Hardware
23 guides tagged “Hardware”.
- Egress vs Ingress Bandwidth: What's the Difference? Ingress is data coming in and is usually free; egress is data going out and is billed. Direction, not volume, is what shows up on the invoice.
- How Many vCPUs Does My App Actually Need? Size from the process model and measured CPU time. A single-process web app often needs one or two vCPUs; the database needs headroom and fast disk.
- NPU vs GPU: What's the Difference? An NPU is a fixed-function accelerator for low-power inference; a GPU is programmable and built for training and flexible serving.
- vCPU vs Core: What's the Difference? A core is physical; a vCPU is a scheduling unit the hypervisor exposes, often one hyperthread. The mapping depends on the instance family.
- What Is a TOPS Score (and What Does It Actually Measure)? TOPS is a peak throughput figure that depends on precision, sparsity, and memory. It is not a measure of real model performance.
- Hardware for AI Builders: How the Machine Your Code Runs On Actually Works A zero-jargon guide to computer hardware for people who build with AI: what a CPU, RAM, SSD, GPU, and data-centre server actually do, and how to size a machine for your app.
- Latency Numbers Every Developer Should Know The order-of-magnitude costs of memory, disk, network, and API calls — and why each step up the ladder is a different kind of slow.
- What Is a vCPU (and Why Isn't It a Real Core)? A vCPU is a slice of a physical core shared with other customers. Learn about oversubscription, steal time, and burstable instances before you size anything.
- ARM vs x86: Why Your Laptop, Your Server, and Your Cloud Disagree Apple Silicon, AWS Graviton, and Ampere made ARM servers real. Learn what changes when your binary no longer matches your host, and when ARM saves money.
- What Is a CPU (and Why More Cores Isn't Always Faster)? A CPU executes instructions one at a time per core. Learn what cores, clock speed, cache, and instruction sets actually do, and why a slower-looking chip often wins.
- Cores vs Threads: What Hyperthreading Actually Buys You A core executes; a thread is a stream of work. SMT lets two streams share one core, which helps some workloads and slows others. Here is what the spec sheet means.
- What Is RAM (and Why Does Your App Run Out of Memory)? RAM is the fast, temporary workspace every running program shares. Learn how capacity, bandwidth, and ECC differ, and why hitting the memory ceiling kills your app instantly.
- How a Program Actually Runs: From Your Python to the CPU What happens between pressing enter and the CPU doing work: bytecode, compilers, syscalls, the kernel, and why containers do not change any of it.
- What Is an NPU (and Do You Need an 'AI PC')? NPUs are small fixed-function AI accelerators now shipping in laptops and phones. Learn what they are good at, what TOPS means, and why they are not a substitute for a GPU.
- Bits, Bytes, and Bandwidth: The Units Behind Every Spec Sheet Why a 512 GB drive shows 476 GB, why 1 Gbps is not 1 GB/s, and how to read a hardware spec sheet without being fooled by the units.
- GPU vs CPU for AI: When Is a GPU Actually Worth It? Training needs GPUs, inference often does not, and an API call beats both for most small projects. A cost and capability comparison for AI builders.
- What Is Egress Bandwidth (and Why It Shows Up on the Bill)? Cloud providers meter traffic leaving their network far more strictly than traffic entering it. Egress is the line item that turns a free app into a monthly bill.
- SSD vs HDD vs NVMe: What Storage Speed Actually Means Hard drives, SATA SSDs, and NVMe drives differ by more than a factor of a hundred. Learn which number matters for your workload and where cloud storage quietly hurts.
- PC vs Data Center Server: Why a Server Looks Slower and Costs More A server is not a bigger PC. ECC memory, redundant power, out-of-band management, and 1U chassis explain why it is slower per core, louder, and far more expensive.
- What Is a GPU (and Why Is It Different From a CPU)? A GPU is thousands of simple cores built to do the same operation on huge blocks of numbers at once. Learn why that is exactly what AI needs and when it is the wrong tool.
- How Much RAM Does a Web App Actually Need? A practical way to size memory from the runtime, the workload, and the concurrency — instead of guessing and watching the process get killed.
- What Is VRAM (and Why Does a Model Not Fit)? GPU memory decides which AI models you can run. Learn the arithmetic for weights, cache, and context, and the practical ways to make a model fit.
- How to Choose a VPS Size for an AI-Built App Start small, measure the right three numbers, and resize on evidence. A practical sizing procedure for the machine that runs your AI-generated app.