Amazon EC2 加速计算实例的规格 - Amazon EC2
实例系列和实例类型实例系列摘要性能规格网络规格亚马逊EBS规格实例存储规格安全规格
Amazon Web Services 文档中描述的 Amazon Web Services 服务或功能可能因区域而异。要查看适用于中国区域的差异，请参阅 中国的 Amazon Web Services 服务入门 (PDF)。

本文属于机器翻译版本。若本译文内容与英语原文存在差异，则一律以英文原文为准。

Amazon EC2 加速计算实例的规格

加速计算实例使用硬件加速器或协处理器来执行诸如浮点数计算、图形处理或数据模式匹配之类的功能，其效率要高于在其上运行的软件所能达到的效率。CPUs

有关此类别的上一代实例类型（例如 G3 实例）的信息，请参阅 Amazon EC2 上一代实例的规格。

实例系列和实例类型

实例系列 可用实例类型
DL1 dl1.24xlarge
DL2q dl2q.24xlarge
F1 f1.2xlarge | f1.4xlarge | f1.16xlarge
F2 f2.12xlarge | f2.48xlarge
G4ad g4ad.xlarge | g4ad.2xlarge | g4ad.4xlarge | g4ad.8xlarge | g4ad.16xlarge
G4dn g4dn.xlarge | g4dn.2xlarge | g4dn.4xlarge | g4dn.8xlarge | g4dn.12xlarge | g4dn.16xlarge | g4dn.metal
G5 g5.xlarge | g5.2xlarge | g5.4xlarge | g5.8xlarge | g5.12xlarge | g5.16xlarge | g5.24xlarge | g5.48xlarge
G5g g5g.xlarge | g5g.2xlarge | g5g.4xlarge | g5g.8xlarge | g5g.16xlarge | g5g.metal
G6 g6.xlarge | g6.2xlarge | g6.4xlarge | g6.8xlarge | g6.12xlarge | g6.16xlarge | g6.24xlarge | g6.48xlarge
G6e g6e.xlarge | g6e.2xlarge | g6e.4xlarge | g6e.8xlarge | g6e.12xlarge | g6e.16xlarge | g6e.24xlarge | g6e.48xlarge
Gr6 gr6.4xlarge | gr6.8xlarge
Inf1 inf1.xlarge | inf1.2xlarge | inf1.6xlarge | inf1.24xlarge
Inf2 inf2.xlarge | inf2.8xlarge | inf2.24xlarge | inf2.48xlarge
P2 p2.xlarge | p2.8xlarge | p2.16xlarge
P3 p3.2xlarge | p3.8xlarge | p3.16xlarge
P3dn p3dn.24xlarge
P4d p4d.24xlarge
P4de p4de.24xlarge
P5 p5.48xlarge
P5e p5e.48xlarge
p5en p5en.48xlarge
Trn1 trn1.2xlarge | trn1.32xlarge
Trn1n trn1n.32xlarge
trn2 trn2.48xlarge
trn2u trn2u.48xlarge
VT1 vt1.3xlarge | vt1.6xlarge | vt1.24xlarge

实例系列摘要

实例系列 虚拟机监控器 处理器类型（架构） 裸机实例可用性 专属主机支持 竞价型实例支持 休眠支持 支持的操作系统
DL1 Nitro v3 Intel (x86_64) ✗ ✓ ✓ ✗ Linux
DL2q Nitro v3 Intel (x86_64) ✗ ✓ ✓ ✗ Linux
F1 Xen Intel (x86_64) ✗ ✓ ✓ ✗ Linux
F2 Nitro v4 AMD (x86_64) ✗ ✓ ✓ ✗ Linux
G4ad Nitro v3 AMD (x86_64) ✗ ✓ ✓ ✗ Windows | Linux
G4dn Nitro v3 Intel (x86_64) ✓ ✓ ✓ ✗ Windows | Linux
G5 Nitro v3 AMD (x86_64) ✗ ✓ ✓ ✗ Windows | Linux
G5g Nitro v2 Amazon Graviton (arm64) ✓ ✓ ✓ ✗ Linux
G6 Nitro v4 AMD (x86_64) ✗ ✓ ✓ ✗ Windows | Linux
G6e Nitro v4 AMD (x86_64) ✗ ✓ ✓ ✗ Windows | Linux
Gr6 Nitro v4 AMD (x86_64) ✗ ✗ ✓ ✗ Windows | Linux
Inf1 Nitro v3 Intel (x86_64) ✗ ✓ ✓ ✗ Linux
Inf2 Nitro v4 AMD (x86_64) ✗ ✓ ✓ ✗ Linux
P2 Xen Intel (x86_64) ✗ ✓ ✓ ✗ Windows | Linux
P3 Xen Intel (x86_64) ✗ ✓ ✓ ✗ Windows | Linux
P3dn Nitro v3 Intel (x86_64) ✗ ✓ ✓ ✗ Windows | Linux
P4d Nitro v3 Intel (x86_64) ✗ ✓ ✓ ✗ Linux
P4de Nitro v3 Intel (x86_64) ✗ ✓ ✓ ✗ Linux
P5 Nitro v4 AMD (x86_64) ✗ ✗ ✓ ✗ Linux
P5e Nitro v4 AMD (x86_64) ✗ ✓ ✓ ✗ Linux
p5en Nitro v5 Intel (x86_64) ✗ ✗ ✓ ✗ Linux
Trn1 Nitro v4 Intel (x86_64) ✗ ✓ ✓ ✗ Linux
Trn1n Nitro v4 Intel (x86_64) ✗ ✗ ✓ ✗ Linux
trn2 Nitro v5 Intel (x86_64) ✗ ✗ ✓ ✗ Linux
trn2u Nitro v5 Intel (x86_64) ✗ ✗ ✗ ✗ Linux
VT1 Nitro v3 Intel (x86_64) ✗ ✓ ✓ ✗ Linux

