title:home
aliases:[alonge.dev, aiops w/ george]
tags:[index, vault-root]
updated:2026-03-06
hi, i'm george.
i'm an engineer, certified hypnotherapist, technical writer, and guitarist. i write about aiops, gpus, devops, kubernetes, and the small reliability lessons i pick up along the way. occasionally, i also write about my personal take on philosophy. this is my open vault.
aiopsgpudevopskubernetesreliabilityphilosophy
##latest essays
cuda toolkit: compilers, libraries, and the host-device model
the software stack that makes gpu programming possible - nvcc compiler pipeline, cublas/cudnn libraries, the host-device memory model, cuda execution flow, and the profiling tools every sre should know.
mar 06, 20269 min · gpu, cuda
inside the sm: warps, partitions, and how gpus schedule work
a deep dive into gpu streaming multiprocessors: sm partitions, warp execution, the simt model, thread divergence, and nvlink interconnect - the hardware mechanics behind gpu compute.
mar 01, 202610 min · gpu, sm
gpu performance: bandwidth, throughput, and what the specs actually mean
how to read gpu specs without getting misled. memory bandwidth, tflops, data type precision, tensor cores, and compute capability - what each metric actually tells you about gpu performance.
feb 27, 202610 min · gpu, performance
a bus for math: an sre's first look at gpu architecture
an sre's exploration of gpu architecture, from cuda fundamentals to streaming multiprocessors. why gpus process data massively in parallel and when to reach for one.
feb 22, 20267 min · gpu, cuda
kustomize for kubernetes: efficiently managing multi-environment apps with kustomize - part 1
navigate multiple application environments in kubernetes with this practical tutorial on kustomize for efficient configuration management.
jun 25, 20232 min · kubernetes, kustomize
##currently
now
writing through gpu fundamentals for SREs and rebuilding this site as a quieter, file-backed workspace for the essays.mailing list
open this note in your inbox.
i send each new essay the morning it ships. one email a week. no spam.
##backlinks
linked from this note 3
cuda toolkit: compilers, libraries, and the host-device modelthe software stack that makes gpu programming possible - nvcc compiler pipeline, cublas/cudnn libraries, the host-device memory model, cuda execution flow, and the profiling tools every sre should know.
inside the sm: warps, partitions, and how gpus schedule worka deep dive into gpu streaming multiprocessors: sm partitions, warp execution, the simt model, thread divergence, and nvlink interconnect - the hardware mechanics behind gpu compute.
gpu performance: bandwidth, throughput, and what the specs actually meanhow to read gpu specs without getting misled. memory bandwidth, tflops, data type precision, tensor cores, and compute capability - what each metric actually tells you about gpu performance.