Typically there are NUMA-aware memory allocators. I don't believe glibc is but mimalloc and the non gperftools tcmalloc should be. The Linux kernel is NUMA-aware and will try to avoid shifting work to a different NUMA node but it may depending on what else is happening - it's a generic algorithm trying to balance CPU utilization, latency, and memory bandwidth. If you're using a higher level language like Go or Java then there may be more involved. Java has a NUMA aware allocator while Go's design requires it to also have a NUMA aware scheduler which I don't believe it does.
You could go extreme and start pinning your threads to specific CPUs to tune the code manually instead of relying on the kernel if you know memory bandwidth is extremely important and you won't have much CPU contention to worry about in terms of getting work scheduled in a timely manner.
All that being said, you typically also need to design your application from the ground up to be NUMA aware to take full advantage so that you can set up your allocations to happen on the right zone & whatnot.
The short answer is a) it's interesting because the author is very notable in the .NET space and b) he's not exactly known for admitting to being wrong about anything.
These words hit close to home. My dad just passed away from cancer/diabetes in Florida and had to endure the "barbaric torture of decay and failure". Basically 4 months suffering in bed until he eventually denied eating anything and his liver failed. I asked the nurses repeatedly if there was anything to help him go or pass and there was nothing. Something's got to change in the US, we treat animals better than humans at the end of life
Same situation in Italy. Unfortunately. Anyone wanting a be euthanised has to travel Switzerland, which is not always feasible at end-of-life (also considering that the whole procedure costs ~20k$).
I spent my civil service as an EMT and only once I saw a doctor giving the daughter of a terminally ill woman a piece of advice I will never forget (this woman was in such bad conditions, she simply wasn't there anymore and spent 100% of her time with morphin-induced allucinations): after giving her the morphin prescription he told her that "nobody is going to go behind you if you give her an overdose" - implying that she could end her suffering but she had to be the one killing her.
It was by far one of the hardest moment of my life, and I was just a passive witness. It was just brutal.
I've heard similar. Also with family members being taught how to use the morphine IV dripper, and what a fatal dose looks like.
The alternative is for patients to refuse food and water until they die.
I'd been thinking about this a lot because my father was diagnosed with a terminal illness 3+ years ago. We'd discussed euthanasia, and he definitely didn't want to spend his final 6-12 months suffering.
Because he didn't die, the docs took another look and realised he'd been misdiagnosed (and yes, we got 2nd & 3rd opinions after the initial diagnosis).
Insane. In your case, I'm glad your dad was misdiagnosed for the good. I can't imagine how this misdiagnose impacted your life and your family during the last few years.
I'm sorry you had to experience that. Some states do allow euthanasia, it's not a federal ban. Unfortunately by the time you know you need it, mobility is often difficult. I'm still young and I desperately hope something changes before I have to answer this question myself. If not, I may move to a better state if I'm lucky enough for the decline to be slow. Meanwhile I'm frightened for my family and loved ones.
Yes, I don't think people understand that the last step of a slow death is starvation until the person stops breathing, at which point, you must fight the urge to revive them. It's brutal. I just watched it happen before my very eyes this week. I will never forget it.
How does that qualify as a software bug? Perkin-Elmer's custom null corrector was misaligned so the mirror was figured into the wrong shape. Edit: if anything, it was an organizational failure—PE chose to ignore other measurements that showed the mirror was the wrong shape.
The r6i.metal instance says up to 50 Gbps on Network bandwidth: https://aws.amazon.com/ec2/instance-types/r6i/