性能规格

实例类型 可突增 内存（GiB） 处理器 vCPUs CPU 核心 每核心线程数 加速器 加速器内存
DL1
dl1.24xlarge ✗ 768.00 Intel Xeon P-8275CL 96 48 2 8 x Habana Gaudi HL-205 GPU 256 GiB (8 x 32 GiB)
DL2q
dl2q.24xlarge ✗ 768.00 Intel Xeon Cascade Lake 96 48 2 8 x Qualcomm Qualcomm AI100 inference accelerator 125 GiB (8 x 15 GiB)
F1
f1.2xlarge ✗ 122.00 Intel Xeon E5-2686v4 8 4 2 1 x Xilinx Virtex UltraScale (VU9P) FPGA 64 GiB (1 x 64 GiB)
f1.4xlarge ✗ 244.00 Intel Xeon E5-2686v4 16 8 2 2 x Xilinx Virtex UltraScale (VU9P) FPGA 128 GiB (2 x 64 GiB)
f1.16xlarge ✗ 976.00 Intel Xeon E5-2686v4 64 32 2 8 x Xilinx Virtex UltraScale (VU9P) FPGA 512 GiB (8 x 64 GiB)
F2
f2.12xlarge ✗ 512.00 AMD EPYC 7R13 48 24 2 2 x Xilinx Virtex UltraScale+ (VU47P) FPGA 160 GiB (2 x 80 GiB)
f2.48xlarge ✗ 2048.00 AMD EPYC 7R13 192 96 2 8 x Xilinx Virtex UltraScale+ (VU47P) FPGA 640 GiB (8 x 80 GiB)
G4ad
g4ad.xlarge ✗ 16.00 2nd Gen AMD EPYC 7R32 4 2 2 1 x AMD Radeon Pro V520 GPU 8 GiB (1 x 8 GiB)
g4ad.2xlarge ✗ 32.00 2nd Gen AMD EPYC 7R32 8 4 2 1 x AMD Radeon Pro V520 GPU 8 GiB (1 x 8 GiB)
g4ad.4xlarge ✗ 64.00 2nd Gen AMD EPYC 7R32 16 8 2 1 x AMD Radeon Pro V520 GPU 8 GiB (1 x 8 GiB)
g4ad.8xlarge ✗ 128.00 2nd Gen AMD EPYC 7R32 32 16 2 2 x AMD Radeon Pro V520 GPU 16 GiB (2 x 8 GiB)
g4ad.16xlarge ✗ 256.00 2nd Gen AMD EPYC 7R32 64 32 2 4 x AMD Radeon Pro V520 GPU 32 GiB (4 x 8 GiB)
G4dn
g4dn.xlarge ✗ 16.00 Intel Xeon P-8259L 4 2 2 1 x NVIDIA T4 GPU 16 GiB (1 x 16 GiB)
g4dn.2xlarge ✗ 32.00 Intel Xeon P-8259L 8 4 2 1 x NVIDIA T4 GPU 16 GiB (1 x 16 GiB)
g4dn.4xlarge ✗ 64.00 Intel Xeon P-8259L 16 8 2 1 x NVIDIA T4 GPU 16 GiB (1 x 16 GiB)
g4dn.8xlarge ✗ 128.00 Intel Xeon P-8259L 32 16 2 1 x NVIDIA T4 GPU 16 GiB (1 x 16 GiB)
g4dn.12xlarge ✗ 192.00 Intel Xeon P-8259L 48 24 2 4 x NVIDIA T4 GPU 64 GiB (4 x 16 GiB)
g4dn.16xlarge ✗ 256.00 Intel Xeon P-8259L 64 32 2 1 x NVIDIA T4 GPU 16 GiB (1 x 16 GiB)
g4dn.metal ✗ 384.00 Intel Xeon P-8259L 96 48 2 8 x NVIDIA T4 GPU 128 GiB (8 x 16 GiB)
G5
g5.xlarge ✗ 16.00 2nd Gen AMD EPYC 7R32 4 2 2 1 x NVIDIA A10G GPU 24 GiB (1 x 24 GiB)
g5.2xlarge ✗ 32.00 2nd Gen AMD EPYC 7R32 8 4 2 1 x NVIDIA A10G GPU 24 GiB (1 x 24 GiB)
g5.4xlarge ✗ 64.00 2nd Gen AMD EPYC 7R32 16 8 2 1 x NVIDIA A10G GPU 24 GiB (1 x 24 GiB)
g5.8xlarge ✗ 128.00 2nd Gen AMD EPYC 7R32 32 16 2 1 x NVIDIA A10G GPU 24 GiB (1 x 24 GiB)
g5.12xlarge ✗ 192.00 2nd Gen AMD EPYC 7R32 48 24 2 4 x NVIDIA A10G GPU 96 GiB (4 x 24 GiB)
g5.16xlarge ✗ 256.00 2nd Gen AMD EPYC 7R32 64 32 2 1 x NVIDIA A10G GPU 24 GiB (1 x 24 GiB)
g5.24xlarge ✗ 384.00 2nd Gen AMD EPYC 7R32 96 48 2 4 x NVIDIA A10G GPU 96 GiB (4 x 24 GiB)
g5.48xlarge ✗ 768.00 2nd Gen AMD EPYC 7R32 192 96 2 8 x NVIDIA A10G GPU 192 GiB (8 x 24 GiB)
G5g
g5g.xlarge ✗ 8.00 Amazon Graviton2 Processor 4 4 1 1 x NVIDIA T4g GPU 16 GiB (1 x 16 GiB)
g5g.2xlarge ✗ 16.00 Amazon Graviton2 Processor 8 8 1 1 x NVIDIA T4g GPU 16 GiB (1 x 16 GiB)
g5g.4xlarge ✗ 32.00 Amazon Graviton2 Processor 16 16 1 1 x NVIDIA T4g GPU 16 GiB (1 x 16 GiB)
g5g.8xlarge ✗ 64.00 Amazon Graviton2 Processor 32 32 1 1 x NVIDIA T4g GPU 16 GiB (1 x 16 GiB)
g5g.16xlarge ✗ 128.00 Amazon Graviton2 Processor 64 64 1 2 x NVIDIA T4g GPU 32 GiB (2 x 16 GiB)
g5g.metal ✗ 128.00 Amazon Graviton2 Processor 64 64 1 2 x NVIDIA T4g GPU 32 GiB (2 x 16 GiB)
G6
g6.xlarge ✗ 16.00 AMD EPYC 7R13 4 2 2 1 x NVIDIA L4 GPU 22 GiB (1 x 22 GiB)
g6.2xlarge ✗ 32.00 AMD EPYC 7R13 8 4 2 1 x NVIDIA L4 GPU 22 GiB (1 x 22 GiB)
g6.4xlarge ✗ 64.00 AMD EPYC 7R13 16 8 2 1 x NVIDIA L4 GPU 22 GiB (1 x 22 GiB)
g6.8xlarge ✗ 128.00 AMD EPYC 7R13 32 16 2 1 x NVIDIA L4 GPU 22 GiB (1 x 22 GiB)
g6.12xlarge ✗ 192.00 AMD EPYC 7R13 48 24 2 4 x NVIDIA L4 GPU 357 GiB (4 x 89 GiB)
g6.16xlarge ✗ 256.00 AMD EPYC 7R13 64 32 2 1 x NVIDIA L4 GPU 22 GiB (1 x 22 GiB)
g6.24xlarge ✗ 384.00 AMD EPYC 7R13 96 48 2 4 x NVIDIA L4 GPU 357 GiB (4 x 89 GiB)
