Unsurpassed acceleration for solving the most complex computational tasks of AI, data analysis and HPC
Graphics servers with Tesla A100
All graphics servers with Tesla A100 are based on two Intel Xeon Gold 6240R CPUs with a base clock speed of 2.4 GHz and a maximum clock speed with Turbo Boost technology of 4.0 GHz.
Each processor contains two Intel® AVX-512 units and supports Intel® AVX-512 Deep Learning Boost functions. This set of instructions speeds up multiplication and addition operations with reduced accuracy, which are used in many internal cycles of the deep learning algorithm.
Each server has 768 GB of DDR4 ECC Reg 2933 MHz RAM. Local storage with a total capacity of 3200 GB is organized on Intel® solid-state drives, designed specifically for data centers.
GPU Tesla A100
Tesla A100 GPU provides unsurpassed acceleration for AI tasks, data analysis and for solving the most complex computing tasks. The A100 is the most productive integrated platform for AI and HPC, allowing you to get real-time results and deploy scalable solutions.
When training deep learning algorithms, tensor cores with Tensor Float (TF32) support increase performance by 20 times, without requiring changes in the code, and speed up the automatic function of working with different accuracy and FP16 by 2 times.
Double-precision tensor cores provide the greatest performance for high-performance computing since the advent of the GPU. HPC applications can also use TF32 to achieve up to 11 times the throughput for precise operations.
Data science specialists need to analyze, visualize large data sets and extract valuable information from them. To cope with workloads, servers equipped with A100 graphics accelerators provide the necessary computing power thanks to large amounts of high-speed memory with high bandwidth.
Video memory capacity
80 GB
Type of video memory
HBM2e
Memory bandwidth
1935 Gb/s
Encode/decode
1 encoder, 2 decoder (+AV1 decode)
GPU performance benchmarks
Performance benchmarks results in a virtual environment for 1 Tesla A100 graphics card.
OctaneBench 2020
up to
500
pts
Matrix multiply example
4300
GFlop/s
Hashcat bcrypt
117 000
H/s
Basic configurations with Tesla A100 80 GB
Prices:
Subscribe to the availability notification
Specify the number of required flavors . When they become available, you will receive a notification by email.
OK
Cancel
You successfully subscribed on notification.
You already subscribed on notification.
You already have reached the limit of TeslaA100 flavors.
Each physical core or GPU adapter assigned only to a single client. It means that:
Available vCPU time is 100%
Physical pass through of GPU inside a VM
Less storage and network load on hypervisors, more storage and network performance for a client.
Up to 75 000 IOPS1 for the RANDOM READ and up to 20 000 IOPS for the RANDOM WRITE for the Virtual Machines with local SSDs.
Up to 22 500 IOPS1 for the RANDOM READ and up to 20 000 IOPS for the RANDOM WRITE for the Virtual Machines with block storage Volumes.
You can be sure that Virtual Machines are not sharing vCPU or GPU among each other.
IOPS — Input/Output Operations Per Second.
Answers to frequently asked questions
What is the minimum rental period for a virtual GPU-server?
You can rent a virtual server for any period. Make a payment for any amount from 1.3 $ and work within the prepaid balance. When the work is completed, delete the server to stop spending money.
How quickly can I get started with a virtual GPU-server?
You create GPU-servers yourself in the control panel, choosing the hardware configuration and operating system. As a rule, the ordered capacities are available for use within a few minutes.
What operating systems can be installed on a virtual GPU-server?
You can choose from basic images: Windows Server 2019, Windows Server 2022, Ubuntu, Debian, CentOS, Fedora, OpenSUSE. Or use a pre-configured image from the Marketplace.
All operating systems are installed automatically when the GPU-server is created.
How to connect to a virtual GPU-server?
By default, we provide connection to Windows-based servers via RDP, and for Linux-based servers-via SSH.
You can configure any connection method that is convenient for you yourself.
Is it possible to rent a virtual GPU-server with an custom configuration?
Yes, it is possible. Contact our round-the-clock support service (https://t.me/immerscloudsupport) and tell us what configuration you need.
Why immers.cloud?
Cheapest GPU rates
Find cheaper — get a discount!
Discounts for prepayment
25% and 50% discount on prepayment for 1 and 3 months
Second billing
Use virtual machines just as much, as needed
No waiting
Automatic OS installation, virtual machines are ready in a few minutes
Free Internet
Up to 1 Gb/s incoming and outgoing traffic for free