ECS Types
Latest-generation Intel® Xeon® Scalable processors enable powerful and reliable
computing, and Huawei 25GE high-speed intelligent NICs provide ultra-high network
bandwidth and PPS.
Websites and web applications with high requirements on computing and network performance
Generalized databases and cache servers
Medium- and heavy-load enterprise applications
- vCPU/Memory Ratio 1:2 | 1:4
- vCPUs 2-64
- Base/Turbo Frequency 3.0/3.4 GHz
- Maximum PPS 10,000,000
- Maximum Intranet Bandwidth 40 Gbit/s
Websites and web applications with high requirements on computing and network performance; generalized databases and cache servers; medium- and heavy-load enterprise applications
Latest-generation Intel® Xeon® Scalable processors and Huawei 25GE high-speed intelligent NICs provide powerful and stable computing and network performance.
Websites and web applications with high requirements on computing and network performance
Generalized databases and cache servers
Medium- and heavy-load enterprise applications
Gaming and rendering
- vCPU/Memory Ratio 1:2
- vCPUs 2-64
- Base/Turbo Frequency 2.6/3.5 GHz
- Maximum PPS 8,500,000
- Maximum Intranet Bandwidth 30 Gbit/s
Websites and web applications with high requirements on computing and network
performance; generalized databases and cache servers; medium- and heavy-load
enterprise applications; gaming and rendering
Latest-generation Intel® Xeon® Scalable processors and Huawei 25GE
high-speed intelligent NICs cost-effectively provide high network bandwidth
and PPS.
Websites and web applications with high requirements on PPS
Small-scale databases and cache servers
Light- and medium-load enterprise applications
- vCPU/Memory Ratio 1:1 | 1:2 | 1:4
- vCPUs 1-8
- Base/Turbo Frequency 2.6/3.5 GHz
- Maximum PPS 500,000
- Maximum Intranet Bandwidth 3 Gbit/s
Websites and web applications with high requirements on PPS; small-scale
databases and cache servers; light- and medium-load enterprise applications
Intel® Xeon® Scalable processors provide high network performance and stability.
Websites and web applications with high requirements on computing and network performance
Generalized databases and cache servers
Medium- and heavy-load enterprise applications
- vCPU/Memory Ratio 1:2 | 1:4
- vCPUs 2-60
- Base/Turbo Frequency 3.0/3.4 GHz
- Maximum PPS 5,000,000
- Maximum Intranet Bandwidth 17 Gbit/s
Websites and web applications with high requirements on computing and network performance; generalized databases and cache servers; medium- and heavy-load enterprise applications
Intel® Xeon® Scalable processors enable powerful and reliable computing, and
Huawei 25GE high-speed intelligent NICs provide ultra-high network bandwidth
and PPS.
High PPS scenarios
Enterprise-class applications with high network requirements, such as on-screen video comments and telecom business forwarding
Small- and medium-scale databases, cache servers, and search clusters
Data analysis and computing
- vCPU/Memory Ratio 1:2 | 1:4
- vCPUs 2-60
- Base/Turbo Frequency 3.0/3.4 GHz
- Maximum PPS 10,000,000
- Maximum Intranet Bandwidth 40 Gbit/s
High PPS scenarios; enterprise-class applications with high network requirements, such as on-screen video comments and telecom business forwarding; small- and medium-scale databases, cache servers, and search clusters; data analysis and computing
Intel® Xeon® Scalable processors provide high performance at low costs.
Websites and web applications
Small-scale databases and cache servers
Light- and medium-load enterprise applications
- vCPU/Memory Ratio 1:1 | 1:2 | 1:4
- vCPUs 1-16
- Base/Turbo Frequency 2.2/3.0 GHz
- Maximum PPS 300,000
- Maximum Intranet Bandwidth 4 Gbit/s
Websites and web applications; small-scale databases and cache servers; light- and medium-load enterprise applications
Baseline level of CPU performance with the ability to burst while keeping costs low, with performance limited by CPU baseline and credits.
Microservices
Low-latency interactive applications
Small- and medium-scale databases
Virtual desktops
Generalized-load websites and web applications, including development, build, and stage environments, code repositories, and product prototypes
- vCPU/Memory Ratio 1:1 | 1:2 | 1:4
- vCPUs 1-16
- Base/Turbo Frequency 2.2/3.0 GHz
- Maximum PPS 600,000
- Maximum Intranet Bandwidth 3 Gbit/s
Microservices; low-latency interactive applications; small- and medium-scale
databases; virtual desktops; generalized-load websites and web applications,
including development, build, and stage environments, code repositories, and
product prototypes
Latest-generation Intel® Xeon® Scalable processors enable powerful and
reliable computing; Huawei 25GE high-speed intelligent NICs provide
ultra-high network bandwidth and PPS; up to 512 GB DDR4 SDRAM is ideal
for memory-intensive applications.