g6.48xlarge ✗ 768.00 AMD EPYC 7R13 192 96 2 8 x NVIDIA L4 GPU 1430 GiB (8 x 178 GiB)
G6e
g6e.xlarge ✗ 32.00 AMD EPYC 7R13 4 2 2 1 x NVIDIA L40S GPU 44 GiB (1 x 44 GiB)
g6e.2xlarge ✗ 64.00 AMD EPYC 7R13 8 4 2 1 x NVIDIA L40S GPU 44 GiB (1 x 44 GiB)
g6e.4xlarge ✗ 128.00 AMD EPYC 7R13 16 8 2 1 x NVIDIA L40S GPU 44 GiB (1 x 44 GiB)
g6e.8xlarge ✗ 256.00 AMD EPYC 7R13 32 16 2 1 x NVIDIA L40S GPU 44 GiB (1 x 44 GiB)
g6e.12xlarge ✗ 384.00 AMD EPYC 7R13 48 24 2 4 x NVIDIA L40S GPU 715 GiB (4 x 178 GiB)
g6e.16xlarge ✗ 512.00 AMD EPYC 7R13 64 32 2 1 x NVIDIA L40S GPU 44 GiB (1 x 44 GiB)
g6e.24xlarge ✗ 768.00 AMD EPYC 7R13 96 48 2 4 x NVIDIA L40S GPU 715 GiB (4 x 178 GiB)
g6e.48xlarge ✗ 1536.00 AMD EPYC 7R13 192 96 2 8 x NVIDIA L40S GPU 2861 GiB (8 x 357 GiB)
Gr6
gr6.4xlarge ✗ 128.00 AMD EPYC 7R13 16 8 2 1 x NVIDIA L4 GPU 22 GiB (1 x 22 GiB)
gr6.8xlarge ✗ 256.00 AMD EPYC 7R13 32 16 2 1 x NVIDIA L4 GPU 22 GiB (1 x 22 GiB)
Inf1
inf1.xlarge ✗ 8.00 Intel Xeon P-8259L 4 2 2 1 x Amazon Inferentia inference accelerator 8 GiB (1 x 8 GiB)
inf1.2xlarge ✗ 16.00 Intel Xeon P-8259L 8 4 2 1 x Amazon Inferentia inference accelerator 8 GiB (1 x 8 GiB)
inf1.6xlarge ✗ 48.00 Intel Xeon P-8259L 24 12 2 4 x Amazon Inferentia inference accelerator 32 GiB (4 x 8 GiB)
inf1.24xlarge ✗ 192.00 Intel Xeon P-8259L 96 48 2 16 x Amazon Inferentia inference accelerator 128 GiB (16 x 8 GiB)
Inf2
inf2.xlarge ✗ 16.00 AMD EPYC 7R13 4 2 2 1 x Amazon Inferentia inference accelerator 32 GiB (1 x 32 GiB)
inf2.8xlarge ✗ 128.00 AMD EPYC 7R13 32 16 2 1 x Amazon Inferentia2 inference accelerator 32 GiB (1 x 32 GiB)
inf2.24xlarge ✗ 384.00 AMD EPYC 7R13 96 48 2 6 x Amazon Inferentia inference accelerator 192 GiB (6 x 32 GiB)
inf2.48xlarge ✗ 768.00 AMD EPYC 7R13 192 96 2 12 x Amazon Inferentia inference accelerator 384 GiB (12 x 32 GiB)
P2
p2.xlarge ✗ 61.00 Intel Xeon E5-2686v4 4 2 2 1 x NVIDIA K80 GPU 12 GiB (1 x 12 GiB)
p2.8xlarge ✗ 488.00 Intel Xeon E5-2686v4 32 16 2 8 x NVIDIA K80 GPU 96 GiB (8 x 12 GiB)
p2.16xlarge ✗ 732.00 Intel Xeon E5-2686 v4 64 32 2 16 x NVIDIA K80 GPU 192 GiB (16 x 12 GiB)
P3
p3.2xlarge ✗ 61.00 Intel Xeon E5-2686 v4 8 4 2 1 x NVIDIA V100 GPU 16 GiB (1 x 16 GiB)
p3.8xlarge ✗ 244.00 Intel Xeon E5-2686 v4 32 16 2 4 x NVIDIA V100 GPU 64 GiB (4 x 16 GiB)
p3.16xlarge ✗ 488.00 Intel Xeon E5-2686 v4 64 32 2 8 x NVIDIA V100 GPU 128 GiB (8 x 16 GiB)
P3dn
p3dn.24xlarge ✗ 768.00 Intel Xeon Platinum 8175 96 48 2 8 x NVIDIA V100 GPU 256 GiB (8 x 32 GiB)
P4d
p4d.24xlarge ✗ 1152.00 Intel Xeon Platinum 8175 96 48 2 8 x NVIDIA A100 GPU 320 GiB (8 x 40 GiB)
P4de
p4de.24xlarge ✗ 1152.00 Intel Xeon Platinum 8175 96 48 2 8 x NVIDIA A100 GPU 640 GiB (8 x 80 GiB)
P5
p5.48xlarge ✗ 2048.00 AMD EPYC 7R13 192 96 2 8 x NVIDIA H100 GPU 640 GiB (8 x 80 GiB)
P5e
p5e.48xlarge ✗ 2048.00 AMD EPYC 7R13 192 96 2 8 x NVIDIA H200 GPU 1128 GiB (8 x 141 GiB)
p5en
p5en.48xlarge ✗ 2048.00 Intel Xeon Sapphire Rapids 192 96 2 8 x NVIDIA NVIDIA GPU 1128 GiB (8 x 141 GiB)
Trn1
trn1.2xlarge ✗ 32.00 Intel Xeon Ice Lake 8375C 8 4 2 1 x Amazon Trainium accelerators 32 GiB (1 x 32 GiB)
trn1.32xlarge ✗ 512.00 Intel Xeon Ice Lake 8375C 128 64 2 16 x Amazon Trainium accelerators 512 GiB (16 x 32 GiB)
Trn1n
trn1n.32xlarge ✗ 512.00 Intel Xeon Ice Lake 128 64 2 16 x Amazon Trainium accelerators 512 GiB (16 x 32 GiB)
trn2
trn2.48xlarge ✗ 2048.00 Intel Xeon Sapphire Rapids 192 96 2 16 x Amazon Trainium2 accelerators 8192 GiB (16 x 512 GiB)
trn2u
trn2u.48xlarge ✗ 2048.00 Intel Xeon Sapphire Rapids 192 96 2 ✗ ✗
VT1
vt1.3xlarge ✗ 24.00 Intel Cascade Lake P-8259CL 12 6 2 1 x Xilinx U30 media accelerator 24 GiB (1 x 24 GiB)
vt1.6xlarge ✗ 48.00 Intel Cascade Lake P-8259CL 24 12 2 2 x Xilinx U30 media accelerator 48 GiB (2 x 24 GiB)
vt1.24xlarge ✗ 192.00 Intel Cascade Lake P-8259CL 96 48 2 8 x Xilinx U30 media accelerator 192 GiB (8 x 24 GiB)

