ingest: Public Cloud Learning Sessions - EC2 Cost Optimization AWS (2024-05-29)
Source: Cloud & DevOps/Public-Cloud-Learning-Sessions/05_FinOps/public-cloud-learning-sessions-best-practices-for-ec2-cost-optimization-in-aws-2.md Entities: Mike-Dukes, Steele-Taylor, Spot-Invaders Concepts: AWS-Nitro, EC2-Spot-Instances, ECS Concepts updated: Graviton, SpotInstances (added source reference)
This commit is contained in:
53
wiki/concepts/Graviton.md
Normal file
53
wiki/concepts/Graviton.md
Normal file
@@ -0,0 +1,53 @@
|
||||
---
|
||||
title: "Graviton"
|
||||
type: concept
|
||||
tags:
|
||||
- AWS
|
||||
- Cost-Optimization
|
||||
- ARM
|
||||
aliases:
|
||||
- Graviton
|
||||
- Graviton ARM
|
||||
- AWS Graviton
|
||||
last_updated: 2026-05-12
|
||||
---
|
||||
|
||||
## Overview
|
||||
|
||||
Graviton 是 AWS 基于 ARM 架构自研的处理器,相比 Intel/AMD x86 实例提供更高的性价比(最高 40%)和更低的功耗(减少高达 60%)。
|
||||
|
||||
## Benefits
|
||||
|
||||
- **成本更低**:相比同等配置 Intel 实例便宜 20-25%
|
||||
- **能效更高**:功耗显著降低
|
||||
- **性能提升**:对于支持 ARM 的工作负载性能更好
|
||||
|
||||
## Instance Types
|
||||
|
||||
- **M系列**:通用型(M6g/M7g)
|
||||
- **T系列**:突发性(T4g)
|
||||
- **C系列**:计算型(C6g/C7g)
|
||||
- **R系列**:内存优化(R6g/R7g)
|
||||
- **X系列**:内存优化(X2gd)
|
||||
|
||||
## Compatibility
|
||||
|
||||
适用于大多数工作负载:
|
||||
- Web 服务
|
||||
- 容器化应用(EKS/ECS)
|
||||
- 大数据处理
|
||||
- CI/CD 构建
|
||||
- 机器学习推理
|
||||
|
||||
排除场景:
|
||||
- 有状态服务(某些数据库)
|
||||
- 需要特定 x86 指令的应用
|
||||
- Windows 工作负载
|
||||
|
||||
## Related Pages
|
||||
|
||||
- [[FinOps]]
|
||||
- [[SpotInstances]]
|
||||
- [[public-cloud-learning-sessions-best-practices-for-ec2-cost-optimization-in-aws-2]]:Mike Dukes 和 Steele Taylor 详解 Graviton 性价比优势(40% 提升)和能耗优势(60% 降低)
|
||||
- [[ctp-topic-13-cloud-finops-micro-focus-policies-best-practices-to-optimize-the-co]]
|
||||
- [[ctp-topic-63-optimise-resource-cost-using-automation]]
|
||||
Reference in New Issue
Block a user