MPP data warehouse
Distributed MapReduce and Hadoop computing
Distributed file systems
Network file systems
Log or data processing
- vCPU/Memory Ratio 1:8
- vCPUs 2-64
- Base/Turbo Frequency 3.0/3.4 GHz
- Maximum PPS 10,000,000
- Maximum Bandwidth 40 Gbit/s
MPP data warehouse; distributed MapReduce and Hadoop computing; distributed file systems; network file systems; log or data processing
Intel® Xeon® Scalable processors and Huawei 25GE high-speed intelligent NICs
provide ultra-high network bandwidth; up to 512 GB DDR4 SDRAM is ideal
for memory-intensive applications.
High-performance databases
In-memory databases
Distributed memory cache
Data analysis and mining
Hadoop or Spark clusters
- vCPU/Memory Ratio 1:8
- vCPUs 2-60
- Base/Turbo Frequency 3.0/3.4 GHz
- Maximum PPS 10,000,000
- Maximum Intranet Bandwidth 40 Gbit/s
High-performance databases; in-memory databases; distributed memory cache; data
analysis and mining; Hadoop or Spark clusters
Latest-generation hardware is ideal for processing large-memory datasets.
High-performance databases
In-memory databases
Distributed memory cache
Data analysis and mining
Hadoop or Spark clusters
- vCPU/Memory Ratio 1:8
- vCPUs 2-60
- Base/Turbo Frequency 3.0/3.4 GHz
- Maximum PPS 5,000,000
- Maximum Intranet Bandwidth 17 Gbit/s
High-performance databases; in-memory databases; distributed memory cache; data analysis and mining; Hadoop or Spark clusters
Intel® Xeon® Scalable processors enable powerful and reliable computing, and
Huawei 25GE high-speed intelligent NICs provide ultra-high network bandwidth
and PPS.
OLTP and OLAP applications
In-memory databases, such as SAP HANA SoH/S4H and BWoH/B4H
High-performance databases
Distributed memory cache
Big data engines
Data mining
- Memory 348-4096 GB
- vCPUs 28-208
- Base/Turbo Frequency 3.0/3.4 GHz
- Maximum PPS 5,000,000
- Maximum Intranet Bandwidth 25 Gbit/s
OLTP and OLAP applications; in-memory databases, such as SAP HANA SoH/S4H and BWoH/B4H; high-performance databases; distributed memory cache; big data engines; data mining
Intel® Xeon® Scalable processors and Huawei-proprietary NICs
provide high performance and stability.
Frontend clusters
Web servers
Scientific and engineering computing
Advertising
MMO gaming
Video encoding
Distributed analysis
- vCPU/Memory Ratio 1:2 | 1:4
- vCPUs 2-32
- Base/Turbo Frequency 3.2/4.2 GHz
- Maximum PPS 4,000,000
- Maximum Intranet Bandwidth 17 Gbit/s
Frontend clusters; web servers; scientific and engineering computing; advertising; MMO gaming; video encoding; distributed analysis
Latest-generation virtualization platform provides improved computing performance
compared to previous-generation ECSs.
Frontend clusters
Web servers
Scientific and engineering computing
Advertising
MMO gaming
Video encoding
Distributed analysis
- vCPU/Memory Ratio 1:2 | 1:4
- vCPUs 2-32
- Base/Turbo Frequency 2.6/3.5 GHz
- Maximum PPS 600,000
- Maximum Intranet Bandwidth 13 Gbit/s
Frontend clusters; web servers; scientific and engineering computing; advertising; MMO gaming; video encoding; distributed analysis
Intel® Xeon® Scalable processors and high-performance local NVMe SSDs deliver
high IOPS at low latency.
RDS databases
NoSQL databases, including Cassandra, MongoDB, and Redis
Elasticsearch
- vCPU/Memory Ratio 1:8
- vCPUs 8-64
- Base/Turbo Frequency 3.0/3.4 GHz
- Maximum PPS 5,000,000
- Maximum Intranet Bandwidth 25 Gbit/s
ElasticSearch, RDS, and NoSQL databases, including Cassandra, MongoDB, and
Redis
1/2/4/6/7 x 1.6 TB local NVMe SSDs
Intel® Xeon® Scalable processors enable powerful and reliable computing;
Huawei 25GE high-speed intelligent NICs provide ultra-high network bandwidth
and PPS; local SAS disks deliver high bandwidth at low costs.