网络规格

实例类型 基准/突增带宽（Gbps） EFA ENA ENA快递 网卡 网络接口数上限 每个接口的 IP 地址数 IPv6
DL1
dl1.24xlarge 4x100Gb ✓ ✓ ✗ 4 60 50 ✓
DL2q
dl2q.24xlarge 100Gb ✓ ✓ ✗ 1 15 50 ✓
F1
f1.2xlarge 1 Up to 10 Gigabit ✗ ✓ ✗ 1 4 15 ✓
f1.4xlarge 1 Up to 10 Gigabit ✗ ✓ ✗ 1 8 30 ✓
f1.16xlarge 25Gb ✗ ✓ ✗ 1 8 50 ✓
F2
f2.12xlarge 25Gb ✗ ✓ ✗ 1 8 30 ✓
f2.48xlarge 100Gb ✓ ✓ ✗ 1 15 50 ✓
G4ad
g4ad.xlarge 1 2.0 / 10.0 ✗ ✓ ✗ 1 2 4 ✓
g4ad.2xlarge 1 4.167 / 10.0 ✗ ✓ ✗ 1 2 4 ✓
g4ad.4xlarge 1 8.333 / 10.0 ✗ ✓ ✗ 1 3 10 ✓
g4ad.8xlarge 15Gb ✗ ✓ ✗ 1 4 15 ✓
g4ad.16xlarge 25Gb ✗ ✓ ✗ 1 8 30 ✓
G4dn
g4dn.xlarge 1 5.0 / 25.0 ✗ ✓ ✗ 1 3 10 ✓
g4dn.2xlarge 1 10.0 / 25.0 ✗ ✓ ✗ 1 3 10 ✓
g4dn.4xlarge 1 20.0 / 25.0 ✗ ✓ ✗ 1 3 10 ✓
g4dn.8xlarge 50Gb ✓ ✓ ✗ 1 4 15 ✓
g4dn.12xlarge 50Gb ✓ ✓ ✗ 1 8 30 ✓
g4dn.16xlarge 50Gb ✓ ✓ ✗ 1 4 15 ✓
g4dn.metal 100Gb ✓ ✓ ✗ 1 15 50 ✓
G5
g5.xlarge 1 2.5 / 10.0 ✗ ✓ ✗ 1 4 15 ✓
g5.2xlarge 1 5.0 / 10.0 ✗ ✓ ✗ 1 4 15 ✓
g5.4xlarge 1 10.0 / 25.0 ✗ ✓ ✗ 1 8 30 ✓
g5.8xlarge 25Gb ✓ ✓ ✗ 1 8 30 ✓
g5.12xlarge 40Gb ✓ ✓ ✗ 1 15 50 ✓
g5.16xlarge 25Gb ✓ ✓ ✗ 1 8 30 ✓
g5.24xlarge 50Gb ✓ ✓ ✗ 1 15 50 ✓
g5.48xlarge 100Gb ✓ ✓ ✗ 1 7 50 ✓
G5g
g5g.xlarge 1 1.25 / 10.0 ✗ ✓ ✗ 1 4 15 ✓
g5g.2xlarge 1 2.5 / 10.0 ✗ ✓ ✗ 1 4 15 ✓
g5g.4xlarge 1 5.0 / 10.0 ✗ ✓ ✗ 1 8 30 ✓
g5g.8xlarge 12Gb ✗ ✓ ✗ 1 8 30 ✓
g5g.16xlarge 25Gb ✗ ✓ ✗ 1 15 50 ✓
g5g.metal 25Gb ✗ ✓ ✗ 1 15 50 ✓
G6
g6.xlarge 1 2.5 / 10.0 ✗ ✓ ✗ 1 4 15 ✓
g6.2xlarge 1 5.0 / 10.0 ✗ ✓ ✗ 1 4 15 ✓
g6.4xlarge 1 10.0 / 25.0 ✗ ✓ ✗ 1 8 30 ✓
g6.8xlarge 25Gb ✓ ✓ ✗ 1 8 30 ✓
g6.12xlarge 40Gb ✓ ✓ ✗ 1 8 30 ✓
g6.16xlarge 25Gb ✓ ✓ ✗ 1 15 50 ✓
g6.24xlarge 50Gb ✓ ✓ ✗ 1 15 50 ✓
g6.48xlarge 100Gb ✓ ✓ ✓ 1 15 50 ✓
G6e
g6e.xlarge 1 2.5 / 20.0 ✗ ✓ ✗ 1 4 15 ✓
g6e.2xlarge 1 5.0 / 20.0 ✗ ✓ ✗ 1 4 15 ✓
g6e.4xlarge 20Gb ✗ ✓ ✗ 1 8 30 ✓
g6e.8xlarge 25Gb ✓ ✓ ✗ 1 8 30 ✓
g6e.12xlarge 100Gb ✓ ✓ ✓ 1 10 30 ✓
g6e.16xlarge 35 Gb ✓ ✓ ✗ 1 15 50 ✓
g6e.24xlarge 200Gb ✓ ✓ ✓ 2 20 50 ✓
g6e.48xlarge 400 Gb ✓ ✓ ✓ 4 40 50 ✓
Gr6
gr6.4xlarge 1 10.0 / 25.0 ✗ ✓ ✗ 1 8 30 ✓
gr6.8xlarge 25Gb ✓ ✓ ✗ 1 8 30 ✓
Inf1
inf1.xlarge 1 5.0 / 25.0 ✗ ✓ ✗ 1 4 10 ✓
inf1.2xlarge 1 5.0 / 25.0 ✗ ✓ ✗ 1 4 10 ✓
inf1.6xlarge 25Gb ✗ ✓ ✗ 1 8 30 ✓
inf1.24xlarge 100Gb ✓ ✓ ✗ 1 11 30 ✓
Inf2
inf2.xlarge 1 2.083 / 15.0 ✗ ✓ ✗ 1 4 15 ✓
inf2.8xlarge 1 16.667 / 25.0 ✗ ✓ ✗ 1 8 30 ✓
inf2.24xlarge 50Gb ✗ ✓ ✗ 1 15 50 ✓
inf2.48xlarge 100Gb ✗ ✓ ✗ 1 15 50 ✓
P2
p2.xlarge 高 ✗ ✓ ✗ 1 4 15 ✓
p2.8xlarge 10Gb ✗ ✓ ✗ 1 8 30 ✓
p2.16xlarge 25Gb ✗ ✓ ✗ 1 8 30 ✓
P3
p3.2xlarge 1 Up to 10 Gigabit ✗ ✓ ✗ 1 4 15 ✓
p3.8xlarge 10Gb ✗ ✓ ✗ 1 8 30 ✓
p3.16xlarge 25Gb ✗ ✓ ✗ 1 8 30 ✓
P3dn
p3dn.24xlarge 100Gb ✓ ✓ ✗ 1 15 50 ✓
P4d
p4d.24xlarge 4x100Gb ✓ ✓ ✗ 4 60 50 ✓
P4de
p4de.24xlarge 4x100Gb ✓ ✓ ✗ 4 60 50 ✓
P5
p5.48xlarge 3200 Gb ✓ ✓ ✗ 32 64 50 ✓
P5e
p5e.48xlarge 3200 Gb ✓ ✓ ✗ 32 64 50 ✓
p5en
p5en.48xlarge 3200 Gb ✓ ✓ ✗ 16 64 50 ✓
Trn1
trn1.2xlarge 1 3.125 / 12.5 ✗ ✓ ✗ 1 4 15 ✓
trn1.32xlarge 8x100Gb ✓ ✓ ✗ 8 40 50 ✓
Trn1n
trn1n.32xlarge 16x 100 Gb ✓ ✓ ✗ 16 80 50 ✓
trn2
trn2.48xlarge 16x 200 千兆位 ✓ ✓ ✗ 16 32 50 ✓
trn2u
trn2u.48xlarge 16x 200 千兆位 ✓ ✓ ✗ 16 32 50 ✓
VT1
vt1.3xlarge 3.12Gb ✗ ✓ ✗ 1 4 15 ✓
vt1.6xlarge 6.25Gb ✗ ✓ ✗ 1 8 30 ✓
vt1.24xlarge 25Gb ✓ ✓ ✗ 1 15 50 ✓
注意

