Amey Lokare

Dec 17, 2024 • 5 min read • Architecture & Design

Tech Doesn't Fail—Bad Architecture Does

When systems fail, it's easy to blame the technology. But in my experience, technology rarely fails on its own. The real culprit is almost always bad architecture. Here's how to think about system design to prevent failures before they happen.

Architecture System Design Engineering Principles Best Practices Failure Analysis System Reliability ameylokare amey lokare lokare amey

Dec 17, 2024 • 5 min read • Tech Trends & Industry

What Broke in Tech This Month (And Why It Matters)

From AI model outages to cloud pricing changes and open-source drama, December 2024 had its share of tech failures. Let's break down what broke, why it matters, and what we can learn from these incidents.

Tech News AI Outages Cloud Computing Open Source Industry Analysis December 2024 Tech Failures ameylokare amey lokare lokare amey

Dec 16, 2024 • 5 min read • Tech Trends & Industry

The Silent Shift: Engineers Are Moving Back to Self-Hosted Tools

A quiet revolution is happening in engineering teams. After years of cloud-first thinking, many engineers are rediscovering the value of self-hosted solutions for Git, CI/CD, monitoring, and even AI workloads. Here's why this shift is happening and what it means.

Self-Hosting DevOps Git CI/CD Monitoring AI Infrastructure Cloud Migration Engineering Trends ameylokare amey lokare lokare amey

Dec 15, 2024 • 4 min read • Infrastructure & Hardware

ECC RAM: When It Matters and When It's a Waste of Money

Error-Correcting Code (ECC) RAM is often touted as essential for production servers, but the reality is more nuanced. After managing hundreds of servers, I've learned when ECC RAM actually matters and when you're just burning money on unnecessary hardware.

ECC RAM Memory Hardware Server Infrastructure Data Integrity Error Correction Hardware Optimization ameylokare amey lokare lokare amey

Dec 14, 2024 • 4 min read • Infrastructure & Hardware

ZFS Is Amazing—Until It Isn't: Hard Truths About Storage Pools

ZFS gets a lot of hype in the storage world, and for good reason. But after years of running ZFS in production, I've learned that it's not a silver bullet. Here are the hard truths about when ZFS fails, why it fails, and what you should know before committing to it.

ZFS Storage File Systems Data Integrity RAID Backup Disaster Recovery Linux Infrastructure ameylokare amey lokare lokare amey

Dec 13, 2024 • 7 min read • AI & Machine Learning

Why 24GB VRAM Is the New Minimum for Serious AI Work

Modern AI models are breaking on 12GB cards. After running local LLMs, training models, and deploying AI systems, I've learned that 24GB VRAM is now the practical minimum for serious AI work. Here's why, and what it means for your hardware choices—comparing RTX 3090, 4090, and A6000.

A6000 AI Hardware AI Training amey amey lokare ameylokare CUDA Deep Learning GPU GPU Computing Hardware LLM lokare Machine Learning Model Inference NVIDIA RTX 3090 RTX 4090 VRAM

Posts Tagged: ameylokare

Tech Doesn't Fail—Bad Architecture Does

What Broke in Tech This Month (And Why It Matters)

The Silent Shift: Engineers Are Moving Back to Self-Hosted Tools

ECC RAM: When It Matters and When It's a Waste of Money

ZFS Is Amazing—Until It Isn't: Hard Truths About Storage Pools

Why 24GB VRAM Is the New Minimum for Serious AI Work

Dive Deeper into ameylokare