I have no idea what the format of a pointer-to-member is. It sure looks like a small closure.
9,763 karma · joined July 28, 2008
Performance of high-scale/low latency systems, counter-censorship systems, embedded systems, distributed VR systems, web browsers, compilers, some kernel work, even the occasional webapp.
Contact: lally.singh AT gmail
I have no idea what the format of a pointer-to-member is. It sure looks like a small closure.
You have three areas to study:
1. Measurement - makes you define the performance you're looking for and measure it. Until you do this it's mostly a bullshit "make people stop complaining about performance" errand that's too wishy washy to do with more than a few stabs in the dark. With containers and decent capture of samples of your load, a benchmark is pretty straightforward to set up.
2. Modeling - these models are usually little more than measured rates and latencies applied to Little's Law. Pocket-calculator math is often good enough. At worst, an M/M/1 queue.
3. Instrumentation - Figuring out how to attribute your computer's resources (memory, CPU time, iops, etc) to different parts of your code. Tracing libraries, Linux perf, and ebpf can be useful here.
There are a decent number of computers performance books. I like the ones by Jain (great, but AFAICT out of print) and Harchol-Baltar. For work, you shouldn't read them straight through but iterate through parts as you better understand the problem you're trying to solve and start choosing strategies. For the tactical side. Brendon Gregg (sp?) has some decent measurement tool books. Figure out what you want to improve and how to measure that. Then start attributing the existing performance to implementation choices that you can control. Then control those choices (e.g. change algorithm, load balance better, make design trade-offs) to improve performance.
No, you can't live your entire life there, but that was never a great idea. Instead, it's great for software development activities. Org mode and magit are very hard to match outside of EMACS.
The primary things you do in an IDE are debugging, browsing code, writing code, and then writing documentation. With LSP, it's now fantastic at all of those. I can write documentation easily with org mode, and GUD is a really good debugger. And I can extend GDB with Python, to simplify the common stuff I often have to do with my own data structures and types.
With emacs, a web browser, and a terminal, I can do effectively anything I need to do on a computer and do it well.
The hit to Azure's brand for this is probably significant. Who was looking at a migration to Azure before who's now discouraged?
I needed really high quality performance telemetry ("profiling" is a mixed choice) for my PhD. You specify the payload type in an IDL and it generates types and transport code. Fill in a struct and hit save(). If the process is being observed, this'll copy the struct to a lock-free ring buffer that the observer injected in. Includes CPU performance counter support.