(All-cores)
(Single-core)
Specifications
Server Metadata
Vendor ID | gcp |
Server ID | 1741004 |
Name | ct5p-hightpu-2t |
Description | 104 vCPUs, 224 GB RAM, 2 Google TPU |
Family | ct5p |
Hw Virt | - |
Status | active |
Observed At | 2026-10-23T22:06:25.427257 |
Availability
| REGION / ID | SPOT | ONDEMAND |
|---|---|---|
Columbus (US) / us-east5 | 1.8065 USD/h | 8.4 USD/h |
Council Bluffs (US) / us-central1 | 1.8522 USD/h | 8.4 USD/h |
Moncks Corner (US) / us-east1 | 1.8522 USD/h | 8.4 USD/h |
Eemshaven (NL) / europe-west4 | 2.3583 USD/h | 8.505 USD/h |
St. Ghislain (BE) / europe-west1 | 3.3646 USD/h | 8.9189 USD/h |
Dallas (US) / us-south1 | 1.2613 USD/h | 9.558 USD/h |
Processor
vCPUs | 104 |
CPU Allocation | Dedicated |
CPU Architecture | x86_64 |
System Resources and Accelerators
| MEMORY | |
|---|---|
Memory Amount | 224 GiB |
| ACCELERATORS | |
|---|---|
Accelerator Count | 2 |
Accelerator Memory Min | 95 GiB |
Accelerator Memory Total | 190 GiB |
Accelerator Type | TPU |
Accelerator Manufacturer | Google |
Accelerator Family | TPU |
Accelerator Model | v5p |
Accelerators |
|
| STORAGE | |
|---|---|
Storage Size | 0 GB |
Storages |
| NETWORK | |
|---|---|
Inbound Traffic | 0 GB/month |
Outbound Traffic | 0 GB/month |
IPv4 | 0 |
Server Description
An accelerator-optimized server featuring dedicated host processors and dual specialized hardware units with high-capacity memory for deep learning workloads.
Google Cloud Platform ct5p-hightpu-2t is an x86_64 server instance configured with 104 dedicated vCPUs, 224.0 GB of system memory, and two Google TPU v5p accelerators providing 190 GB of total VRAM. The instance does not feature local storage or complimentary public IPv4 addresses, focusing its resources entirely on compute and acceleration. This hardware profile is highly specialized for machine learning workloads, including deep learning training, neural network inference, and complex data science computations. By offloading matrix operations to dedicated TPU hardware, the instance offers high qualitative cost efficiency for specialized AI workloads compared to general-purpose CPU configurations. The system provides approximately 2.15 GB of RAM per core, balancing host processor capabilities with massive accelerator memory.
Economics
Alternatives
Servers of the Same Family
| INSTANCE | vCPUs | MEMORY | Accelerators |
|---|---|---|---|
| ct5p-hightpu-1t | 52 | 112 GiB | 1 |
| ct5p-hightpu-4t | 208 | 448 GiB | 4 |
Similar Servers
| INSTANCE | VENDOR | vCPUs | MEMORY | Accelerators |
|---|