Single HBM stack reads Wikipedia 20x per second, HBM5 to double bandwidth by 2028
lemire · x · 2026-09-29
- AI processors use High Bandwidth Memory; the current HBM4 family exceeds 2 TB/s per stack (up to 64 GB, often 32 GB per stack), with accelerators stacking multiple modules for even higher total bandwidth.
- Perspective: all uncompressed Wikipedia text is 100 GB, so a single HBM stack can read it over 20 times per second.
- HBM5 is expected by 2028 at roughly twice the speed (>4 TB/s per stack), reading Wikipedia 40+ times per second. Since bandwidth is often the bottleneck, 2028 AI chips should be much faster on this basis alone.
More from Infra
- The endless model-hopper cycle: Claude's success overloads servers, users drift back to GPT — haider1 · 2026-09-29
- AI detector startup Pangram saved 2-3 engineer-years and cut compute costs 50-60% — LightningAI · 2026-09-29
- Batam Data Center Strains Residents' Water Supply While Its Desalination Plant Remains on Paper — AryHHAry · 2026-09-29
- Nvidia and AMD lobby Trump to keep China chip sales flowing, opposing AI OVERWATCH Act — pstAsiatech · 2026-09-29
- Chinese Phone Makers Raise Flagship Prices by Up to $446 as Memory Costs Surge — pstAsiatech · 2026-09-29
- Running 1000s of enterprise RAG indexes: a year of lessons in incremental sync — srnsnemil · 2026-09-29