n4d-highmem-80

Compute Engine · N4D family

The n4d-highmem-80 is a general purpose Compute Engine machine type with 80 vCPUs and 640 GiB of memory, running on AMD Turin with up to 50 gbps networking, starting at $3.383 per hour on demand.

general purposex86-64amd
vCPUs
80

dedicated

memory
640 GiB

8 GiB per vCPU

processor
AMD

2.7 GHz

on-demand
$4.447

$3,246/mo · us-central1

spot vm
$2.667

−40% vs on-demand

1-yr committed use
$2.802

−37% vs on-demand

n4d-highmem-80 price by region

17 regions · hourly list prices in USD · On-demand · Spot VM · 1-yr Committed use · 3-yr Committed use

Mumbaiasia-south1cheapestAsia$3.383$1.708$2.129$1.52
Columbusus-east5North America$4.447$2.368$2.802$2.002
Iowaus-central1referenceNorth America$4.447$2.667$2.802$2.002
Northern Virginiaus-east4North America$4.447$2.486$2.802$2.002
Oregonus-west1North America$4.447$1.236$2.802$2.002
South Carolinaus-east1North America$4.447$1.18$2.802$2.002
Delhiasia-south2Asia$4.624$1.364$2.913$2.082
Netherlandseurope-west4Europe$4.672$2.802$2.941$2.099
Tel Avivme-west1Middle East$4.891$1.044$3.082$2.202
Belgiumeurope-west1Europe$4.894$1.468$3.085$2.204
Finlandeurope-north1Europe$4.894$2.116$3.085$2.204
Las Vegasus-west4North America$5.01$1.352$3.155$2.256
Londoneurope-west2Europe$5.068$3.04$3.194$2.283
Frankfurteurope-west3Europe$5.25$2.792$3.306$2.363
Salt Lake Cityus-west3North America$5.347$3.206$3.367$2.405
Singaporeasia-southeast1Asia$5.484$3.07$3.456$2.471
Dammamme-central2Middle East$7.118$1.708$4.482$3.20

Specifications

compute

vCPUs
80
Memory
640 GiB
Memory per vCPU
8 GiB
Processor
AMD Turin
CPU vendor
AMD
Architecture
x86-64
Clock speed
2.7 GHz (turbo 3.5 GHz)
Burstable CPU
No

storage & network

Local storage
Persistent disk only
Network
Up to 50 Gbps
Peak bandwidth
50 Gbps

google cloud details

API name
n4d-highmem-80
Family
N4D
Category
General purpose
Generation
Current
Regions
17 (Asia, Europe, Middle East, North America)
Cheapest region
Mumbai · $3.383/hr
Max persistent disks
128

n4d-highmem-80 equivalents on other clouds

closest match by vCPU, memory, CPU architecture, accelerator and class · Linux on-demand in each provider's reference region

compare side by side→
ProviderInstancevCPUMemoryProcessorOn-demandvs n4d-highmem-80
Google Cloudn4d-highmem-8080640 GiBAMD Turin$4.447–
AWSi7ie.18xlarge72 vCPU (1.1× less) · 576 GiB (1.1× less)72576 GiBIntel Xeon Scalable (Emerald Rapids)$9.356+110%
r5a.24xlarge96 vCPU (1.2×) · 768 GiB (1.2×)96768 GiBAMD EPYC 7571$5.424+22%
AzureL80as v380640 GiBAMD EPYC 7763v (Milan)$6.24+40%
L80as v480640 GiBAMD EPYC 9004 (Genoa)$6.88+55%
AkamaiG8 Dedicated 512x128128 vCPU (1.6×) · 512 GiB (1.3× less)128512 GiBAMD EPYC$6.84+54%
Vultrvbm-32c-755gb-amd64 vCPU (1.3× less) · 768 GiB (1.2×)64768 GiBAMD EPYC 9354P$1.986-55%
vx1-m-96c-768gb96 vCPU (1.2×) · 768 GiB (1.2×)96768 GiBAMD EPYC$3.84-14%