1这些实例具备基准带宽，并且可以使用网络输入/输出积分机制，尽可能突破其基准带宽。其他实例类型可无限期地保持最高性能。有关更多信息，请参阅实例网络带宽。

亚马逊EBS规格

下表显示了哪些实例类型默认由 Amazon EBS 优化，以及哪些可选支持该类型。它还描述了它们EBS经过优化的性能，包括为 Amazon 提供的专用带宽EBS、使用流式读取工作负载和 128 KiB I/O 大小在该专用连接上可以实现的典型IOPS最大聚合吞吐量，以及使用 16 KiB I/O 大小时实例类型可以支持的最大值。未列出的实例类型不支持 Amazon EBS 优化。

重要

实例的EBS性能受实例的性能限制或其所连接卷的聚合性能的限制，以较小者为准。要实现最高EBS性能，实例必须连接的卷的组合性能等于或大于实例的最大性能。例如，要实现 80,000 IOPSr6i.16xlarge，实例必须至少配置16,000IOPS每个5gp3卷（5卷 x 16,000 IOPS = 80,000IOPS）。

我们建议您选择EBS经过优化的实例类型，该类型提供的专用 Amazon EBS 吞吐量超过您的应用程序需求；否则，Amazon EBS 和 Amazon 之间的连接EC2可能会成为性能瓶颈。

