Dedicated GPU Resources

by Bluvalt Business application

Dedicated GPU Resources

by Bluvalt

Description
Plans
Addons
Screenshots
Support
FAQs
Contact Us

- Description -

STC Cloud now offers dedicated GPU instances, for handling high volume of tasks and projects, that require GPU processing power. STC guarantees service availability, uptime and provides around-the-clock support. This service has been designed as a monthly subscription that grants permanent resource availability and priority access, to the most exigent customers. Based on a high-performance, stable, reliable, and scalable infrastructure of DGX-A100 nodes, it offers the following features:

Dedicated NVIDIA A100 GPUs (based on innovative Ampere architecture), linked with NVSwitches to establish connections between GPUs.
40 GB HBM2 of memory per GPU.
16 cores of AMD EPYC processor per GPU.
High-speed NVMe Storage, with redundant SSD and up to 1 TB per GPU.
High-Throughput and low-latency communications with Mellanox 200Gbps.
GPU-to-GPU transfers with NVSwitch and NVLink.
Host-to-Device transfer speeds with PCI-Express generation 4.0
Multi-Instance GPU (MIG), to support physical separation of tasks, slicing one GPU into up to seven distinct hardware instances.

AVAILABILITY99.92% Uptime Guarantee

- Plans -

GPU-1G

Single GPU-type instance NVIDIA A100 Tensor Core GPU optimized for deep learning and high-performance computing workloads. Monthly Commitment (MC).

Items

40 gigabyte GPU Memory
16 Hour CPU Cores
124 gigabyte RAM
1 terabyte Storage

Features

GPUs based on Ampere Architecture
Local and SAN-based storage option
ML applications and tools available
Upload your own images

GPU-2G

Dual GPU-type instance NVIDIA A100 Tensor Core GPU optimized for deep learning and high performance computing workloads. Monthly Commitment (MC).

Items

80 gigabyte GPU Memory
32 Hour CPU Cores
248 gigabyte RAM
2 terabyte Storage

Features

GPUs based on Ampere Architecture
Local and SAN-based storage option
ML applications and tools available
Upload your own images

GPU-4G

Quadruple GPU-type instance NVIDIA A100 Tensor Core GPU optimized for deep learning and high performance computing workloads. Monthly Commitment (MC).

Items

160 gigabyte GPU Memory
64 Hour CPU Cores
496 gigabyte RAM
4 terabyte Storage

Features

GPUs based on Ampere Architecture
Local and SAN-based storage option
ML applications and tools available
Upload your own images

GPU-8G

Full instance of DGX A100 providing 8 Tensor Core GPU Optimized for more demanding deep learning and high performance computing workloads Monthly Commitment (MC).

Items

320 gigabyte GPU Memory
128 Hour CPU Cores
992 gigabyte RAM
8 terabyte Storage

Features

GPUs based on Ampere Architecture
Local and SAN-based storage option
ML applications and tools available
Upload your own images

GPU-1G-YC

Single GPU-type instance NVIDIA A100 Tensor Core GPU optimized for deep learning and high-performance computing workloads. Yearly Commitment (YC)

Items

40 gigabyte GPU Memory
16 Hour CPU Cores
124 gigabyte RAM
1 terabyte Storage

Features

GPUs based on Ampere Architecture
Local and SAN-based storage option
ML applications and tools available
Upload your own images

GPU-2G-YC

Dual GPU-type instance NVIDIA A100 Tensor Core GPU optimized for deep learning and high performance computing workloads. Yearly Commitment (YC).

Items

80 gigabyte GPU Memory
32 Hour CPU Cores
248 gigabyte RAM
4 terabyte Storage

Features

GPUs based on Ampere Architecture
Local and SAN-based storage option
ML applications and tools available
Upload your own images

GPU-4G-YC

Quadruple GPU-type instance NVIDIA A100 Tensor Core GPU optimized for deep learning and high performance computing workloads Yearly Commitment (YC).

Items

160 gigabyte GPU Memory
64 Hour CPU Cores
496 gigabyte RAM
4 terabyte Storage

Features

GPUs based on Ampere Architecture
Local and SAN-based storage option
ML applications and tools available
Upload your own images

GPU-8G-YC

Full instance of DGX A100 providing 8 Tensor Core GPU Optimized for more demanding deep learning and high performance computing workloads Yearly Commitment (YC).

Items

320 gigabyte GPU Memory
128 Hour CPU Cores
992 gigabyte RAM
8 terabyte Storage

Features

GPUs based on Ampere Architecture
Local and SAN-based storage option
ML applications and tools available
Upload your own images

- Add-ons -

Extend your subscription plan with the following add-ons

Answer your business needs and
be ready in seconds.

Create organization account

- How it works -

Register your organization with STC Cloud Cloud Marketplace
Add designated system administator

- Service Screenshots -

- Support Channels Included -

Email Phone Website

Coverage Details

Covered by Support

Provide appropriate virtualization infrastructure (GPU, CPU RAM, Storage and Connectivity), as specified by the GPUaaS offering.
Security & Availability of physical GPU cloud servers and physical network devices.
Security & Availability of GPU Cloud Infrastructure Management Layer.
Provide auto-provisioning of VTM virtual appliances, install & cancel licenses as per subscription, and automate the initial system setup, including high-availability (HA) for the virtual appliances.
Provide auto-provisioning of virtual IPs.
Ensure that customer data is not processed for any purpose independent of the instructions of the customer and follow STC privacy practices of collection, usage, storage and disposal.
Provide notification of scheduled maintenance at least 1 month in advance for any major changes that may affect the customers, and 1 week to 24 hours for emergency maintenance, as applicable.
Inform to the Client regarding unscheduled service outages due to troubleshooting, disruptions or as otherwise necessary.

Not Covered by Support

Perform all of the necessary security configuration and management tasks.
Handle the management tasks of the guest operating system (including updates and security patches), any application software or utilities installed by the customer on the instances, and the configuration of the STC-provided firewall on each instance.
Manage their data (including encryption options), classifying their assets, and using IAM tools to apply the appropriate permissions.
Determine external and internal security and related compliance requirements and objectives, and consider industry frameworks like the NIST Cybersecurity Framework (CSF) and ISO.
Communicate with STC support for any incidents, service requests and any service credit claims, from registered account in the GPU Cloud Support portal.
Ensure reasonable availability of Customer representative(s) to communicate with STC while resolving a service related request during business hours.
Provide all necessary information and assistance related to service performance that allows the STC team to meet the performance standards as outlined in this Agreement.

- FAQ's -

- Contact Sales -

Please don't hesitate to contact us if you would like to know more about this service.

Dedicated GPU Resources