Inside Graviton4: How AWS's New Chip Cuts EC2 Costs by 20% for Compute-Heavy Workloads
AWS has launched Graviton4, its fourth-generation custom chip, designed to reduce EC2 costs by 20% for compute-heavy workloads. The chip is built on TSMC's 3nm process and features ARM's Neoverse V2 architecture, delivering higher performance and efficiency. It is optimized for use cases like HPC, media encoding, and batch processing on ARM64-compatible Linux environments.
- ▪Graviton4 is fabricated on TSMC's 3nm process node, an advancement from the 5nm process used in Graviton3.
- ▪The chip features 72 Neoverse V2 cores, 2MB L2 cache per core, and supports DDR5-6400 memory with 12 channels.
- ▪It delivers 40% better performance per watt than Graviton3, contributing to lower operational costs.
- ▪AWS reports up to 30% higher throughput in video encoding and 28% faster job completion in batch processing with Graviton4.
- ▪Graviton4 does not currently support Windows workloads and is best suited for compute-bound applications on ARM64 Linux distributions.
Opening excerpt (first ~120 words) tap to expand
try { if(localStorage) { let currentUser = localStorage.getItem('current_user'); if (currentUser) { currentUser = JSON.parse(currentUser); if (currentUser.id === 3900225) { document.getElementById('article-show-container').classList.add('current-user-is-article-author'); } } } } catch (e) { console.error(e); } ANKUSH CHOUDHARY JOHAL Posted on May 1 • Originally published at johal.in Inside Graviton4: How AWS's New Chip Cuts EC2 Costs by 20% for Compute-Heavy Workloads #inside #graviton4 #awss #chip Inside Graviton4: How AWS's New Chip Cuts EC2 Costs by 20% for Compute-Heavy Workloads AWS recently launched Graviton4, its fourth-generation custom silicon chip built specifically for EC2 workloads.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at DEV.to (Top).
Discussion
0 commentsMore from DEV.to (Top)
