a2-highgpu-8g

Compute Engine · A2 family

The a2-highgpu-8g is a gpu & accelerated Compute Engine machine type with 96 vCPUs and 680 GiB of memory, 8 × NVIDIA A100 40GB, running on Intel Cascade Lake with up to 100 gbps networking, starting at $29.39 per hour on demand.

gpu & acceleratedx86-64intelgpu
vCPUs
96

dedicated

memory
680 GiB

7.08 GiB per vCPU

gpu
8× A100 40GB

320 GiB total

on-demand
$29.39

$21,453/mo · us-central1

spot vm
$15.91

−46% vs on-demand

1-yr committed use
$18.51

−37% vs on-demand

a2-highgpu-8g price by region

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

Iowaus-central1cheapestreferenceNorth America$29.39$15.91$18.51$10.29
Oregonus-west1cheapestNorth America$29.39$15.91$18.51$10.29
South Carolinaus-east1cheapestNorth America$29.39$15.91$18.51$10.29
Netherlandseurope-west4Europe$29.98$13.92$18.89$10.49
Las Vegasus-west4North America$31.47$14.61$19.66$11.06
Singaporeasia-southeast1Asia$32.11$19.22$20.06$11.28
Tel Avivme-west1Middle East$32.33$19.40$20.37$11.31
Seoulasia-northeast3Asia$32.40$19.30$20.24$11.38
Tokyoasia-northeast1Asia$32.40$19.34$20.24$11.38
Salt Lake Cityus-west3North America$34.60$18.64$21.95$11.73

Specifications

compute

vCPUs
96
Memory
680 GiB
Memory per vCPU
7.08 GiB
Processor
Intel Cascade Lake
CPU vendor
Intel
Architecture
x86-64
Clock speed
2.2 GHz (turbo 2.9 GHz)
Burstable CPU
No

storage & network

Local storage
Persistent disk (local SSD attachable)
Network
Up to 100 Gbps
Peak bandwidth
100 Gbps

accelerators

Count
8
Memory (total)
320 GiB
Architecture
Ampere
Vendor
NVIDIA

google cloud details

API name
a2-highgpu-8g
Family
A2
Category
GPU & accelerated
Generation
Current
Regions
10 (Asia, Europe, Middle East, North America)
Cheapest region
Iowa · $29.39/hr
CoreMark score
1,269,327
Max persistent disks
128

a2-highgpu-8g 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 a2-highgpu-8g
Google Clouda2-highgpu-8g96680 GiB8 × NVIDIA A100 40GB$29.39–
AWSp4d.24xlarge1152 GiB (1.69×)961,152 GiB8 × NVIDIA A100 40GB$21.96-25%
p3dn.24xlarge768 GiB (1.13×) · NVIDIA V10096768 GiB8 × NVIDIA V100$31.21+6%
AzureND96asr A100 v4900 GiB (1.32×)96900 GiB8 × NVIDIA A100 40GB$27.20-7%

A2 family sizes

all about A2→
SizevCPUMemoryAcceleratorLocal storageOn-demandMonthlySpot1-yr
a2-highgpu-1g1285 GiB1 × NVIDIA A100 40GB–$3.673$2,682$1.989$2.314
a2-ultragpu-1g12170 GiB1 × NVIDIA A100 80GB375 GB NVME$5.069$3,700$2.977–
a2-highgpu-2g24170 GiB2 × NVIDIA A100 40GB–$7.347$5,363$3.978$4.628
a2-ultragpu-2g24340 GiB2 × NVIDIA A100 80GB750 GB NVME (2 disks)$10.14$7,400$5.954–
a2-highgpu-4g48340 GiB4 × NVIDIA A100 40GB–$14.69$10,726$7.957$9.257
a2-ultragpu-4g48680 GiB4 × NVIDIA A100 80GB1.5 TB NVME (4 disks)$20.28$14,801$11.91–
a2-highgpu-8g96680 GiB8 × NVIDIA A100 40GB–$29.39$21,453$15.91$18.51
a2-ultragpu-8g961,360 GiB8 × NVIDIA A100 80GB3 TB NVME (8 disks)$40.55$29,602$23.81–
a2-megagpu-16g961,360 GiB16 × NVIDIA A100 40GB–$55.74$40,690$30.18$35.12

a2-highgpu-8g FAQ

How much does a2-highgpu-8g cost?

a2-highgpu-8g costs $29.39 per hour on demand with Linux in Iowa (us-central1), about $21,453 per month. Across 10 regions the Linux on-demand price ranges from $29.39 in Iowa to $34.60 in Salt Lake City. Spot VM capacity costs $15.91 per hour (46% less), but can be reclaimed. A 1-yr Committed use commitment brings it to $18.51 per hour. With Windows it costs $33.80 per hour.

How many vCPUs and how much memory does a2-highgpu-8g have?

a2-highgpu-8g has 96 vCPUs and 680 GiB of memory (7.08 GiB per vCPU), running on Intel Cascade Lake. It is a gpu & accelerated machine type in the A2 family.

What GPU does a2-highgpu-8g have?

a2-highgpu-8g has 8 × NVIDIA A100 40GB with 320 GiB of accelerator memory in total.

What are the equivalents of a2-highgpu-8g on other clouds?

The closest matches by vCPU, memory, CPU architecture, accelerator and instance class are on AWS, p4d.24xlarge (96 vCPU, 1,152 GiB, $21.96/hour in US East (N. Virginia)); on Azure, ND96asr A100 v4 (96 vCPU, 900 GiB, $27.20/hour in East US).