot of time deep-diving into advanced performance topics — vectorization, cache hierarchies, memory bandwidth, you name it. But not everyone is ready to jump straight into assembly listings and microarchitectural details. This post is for the beginners. For the fresh graduates and junior developers who are just starting… Read| Johnny's Software Lab
In this blog, we will provide step by step instruction on how to reproduce AMD's MLPerf Inference v5.1 Submission| ROCm Blogs
You can use multiple plugins for caching, image optimization, and a Content Delivery Network but there's also a cloud-based solution that does it all. In this Nitropack review, I'm going over its functionality, performance, and cost-effectiveness for WordPress websites. The post "NitroPack Review: WordPress Site Speed Optimization Made Easy" first appeared on WP Mayor.| WP Mayor
Why UGC sells and should be your top marketing priority and how businesses can harness it to amplify their advertising efforts.| Billo
Sleep and exercise are two pillars of a healthy lifestyle, but what happens when they clash? If you’ve ever tossed and turned all night only to wake up groggy and wonder whether you should still hit the gym, you’re not alone. The question of whether to exercise when you don’t get enough sleep is a common dilemma. On one hand, physical activity can boost energy and mood; on the other, pushing your body when it’s already running on empty might do more harm than good.… The post Running...| Whole Life Challenge
Benchmarking software is a tricky business. For robust results, you need dedicated hardware that only runs the benchmarking suite under controlled conditions. No other processes! No OS updates! Nothin| Quansight Labs
At Grab, our decade-old Go monorepo had become a 214GB monster with 13 million commits, causing 4-minute replication delays and crippling developer productivity. Through custom migration tooling and strategic history pruning, we achieved a 99.9% reduction in commits while preserving all critical functionality. The result? 36% faster clones, eliminated single points of failure, and a 99.4% improvement in replication performance—transforming our biggest infrastructure bottleneck into a develo...| Grab Tech
When the Atlas Monkey encounters a massive cache invalidation event threatening reality itself, the crew must perform the galaxy's most dangerous database migration to save a trapped vessel and prevent a cascade failure across three sectors.| Seuros Blog - Navigation Logs from the Ruby Nebula
When the backend developer said "screw it, let's just give the frontend direct database access through GraphQL," the Atlas Monkey fleet faced its worst performance disaster yet. Ships crashed from N+1 query storms, security breaches exposed critical data, and what seemed like developer convenience became a maintenance nightmare. Captain Seuros investigates why GraphQL often creates more problems than it solves.| Seuros Blog - Navigation Logs from the Ruby Nebula
Captain Seuros discovers that small Ruby modernization gains compound exponentially - from 0.58% improvements in state_machines to fleet-wide performance that challenges Rust battleships. The mathematics of incremental engineering and strategic patience.| Seuros Blog - Navigation Logs from the Ruby Nebula
After years of comfortable cruising through the Ruby nebula, I discovered something that changed everything. This is the story of why a die-hard Rubyist decided to learn Go—and what I found on the other side.| Seuros Blog
Smith Concepts' bike is a fusion of style and performance—a personal interpretation of a genre that holds a special place in Jared's heart.| Hot Bike Magazine
Serilog is one of the most popular structured logging libraries for .NET, offering excellent performance and flexibility. While Serilog comes with many built-in sinks for common destinations like files, databases, and cloud services, we created a custom sink to guarantee compatibility with an existing legacy logging solution. However as we noticed some performance issues, we decided to rewrite the implementation to use a batched sink.| The art of simplicity
Adams State will make its presence known downtown during First Friday starting at 5 p.m. and ending when the last ember dies and the smoke clears from the bonfire. The post Homecoming brings back favorites and adds new events appeared first on Adams State University.| Adams State University
“It’s a fun way to kick off our 100th year and the Halloween season,” Dr. Taylor said. The post Adams State Theatre presents Stephen King’s Misery appeared first on Adams State University.| Adams State University
The Moonbit team recently published a blog post claiming their language runs "30% faster than Rust" for FFT workloads. This is a lie by omission. They benchmarked against a deliberately crippled Rust implementation that no competent programmer would write.| bitemyapp.com
Boost desktop app development with faster builds, safer multithreading, plug-in design, text configs, and shared libraries for performance & maintainability.| KDAB
Stop blocking Ruby threads with Kafka. WaterDrop's transparent async support using fibers delivers 5x throughput without callbacks or complexity.| Closer to Code
If there is one thing the Unreal Engine doesn’t suffer from, it is it’s lack of features. Over the past years it has been transformed from “just” a game engine into an ̵…| Larst Of Us
LogSeam transforms log management with Tigris, achieving massive cost savings and performance improvements in their security data lake architecture.| Tigris Object Storage Blog