实例类型 基准/最大带宽（Mbps） 基准/最大吞吐量 (MB/s, 128 KiB I/O) 基准/最大值IOPS（16 KiB I/O） NVMe EBS优化 2
DL1
dl1.24xlarge 19000.00 2375.00 80000.00 ✓ default
DL2q
dl2q.24xlarge 19000.00 2375.00 80000.00 ✓ default
F1
f1.2xlarge 1700.00 212.50 12000.00 ✗ default
f1.4xlarge 3500.00 437.50 44000.00 ✗ default
f1.16xlarge 14000.00 1750.00 75000.00 ✗ default
F2
f2.12xlarge 15000.00 1875.00 60000.00 ✓ default
f2.48xlarge 60000.00 7500.00 240000.00 ✓ default
G4ad
g4ad.xlarge 1 400.00/3170.00 50.00/396.25 1700.00/13333.00 ✓ default
g4ad.2xlarge 1 800.00/3170.00 100.00/396.25 3400.00/13333.00 ✓ default
g4ad.4xlarge 1 1580.00/3170.00 197.50/396.25 6700.00/13333.00 ✓ default
g4ad.8xlarge 3170.00 396.25 13333.00 ✓ default
g4ad.16xlarge 6300.00 787.50 26667.00 ✓ default
G4dn
g4dn.xlarge 1 950.00/3500.00 118.75/437.50 3000.00/20000.00 ✓ default
g4dn.2xlarge 1 1150.00/3500.00 143.75/437.50 6000.00/20000.00 ✓ default
g4dn.4xlarge 4750.00 593.75 20000.00 ✓ default
g4dn.8xlarge 9500.00 1187.50 40000.00 ✓ default
g4dn.12xlarge 9500.00 1187.50 40000.00 ✓ default
g4dn.16xlarge 9500.00 1187.50 40000.00 ✓ default
g4dn.metal 19000.00 2375.00 80000.00 ✓ default
G5
g5.xlarge 1 700.00/3500.00 87.50/437.50 3000.00/15000.00 ✓ default
g5.2xlarge 1 850.00/3500.00 106.25/ 437.50 3500.00/15000.00 ✓ default
g5.4xlarge 4750.00 593.75 20000.00 ✓ default
g5.8xlarge 16000.00 2000.00 65000.00 ✓ default
g5.12xlarge 16000.00 2000.00 65000.00 ✓ default
g5.16xlarge 16000.00 2000.00 65000.00 ✓ default
g5.24xlarge 19000.00 2375.00 80000.00 ✓ default
g5.48xlarge 19000.00 2375.00 80000.00 ✓ default
G5g
g5g.xlarge 1 1188.00/4750.00 148.50/593.75 6000.00/20000.00 ✓ default
g5g.2xlarge 1 2375.00/4750.00 296.88/593.75 12000.00/20000.00 ✓ default
g5g.4xlarge 4750.00 593.75 20000.00 ✓ default
g5g.8xlarge 9500.00 1187.50 40000.00 ✓ default
g5g.16xlarge 19000.00 2375.00 80000.00 ✓ default
g5g.metal 19000.00 2375.00 80000.00 ✓ default
G6
g6.xlarge 1 1000.00/5000.00 125.00/625.00 4000.00/20000.00 ✓ default
g6.2xlarge 1 2000.00/5000.00 250.00/625.00 8000.00/20000.00 ✓ default
g6.4xlarge 8000.00 1000.00 32000.00 ✓ default
g6.8xlarge 16000.00 2000.00 64000.00 ✓ default
g6.12xlarge 20000.00 2500.00 80000.00 ✓ default
g6.16xlarge 20000.00 2500.00 80000.00 ✓ default
g6.24xlarge 30000.00 3750.00 120000.00 ✓ default
g6.48xlarge 60000.00 7500.00 240000.00 ✓ default
G6e
g6e.xlarge 1 1000.00/5000.00 125.00/625.00 4000.00/20000.00 ✓ default
g6e.2xlarge 1 2000.00/5000.00 250.00/625.00 8000.00/20000.00 ✓ default
g6e.4xlarge 8000.00 1000.00 32000.00 ✓ default
g6e.8xlarge 16000.00 2000.00 64000.00 ✓ default
g6e.12xlarge 20000.00 2500.00 80000.00 ✓ default
g6e.16xlarge 20000.00 2500.00 80000.00 ✓ default
g6e.24xlarge 30000.00 3750.00 120000.00 ✓ default
g6e.48xlarge 60000.00 7500.00 240000.00 ✓ default
Gr6
gr6.4xlarge 8000.00 1000.00 32000.00 ✓ default
gr6.8xlarge 16000.00 2000.00 64000.00 ✓ default
Inf1
inf1.xlarge 1 1190.00/4750.00 148.75/593.75 4000.00/20000.00 ✓ default
inf1.2xlarge 1 1190.00/4750.00 148.75/593.75 6000.00/20000.00 ✓ default
inf1.6xlarge 4750.00 593.75 20000.00 ✓ default
inf1.24xlarge 19000.00 2375.00 80000.00 ✓ default
Inf2
inf2.xlarge 1 1250.00/10000.00 156.25/1250.00 6000.00/40000.00 ✓ default
inf2.8xlarge 10000.00 1250.00 40000.00 ✓ default
inf2.24xlarge 30000.00 3750.00 120000.00 ✓ default
inf2.48xlarge 60000.00 7500.00 240000.00 ✓ default
P2
p2.xlarge 750.00 93.75 6000.00 ✗ default
p2.8xlarge 5000.00 625.00 32500.00 ✗ default
p2.16xlarge 10000.00 1250.00 65000.00 ✗ default
P3
p3.2xlarge 1750.00 218.75 10000.00 ✗ default
p3.8xlarge 7000.00 875.00 40000.00 ✗ default
p3.16xlarge 14000.00 1750.00 80000.00 ✗ default
P3dn
p3dn.24xlarge 19000.00 2375.00 80000.00 ✓ default
P4d
p4d.24xlarge 19000.00 2375.00 80000.00 ✓ default
P4de
p4de.24xlarge 19000.00 2375.00 80000.00 ✓ default
P5
p5.48xlarge 80000.00 10000.00 260000.00 ✓ default
P5e
p5e.48xlarge 80000.00 10000.00 260000.00 ✓ default
p5en
p5en.48xlarge 100000.00 12500.00 400000.00 ✓ default
Trn1
trn1.2xlarge 1 5000.00/20000.00 625.00/2500.00 16250.00/65000.00 ✓ default
trn1.32xlarge 80000.00 10000.00 260000.00 ✓ default
Trn1n
trn1n.32xlarge 80000.00 10000.00 260000.00 ✓ default
trn2
trn2.48xlarge 80000.00 10000.00 260000.00 ✓ default
trn2u
trn2u.48xlarge 80000.00 10000.00 260000.00 ✓ default
VT1
vt1.3xlarge 1 2375.00/4750.00 296.88/593.75 10000.00/20000.00 ✓ default
vt1.6xlarge 4750.00 593.75 20000.00 ✓ default
vt1.24xlarge 19000.00 2375.00 80000.00 ✓ default
注意

1 这些实例可以至少每 24 小时支持一次 30 分钟的最大性能，之后会恢复到基线性能。其他实例保持最高性能的时间不确定。如果工作负载需要在超过 30 分钟的时长内保持最大性能，请选择以下任一实例。

2 default 表示默认情况下启用实例进行EBS优化。 supported表示可以选择启用实例以进行EBS优化。有关更多信息，请参阅 Amazon EBS 优化实例。

