Compute · N1 Instance Families

High-Performance
Compute Built for
Your Workload.

Three instance families. One bare-metal substrate. Pick the memory ratio that fits the workload, pay up to 37% less than the equivalent on AWS.

n1-highcpu37%
2 GB / vCPU
n1-standard26%
4 GB / vCPU
n1-highmem37%
8 GB / vCPU
Compute illustration
Why Nirvana Compute

Compute You'd Build.
If You Owned the Datacenter.

01
Instance Types

3 Instance Families

n1-highcpu, n1-standard, n1-highmem. Match the memory ratio to your workload, not the other way around.

02
Silicon

High-Clock AMD EPYC

Dedicated cores on AMD EPYC 9354P. No oversubscription. No noisy neighbors. No core sharing.

03
Virtualization

Lightweight Hypervisor

Custom hypervisor with minimal virtualization overhead. Tuned for latency-sensitive workloads, spin up in seconds, zero hardware to manage.

04
Compatibility

GCP / AWS Compatible

Familiar shape naming (n1-*). Drop-in for Compute Engine and EC2 workloads, keep your tooling.

05
Orchestration

VMs and NKS

Run instances directly or as nodes inside Nirvana Kubernetes Service. Same shapes, same pricing.

06
Pricing

Up to 37% Cheaper

Beat AWS equivalents by up to 37%. No commits required, no hidden egress.

Instance Families //

Pick the Shape
That Fits the Workload.

Every n1 shape runs on the same high-clock AMD EPYC substrate. The only thing that changes is the memory ratio, and the price.

Instance //
n1-highcpu
Compute-Bound.
vs AWS
37%
c6a
Ratio
2 GB / vCPU

Built for compute-bound work. CI/CD pipelines, video encoding, cryptographic verification, and blockchain validation — workloads where CPU cycles matter more than memory. Sizes range from n1-highcpu-2 (2 vCPU, 4 GB) to n1-highcpu-96 (96 vCPU, 192 GB).

vCPUs2 – 96
Memory4 – 192 GB
NetworkUp to 2 Gbps
Best For
  • CI/CD pipelines
  • Video encoding
  • Cryptographic verification
  • Blockchain validation
Most Popular //
n1-standard
General-Purpose.
vs AWS
26%
m6a
Ratio
4 GB / vCPU

The general-purpose family. Web servers, application backends, relational databases, and microservice fleets. This is where most teams start. Sizes range from n1-standard-2 (2 vCPU, 8 GB) to n1-standard-192 (192 vCPU, 768 GB).

vCPUs2 – 192
Memory8 – 768 GB
NetworkUp to 2 Gbps
Best For
  • Web servers & API gateways
  • Application backends
  • Relational databases
  • Microservice fleets
Instance //
n1-highmem
Memory-Optimized.
vs AWS
37%
r6a
Ratio
8 GB / vCPU

Memory-optimized for workloads that need large memory-to-CPU ratios. In-memory caches (Redis, Memcached), analytical databases (ClickHouse, DuckDB), blockchain archive nodes, and data warehousing. Sizes range from n1-highmem-2 (2 vCPU, 16 GB) to n1-highmem-96 (96 vCPU, 768 GB).

vCPUs2 – 96
Memory16 – 768 GB
NetworkUp to 2 Gbps
Best For
  • Redis / Memcached
  • ClickHouse / DuckDB
  • Blockchain archive nodes
  • Data warehousing
Shape Catalog //

Every shape in the catalog.

n1-standard-2
standard26%
vCPUs: 2Memory: 8 GBRatio: 4 GB/vCPUNetwork: 0.5 Gbps
n1-standard-4
standard26%
vCPUs: 4Memory: 16 GBRatio: 4 GB/vCPUNetwork: 0.5 Gbps
n1-standard-8
standard26%
vCPUs: 8Memory: 32 GBRatio: 4 GB/vCPUNetwork: 1 Gbps
n1-standard-16
standard26%
vCPUs: 16Memory: 64 GBRatio: 4 GB/vCPUNetwork: 1 Gbps
n1-standard-32
standard26%
vCPUs: 32Memory: 128 GBRatio: 4 GB/vCPUNetwork: 1 Gbps
n1-standard-48
standard26%
vCPUs: 48Memory: 192 GBRatio: 4 GB/vCPUNetwork: 2 Gbps
n1-standard-64
standard26%
vCPUs: 64Memory: 256 GBRatio: 4 GB/vCPUNetwork: 2 Gbps
n1-standard-96
standard26%
vCPUs: 96Memory: 384 GBRatio: 4 GB/vCPUNetwork: 2 Gbps
n1-standard-128
standard26%
vCPUs: 128Memory: 512 GBRatio: 4 GB/vCPUNetwork: 2 Gbps
n1-standard-192
standard26%
vCPUs: 192Memory: 768 GBRatio: 4 GB/vCPUNetwork: 2 Gbps
Benchmark //
Nirvana n1 instances match the equivalent AWS c6a, m6a and r6a shapes on identical workloads, at up to 37% lower cost.
Savings calculated against AWS on-demand list pricing for equivalent c6a / m6a / r6a shapes in us-east-1. No commits, no hidden egress.
Egress within a VPC and across Nirvana Connect is free. Public-internet egress is metered at $0.02/GB.
3.8 GHz
Clock Speed
AMD EPYC peak boost
−37%
Max Savings
vs AWS c6a / r6a
<60s
Provisioning
API call to ready
Private Egress
inter-VM in VPC + Connect

Powering AI, blockchain, and
real-time systems.

Talk to Sales