Lubegard has introduced two new products: Lubegard Complete Type F Automatic Transmission Fluid and Type F ATF Protectant.| Transmission Digest
I will be presenting our new paper, “HoliPaxos: Towards More Predictable Performance in State Machine Replication,” at the VLDB’25. Feel free to ping me if you are there and want to chat!| Aleksey Charapko
In the rapidly-evolving world of artificial intelligence (AI) and machine learning (ML), speed and scalability are paramount. The ability to process massive amounts of data in real-time is a critical requirement for organizations looking to leverage AI/ML for competitive advantage. Whether it's training large machine learning models, running complex| MinIO Blog
Suppose that you have a long string and you want to insert line breaks every 72 characters. You might need to do this if you need to write a public cryptographic key to a text file.| Daniel Lemire's blog
Posted by Mayuri Khinvasara Khabya - Developer Relations Engineer (LinkedIn and X)| Android Developers Blog
The "big data king" that enterprises spent billions on is dying. Not from lack of trying, but because cloud-native alternatives now beat it at its own game. AIStor is faster and cheaper than HDFS. The revolution is happening in quarters, not decades.| MinIO Blog
After a user signaled a performance issue in a Matrix client, we have added new tracing timers in the Matrix Rust SDK to spot the problem. Once found, we have fixed an SQL query improving the throughput from 19k to 251k events/sec, and the speed from 502ms to 39ms. Then after another creative patch, the throughput has been improved to 4.2M events/sec, and the speed to 2ms.| mnt.io
Critical sections don’t have to be scheduler-bottlenecked.| protty
Sometimes you can trick the compiler into generating more efficient code. How can you test this optimization continues to be applied?| Python⇒Speed
The Apple M1 ARM processor family incorporates two memory consistency models: the conventional ARM weak memory ordering and the Total store ordering (…| www.sciencedirect.com
How often can you honestly say your L&D efforts have created lasting behavior change – not just course completions? With four decades of learning expertise, Andrew Gibbons presents his Six Is of Learning Model: a practical framework that helps L&D cultivate sustained performance that learners, ultimately, drive themselves.| TrainingZone
Do you know why React projects fail? React and Angular are "mix & match" frameworks that lead to further complications down the line. Find out why.| The Isomorphic Software Blog
Posted by Alan Kelly, Software EngineerOne of our previous articles, Optimizing TensorFlow Lite Runtime Memory, discusses how TFLite’s memory arena minimizes memory usage by sharing buffers between tensors. This means we can run models on even smaller edge devices. In today’s article, I will describe the performance optimization of the memory arena initialization so that our users get the benefit of low memory usage with little additional overhead.| The TensorFlow Blog
Intro| steelcake.com
AIStore + HuggingFace: Distributed Downloads for Large-Scale Machine Learning| AIStore
AIStore v3.28 introduces a unified rate-limiting capability that works at both the frontend (client-facing) and backend (cloud-facing) layers. It enables proactive control to prevent hitting limits and reactive handling when limits are encountered — all configurable at both the cluster and bucket levels, with zero performance overhead when disabled.| AIStore
In our daily work with talent leaders and solution providers, we run into some incredible technology. These tools available to organizations right now can help them hire, develop, and retain their people, and the platforms and systems are amazing in their ability to support intelligent decisions, personalized actions, and more. According to nearly 10 years […]| Lighthouse Research & Advisory
Note: Our team is working closely with UNLEASH this year on a series of activities, including a quarterly column penned by yours truly. This is the first in the series as I bring research to show that the newest fad, generative AI, may have long-lasting and dangerous side effects. What if the tools used by […]| Lighthouse Research & Advisory
The Performance Myth reduces individuals to commodities or “performers,” whose value is contingent on meeting predefined metrics. This leads to a workplace culture where employees are c…| Paul Taylor
Chances are you’re already on your way towards smashing your degree with flying colours. However, learning how to handle your money is a critical skill that you’ll need to know as a student. Whether you squirrel your precious cash away inside your mattress ‘gangsta’ style or embrace the new age with the wealth of money-saving […] The post 9 Money-Saving Tips For New Students appeared first on BIMM University Blog.| BIMM University Blog
Did you know that macOS has a secret setting that can make Rust builds faster? It can also make Rust tests faster. It probably even has similar effects for other compiled languages such as C, C++, Go, and Swift. It sounds crazy, but read on…| Nicholas Nethercote
How we made Kagi Assistant load twice as fast| jacobwinters.com
Dens Sumesh - Software Engineer| densumesh.dev
I deal with a lot of servers at work, and one thing everyone wants to know about their servers is how close they are to being at max utilization. It should be easy, right? Just pull up top or another system monitor tool, look at network, memory and CPU utilization …| Brendan Long