Product Introduction

Release Notes

Overview

Strengths

Scenarios

Instance Specifications

Billing Overview

Instance Regions

Purchasing Hyper Computing Cluster Instances

Billing

Getting Started

Managing Hyper Computing Cluster

Installing nvidia-fabricmanager Service on GPU Instance

Installation Instructions of TCCL on GPU Instances

Installing RDMA Millisecond-Level Monitoring Component on GPU Instances

User Guide

API Document

FAQs

RDMA Network Configuration Component RDMA-agent Description

Contact Us

Purchasing Methods

https://www.tencentcloud.com/document/product/1236/62816

Product Pricing

https://intl.cloud.tencent.com/pricing/cvm/overview

You can learn about the basic concepts and usages of Tencent Hyper Computing Cluster to help you quickly understand and efficiently use it.

Tencent Hyper Computing Cluster Overview

https://www.tencentcloud.com/document/product/1236/62807

Tencent Hyper Computing Cluster offers multiple billing modes. You can flexibly select the appropriate one based on business scenarios to optimize costs.

Billing Modes

https://www.tencentcloud.com/document/product/1236/62814

Tencent Hyper Computing Cluster provides multiple instance families, including GPU, Standard, and Compute. Each family is optimized for different dimensions such as CPU, GPU, and networking, you can select suitable computing resources based on your business scenarios.

https://www.tencentcloud.com/document/product/1236/62812

Learn about the availability of Tencent Hyper Computing Cluster instance specifications in different regions to help you plan resource deployment.

https://www.tencentcloud.com/document/product/1236/62815

This section introduces the related concepts, billing methods, and instance specifications of Tencent Hyper Computing Cluster, helping you understand the product.

Learning About Tencent Hyper Computing Cluster

Describes how to create, modify, scale, and delete THCC clusters, helping you manage cluster resources.

Managing Tencent Hyper Computing Cluster 

https://www.tencentcloud.com/document/product/1236/62818

Describes three installation methods for configuring the Tencent Collective Communication Library (TCCL) accelerated communication library on GPU instances. You can choose the method that best fits your business scenario.

Installing TCCL Acceleration Communication Library

https://www.tencentcloud.com/document/product/1236/62821

Enables millisecond-level monitoring in RDMA network environments, providing real-time network traffic monitoring and analysis to facilitate performance optimization.

Installing RDMA Millisecond-Level Monitoring Component

https://www.tencentcloud.com/document/product/1236/62822

Describes how to create and manage Tencent Hyper Computing Cluster, configure GPU instances, and install related components, helping you get started quickly.

Tencent Hyper Computing Cluster instances and CVM instances share the same APIs, You can learn about the APIs used for operating and managing core resources.

API Category

https://www.tencentcloud.com/document/product/213/15688

API Explorer is an automated online debugging tool designed for interactive interface operations. You can quickly test APIs and verify the availability and response results of APIs.

Using API Explorer

https://www.tencentcloud.com/document/product/1080/38761

Tencent Cloud Command Line Interface (TCCLI) is a unified tool for managing Tencent Cloud resources, which needs to be installed on local computers and is suitable for deep integration and automation scripts.

Using TCCLI

https://www.tencentcloud.com/document/product/1080/38762

CloudShell is a built-in online command line tool in the Tencent Cloud console, suitable for quickly running commands and performing temporary operations.

Using CloudShell

https://www.tencentcloud.com/document/product/1080/68246

Tencent Hyper Computing Cluster provides developers with tools such as APIs and SDKs, helping them efficiently accomplish integration, development, and Ops.

Developer Guide

Tencent Hyper Computing Cluster takes high-performance CVMs as nodes and interconnects with RDMA (Remote Direct Memory Access), providing high bandwidth and ultra-low latency network services, significantly improving network performance, and meeting the parallel computing requirements of large-scale high-performance computing, artificial intelligence, big data recommendation, and other applications. Tencent Hyper Computing Cluster combines elasticity and flexibility with extreme performance. It is ready to use immediately and can scale out on demand, effectively addressing the pain points of building on-premises supercomputing clusters, such as high investment, long cycles, and difficult Ops.

tencent cloud

Hyper Computing Cluster

Product Introduction

帮助和支持