N4D family sizes

all about N4D→
SizevCPUMemoryLocal storageOn-demandMonthlySpot1-yr
n4d-highcpu-224 GiB–$0.0715$52.17$0.0428$0.045
n4d-standard-228 GiB–$0.0847$61.83$0.0508$0.0534
n4d-highmem-2216 GiB–$0.1112$81.16$0.0667$0.07
n4d-highcpu-448 GiB–$0.1429$104$0.0857$0.09
n4d-standard-4416 GiB–$0.1694$124$0.1016$0.1067
n4d-highmem-4432 GiB–$0.2224$162$0.1334$0.1401
n4d-highcpu-8816 GiB–$0.2858$209$0.1714$0.1801
n4d-standard-8832 GiB–$0.3388$247$0.2032$0.2134
n4d-highmem-8864 GiB–$0.4447$325$0.2667$0.2802
n4d-highcpu-161632 GiB–$0.5717$417$0.3428$0.3601
n4d-standard-161664 GiB–$0.6776$495$0.4063$0.4269
n4d-highmem-1616128 GiB–$0.8894$649$0.5334$0.5603
n4d-highcpu-323264 GiB–$1.143$835$0.6855$0.7203
n4d-standard-3232128 GiB–$1.355$989$0.8126$0.8537
n4d-highmem-3232256 GiB–$1.779$1,299$1.067$1.121
n4d-highcpu-484896 GiB–$1.715$1,252$1.028$1.08
n4d-standard-4848192 GiB–$2.033$1,484$1.219$1.281
n4d-highmem-4848384 GiB–$2.668$1,948$1.60$1.681
n4d-highcpu-6464128 GiB–$2.287$1,669$1.371$1.441
n4d-standard-6464256 GiB–$2.71$1,979$1.625$1.707
n4d-highmem-6464512 GiB–$3.558$2,597$2.134$2.241
n4d-highcpu-8080160 GiB–$2.858$2,087$1.714$1.801
n4d-standard-8080320 GiB–$3.388$2,473$2.032$2.134
n4d-highmem-8080640 GiB–$4.447$3,246$2.667$2.802
n4d-highcpu-9696192 GiB–$3.43$2,504$2.057$2.161
n4d-standard-9696384 GiB–$4.066$2,968$2.438$2.561
n4d-highmem-9696768 GiB–$5.337$3,896$3.20$3.362

Similar machine types on Google Cloud

other Google Cloud families with a similar shape, cheapest first

n4d-highmem-80 FAQ

How much does n4d-highmem-80 cost?

n4d-highmem-80 costs $4.447 per hour on demand with Linux in Iowa (us-central1), about $3,246 per month. Across 17 regions the Linux on-demand price ranges from $3.383 in Mumbai to $7.118 in Dammam. Spot VM capacity costs $2.667 per hour (40% less), but can be reclaimed. A 1-yr Committed use commitment brings it to $2.802 per hour. With Windows it costs $8.127 per hour.

How many vCPUs and how much memory does n4d-highmem-80 have?

n4d-highmem-80 has 80 vCPUs and 640 GiB of memory (8 GiB per vCPU), running on AMD Turin. It is a general purpose machine type in the N4D family.

What is the cheapest region for n4d-highmem-80?

Mumbai (asia-south1) at $3.383 per hour for Linux on demand, 24% less than Iowa. n4d-highmem-80 is sold in 17 Google Cloud regions.

What are the equivalents of n4d-highmem-80 on other clouds?

The closest matches by vCPU, memory, CPU architecture and instance class are on AWS, i7ie.18xlarge (72 vCPU, 576 GiB, $9.356/hour in US East (N. Virginia)); on Azure, L80as v3 (80 vCPU, 640 GiB, $6.24/hour in East US); on Akamai, G8 Dedicated 512x128 (128 vCPU, 512 GiB, $6.84/hour in Newark, NJ); on Vultr, vbm-32c-755gb-amd (64 vCPU, 768 GiB, $1.986/hour in New Jersey).