实例存储规格

下表显示了支持实例类型的实例存储卷配置，以及队列深度饱和时块大小为 4,096 字节的聚合IOPS性能。

实例类型 实例存储卷 实例存储类型 100% 随机读取 IOPS /写入 IOPS 需要初始化1 TRIM支持 2
DL1
dl1.24xlarge 4 x 1000 GB NVMe SSD 1,000,000/800,000 ✓
F1
f1.2xlarge 1 x 470 GB NVMe SSD ✓
f1.4xlarge 1 x 940 GB NVMe SSD ✓
f1.16xlarge 4 x 940 GB NVMe SSD ✓
F2
f2.12xlarge 2 x 940 GB NVMe SSD 800,000/250,000 ✓
f2.48xlarge 8 x 940 GB NVMe SSD 3200,000 /1,000,000 ✓
G4ad
g4ad.xlarge 1 x 150 GB NVMe SSD 10,417/8,333 ✓
g4ad.2xlarge 1 x 300 GB NVMe SSD 20,833/16,667 ✓
g4ad.4xlarge 1 x 600 GB NVMe SSD 41,667/33,333 ✓
g4ad.8xlarge 1 x 1200 GB NVMe SSD 83,333/66,667 ✓
g4ad.16xlarge 2 x 1200 GB NVMe SSD 166,666/133,332 ✓
G4dn
g4dn.xlarge 1 x 125 GB NVMe SSD 42,500/32,500 ✓
g4dn.2xlarge 1 x 225 GB NVMe SSD 42,500/32,500 ✓
g4dn.4xlarge 1 x 225 GB NVMe SSD 85,000/65,000 ✓
g4dn.8xlarge 1 x 900 GB NVMe SSD 250,000/200,000 ✓
g4dn.12xlarge 1 x 900 GB NVMe SSD 250,000/200,000 ✓
g4dn.16xlarge 1 x 900 GB NVMe SSD 250,000/200,000 ✓
g4dn.metal 2 x 900 GB NVMe SSD 500,000/400,000 ✓
G5
g5.xlarge 1 x 250 GB NVMe SSD 40,625/20,313 ✓
g5.2xlarge 1 x 450 GB NVMe SSD 40,625/20,313 ✓
g5.4xlarge 1 x 600 GB NVMe SSD 125,000/62,500 ✓
g5.8xlarge 1 x 900 GB NVMe SSD 250,000/125,000 ✓
g5.12xlarge 1 x 3800 GB NVMe SSD 312,500/156,250 ✓
g5.16xlarge 1 x 1900 GB NVMe SSD 250,000/125,000 ✓
g5.24xlarge 1 x 3800 GB NVMe SSD 312,500/156,250 ✓
g5.48xlarge 2 x 3800 GB NVMe SSD 625,000/312,500 ✓
G6
g6.xlarge 1 x 250 GB NVMe SSD 40,625/20,000 ✓
g6.2xlarge 1 x 450 GB NVMe SSD 40,625/20,000 ✓
g6.4xlarge 1 x 600 GB NVMe SSD 125,000/40,000 ✓
g6.8xlarge 2 x 450 GB NVMe SSD 250,000/80,000 ✓
g6.12xlarge 4 x 940 GB NVMe SSD 312,500/125,000 ✓
g6.16xlarge 2 x 940 GB NVMe SSD 250,000/80,000 ✓
g6.24xlarge 4 x 940 GB NVMe SSD 312,500/156,248 ✓
g6.48xlarge 8 x 940 GB NVMe SSD 625,000/312,496 ✓
G6e
g6e.xlarge 1 x 250 GB NVMe SSD 40,625/20,000 ✓
g6e.2xlarge 1 x 450 GB NVMe SSD 40,625/20,000 ✓
g6e.4xlarge 1 x 600 GB NVMe SSD 125,000/40,000 ✓
g6e.8xlarge 2 x 450 GB NVMe SSD 250,000/80,000 ✓
g6e.12xlarge 2 x 1900 GB NVMe SSD 312,500/125,000 ✓
g6e.16xlarge 2 x 950 GB NVMe SSD 250,000/80,000 ✓
g6e.24xlarge 2 x 1900 GB NVMe SSD 312,500/156,250 ✓
g6e.48xlarge 4 x 1900 GB NVMe SSD 625,000/312,500 ✓
Gr6
gr6.4xlarge 1 x 600 GB NVMe SSD 125,000/40,000 ✓
gr6.8xlarge 2 x 450 GB NVMe SSD 250,000/80,000 ✓
P3dn
p3dn.24xlarge 2 x 900 GB NVMe SSD 700,000/340,000 ✓
P4d
p4d.24xlarge 8 x 1000 GB NVMe SSD 2,000,000/1,600,000 ✓
P4de
p4de.24xlarge 8 x 1000 GB NVMe SSD 2,000,000/1,600,000 ✓
P5
p5.48xlarge 8 x 3800 GB NVMe SSD 4,400,000/2,200,000 ✓
P5e
p5e.48xlarge 8 x 3800 GB NVMe SSD 4,400,000/2,200,000 ✓
p5en
p5en.48xlarge 8 x 3800 GB NVMe SSD 4,400,000/2,200,000 ✓
Trn1
trn1.2xlarge 1 x 474 GB NVMe SSD 107,500/45,000 ✓
trn1.32xlarge 4 x 1900 GB NVMe SSD 1,720,000/720,000 ✓
Trn1n
trn1n.32xlarge 4 x 1900 GB NVMe SSD 1,720,000/720,000 ✓
trn2u
trn2u.48xlarge 4 x 1900 GB NVMe SSD 1,720,000/720,000 ✓

