360SOFTY

Insights

Engineering Insights

Practical writing on software architecture, SaaS products, AI automation, legacy modernisation, and the business of building reliable systems.

RSS

Curated links from external sources — not 360Softy original articles.

ExternalAI
NVIDIA Technical Blog

Reducing High-Bandwidth Memory Bottlenecks in JAX-Based LLM Training with Host Offloading

Large language model (LLM) training workloads increasingly run into GPU memory limits before compute is fully used. Model weights, gradients, optimizer states,... Large language model (LLM) training workloads increasingly run into GPU memory limits before compute is fully used. Model weights, gradients, optimizer states, communication buffers, and intermediate activations all compete for GPU high-bandwidth memory (HBM). As model size, sequence length, and batch size grow, HBM capacity often beco

NVIDIA Technical BlogRead original
ExternalSoftware Engineering
DZone

Service Industry Evolution: Beyond 99.9% Uptime With Evolving Technology

For years, service organizations measured operational efficiency through response time. A machine failed, a ticket dropped, a technician arrived on-site, and the diagnosis and repair resolved the issue. Industries dependent on physical assets accepted this framework because they believed that it was not possible to avoid downtime. The benchmark for operational excellence depended on how quickly teams reacted after disruption occurred. That definition of service reliability has changed dramatical

Work with 360Softy

Building a SaaS product, AI system, or business platform?

Book a free consultation and we will tell you honestly whether we can help.