1如果附加到特定实例的卷未初始化，则会遭受首次写入惩罚。有关更多信息，请参阅优化实例存储卷的磁盘性能。

2 有关更多信息，请参阅实例存储卷TRIM支持。

安全规格

实例类型 EBS 加密 实例存储加密 传输中加密 AMD SEV-SNP 硝基 TPM Nitro Enclaves
DL1
dl1.24xlarge ✓ ✓ ✓ ✗ ✗ ✓
DL2q
dl2q.24xlarge ✓ Instance store not supported ✓ ✗ ✗ ✓
F1
f1.2xlarge ✓ ✓ ✗ ✗ ✗ ✗
f1.4xlarge ✓ ✓ ✗ ✗ ✗ ✗
f1.16xlarge ✓ ✓ ✗ ✗ ✗ ✗
F2
f2.12xlarge ✓ ✓ ✓ ✗ ✓ ✓
f2.48xlarge ✓ ✓ ✓ ✗ ✓ ✓
G4ad
g4ad.xlarge ✓ ✓ ✓ ✗ ✗ ✗
g4ad.2xlarge ✓ ✓ ✓ ✗ ✗ ✗
g4ad.4xlarge ✓ ✓ ✓ ✗ ✗ ✗
g4ad.8xlarge ✓ ✓ ✓ ✗ ✗ ✗
g4ad.16xlarge ✓ ✓ ✓ ✗ ✗ ✗
G4dn
g4dn.xlarge ✓ ✓ ✓ ✗ ✓ ✓
g4dn.2xlarge ✓ ✓ ✓ ✗ ✓ ✓
g4dn.4xlarge ✓ ✓ ✓ ✗ ✓ ✓
g4dn.8xlarge ✓ ✓ ✓ ✗ ✓ ✓
g4dn.12xlarge ✓ ✓ ✓ ✗ ✓ ✓
g4dn.16xlarge ✓ ✓ ✓ ✗ ✓ ✓
g4dn.metal ✓ ✓ ✓ ✗ ✗ ✗
G5
g5.xlarge ✓ ✓ ✓ ✗ ✓ ✓
g5.2xlarge ✓ ✓ ✓ ✗ ✓ ✓
g5.4xlarge ✓ ✓ ✓ ✗ ✓ ✓
g5.8xlarge ✓ ✓ ✓ ✗ ✓ ✓
g5.12xlarge ✓ ✓ ✓ ✗ ✓ ✓
g5.16xlarge ✓ ✓ ✓ ✗ ✓ ✓
g5.24xlarge ✓ ✓ ✓ ✗ ✓ ✓
g5.48xlarge ✓ ✓ ✓ ✗ ✓ ✓
G5g
g5g.xlarge ✓ Instance store not supported ✗ ✗ ✗ ✗
g5g.2xlarge ✓ Instance store not supported ✗ ✗ ✗ ✗
g5g.4xlarge ✓ Instance store not supported ✗ ✗ ✗ ✗
g5g.8xlarge ✓ Instance store not supported ✗ ✗ ✗ ✗
g5g.16xlarge ✓ Instance store not supported ✗ ✗ ✗ ✗
g5g.metal ✓ Instance store not supported ✗ ✗ ✗ ✗
G6
g6.xlarge ✓ ✓ ✓ ✗ ✓ ✓
g6.2xlarge ✓ ✓ ✓ ✗ ✓ ✓
g6.4xlarge ✓ ✓ ✓ ✗ ✓ ✓
g6.8xlarge ✓ ✓ ✓ ✗ ✓ ✓
g6.12xlarge ✓ ✓ ✓ ✗ ✓ ✓
g6.16xlarge ✓ ✓ ✓ ✗ ✓ ✓
g6.24xlarge ✓ ✓ ✓ ✗ ✓ ✓
g6.48xlarge ✓ ✓ ✓ ✗ ✓ ✓
G6e
g6e.xlarge ✓ ✓ ✓ ✗ ✓ ✓
g6e.2xlarge ✓ ✓ ✓ ✗ ✓ ✓
g6e.4xlarge ✓ ✓ ✓ ✗ ✓ ✓
g6e.8xlarge ✓ ✓ ✓ ✗ ✓ ✓
g6e.12xlarge ✓ ✓ ✓ ✗ ✓ ✓
g6e.16xlarge ✓ ✓ ✓ ✗ ✓ ✓
g6e.24xlarge ✓ ✓ ✓ ✗ ✓ ✓
g6e.48xlarge ✓ ✓ ✓ ✗ ✓ ✓
Gr6
gr6.4xlarge ✓ ✓ ✓ ✗ ✓ ✓
gr6.8xlarge ✓ ✓ ✓ ✗ ✓ ✓
Inf1
inf1.xlarge ✓ Instance store not supported ✓ ✗ ✓ ✓
inf1.2xlarge ✓ Instance store not supported ✓ ✗ ✓ ✓
inf1.6xlarge ✓ Instance store not supported ✓ ✗ ✓ ✓
inf1.24xlarge ✓ Instance store not supported ✓ ✗ ✓ ✓
Inf2
inf2.xlarge ✓ Instance store not supported ✓ ✗ ✓ ✓
inf2.8xlarge ✓ Instance store not supported ✓ ✗ ✓ ✓
inf2.24xlarge ✓ Instance store not supported ✓ ✗ ✓ ✓
inf2.48xlarge ✓ Instance store not supported ✓ ✗ ✓ ✓
P2
p2.xlarge ✓ Instance store not supported ✗ ✗ ✗ ✗
p2.8xlarge ✓ Instance store not supported ✗ ✗ ✗ ✗
p2.16xlarge ✓ Instance store not supported ✗ ✗ ✗ ✗
P3
p3.2xlarge ✓ Instance store not supported ✗ ✗ ✗ ✗
p3.8xlarge ✓ Instance store not supported ✗ ✗ ✗ ✗
p3.16xlarge ✓ Instance store not supported ✗ ✗ ✗ ✗
P3dn
p3dn.24xlarge ✓ ✓ ✓ ✗ ✗ ✓
P4d
p4d.24xlarge ✓ ✓ ✓ ✗ ✗ ✓
P4de
p4de.24xlarge ✓ ✓ ✓ ✗ ✗ ✓
P5
p5.48xlarge ✓ ✓ ✓ ✗ ✗ ✓
P5e
p5e.48xlarge ✓ ✓ ✓ ✗ ✓ ✓
p5en
p5en.48xlarge ✓ ✓ ✓ ✗ ✗ ✓
Trn1
trn1.2xlarge ✓ ✓ ✓ ✗ ✗ ✗
trn1.32xlarge ✓ ✓ ✓ ✗ ✗ ✗
Trn1n
trn1n.32xlarge ✓ ✓ ✓ ✗ ✗ ✗
trn2
trn2.48xlarge ✓ Instance store not supported ✓ ✗ ✗ ✓
trn2u
trn2u.48xlarge ✓ ✓ ✓ ✗ ✗ ✓
VT1
vt1.3xlarge ✓ Instance store not supported ✓ ✗ ✗ ✗
vt1.6xlarge ✓ Instance store not supported ✓ ✗ ✗ ✗
vt1.24xlarge ✓ Instance store not supported ✓ ✗ ✗ ✗