MPP data warehouse
Distributed MapReduce and Hadoop computing
Distributed file systems
Network file systems
Log or data processing
- vCPU/Memory Ratio 1:8 | 1:10
- vCPUs 4-56
- Base/Turbo Frequency 3.0/3.4 GHz
- Maximum PPS 5,000,000
- Maximum Intranet Bandwidth 40 Gbit/s
MPP data warehouse; distributed MapReduce and Hadoop computing; distributed file systems; network file systems; log or data processing
2/4/8/12/16/24/28 x 1.8 TB local SAS disks
Intel® Xeon® E5 v4 processors enable rapid access to local storage.
MPP data warehouse
Distributed MapReduce and Hadoop computing
Distributed file systems
Network file systems
Log or data processing
- vCPU/Memory Ratio 1:8
- vCPUs 4-48
- Base/Turbo Frequency 2.6/3.5 GHz
- Maximum PPS 900,000
- Maximum Intranet Bandwidth 13 Gbit/s
MPP data warehouse; distributed MapReduce and Hadoop computing; distributed file systems; network file systems; log or data processing
2/4/8/12/16/24 x 1.8 TB local SAS disks
NVIDIA Tesla T4 GPUs enable real-time AI inference, with up to 130 TeraOPS
(TOPS) of INT8 performance.
AI inference, including machine vision, image classification, and natural language processing
Video encoding
Machine learning
Model training for light-traffic scenarios
- vCPU/Memory Ratio 1:4
- vCPUs 8/16/32
- Base/Turbo Frequency 3.0/3.4 or 2.6/3.5 GHz
- GPU Model NVIDIA T4
- Memory per GPU 16 GB
AI inference, including machine vision, image classification, and natural language processing; video encoding, machine learning,
and model training for light-traffic scenarios
NVIDIA T4 GPU with 16 GB memory
NVIDIA Tesla V100 GPUs enable high-performance graphics processing with
balanced vCPU/memory ratio and high-performance networking.
Desktop cloud
3D rendering
Graphics-intensive remote workstation
Application stream processing
Video encoding
Heavy-load graphics on servers
- vCPU/Memory Ratio 1:4
- vCPUs 32
- Base/Turbo Frequency 3.0/3.4 GHz
- GPU Model NVIDIA V100
- Memory per GPU 16 GB
Desktop cloud; 3D rendering; graphics-intensive remote workstation; application stream processing; video encoding; heavy-load graphics on servers
NVIDIA V100 GPU with 16 GB memory
NVIDIA Tesla V100 GPUs that are interconnected through NVLink are ideal for
deep learning and scientific computing.
Machine learning and computational fluid dynamics
Computational finance
Seismic analysis and molecular modeling
- vCPU/Memory Ratio 1:8
- vCPUs 8/16/32/64
- Base/Turbo Frequency 3.0/3.4 GHz
- GPU Model NVIDIA V100
- Memory per GPU 16 GB
Machine learning and computational fluid dynamics; computational finance; seismic analysis and molecular modeling
Why ECS?
Unleash the power of the cloud to take on workloads of any scale
Elastic, flexible cloud migration that is easy and inexpensive
- You can use a web-based console or open APIs to manage your ECSs.
- You can configure auto scaling policies to add or remove ECS instances automatically.
- ECS provides pay-per-use, yearly/monthly, and spot price billing modes for you to choose from.
Flagship products with robust performance
- ECSs with up to 208 vCPUs and 4,096 GiB of memory are available to deliver industry-leading computing power.
- 10-million PPS performance is industry-leading performance. Intranet bandwidth for individual ECSs reaches up to 44 Gbit/s.
- Thousands of ECSs can be provisioned within minutes.
.png)
A wide range of specifications, catering to your needs
- A broad range of servers are available for you to choose from: x86, Kunpeng, Ascend, and heterogeneous ECSs.
- You can use public, private, shared, and Marketplace images to create ECSs.
- ECSs and images can be deployed with just a few clicks without any configurations for website setup and application development.
A solid data platform, ensuring data reliability
- Nine nines (99.9999999%) of data durability, ensuring your data security and reliability.
- A variety of EVS disk types are available for different needs: extreme SSD, ultra-high I/O, general purpose SSD, and high I/O.
- You can share or encrypt disks, back up disks or servers, and create snapshots to satisfy demands of different services.
.png)
Multiple levels of protection, securing your data in the cloud
- Security vulnerabilities and threats can be detected, and configuration checks, suggestions, and best practices are provided to help you reduce or eliminate losses caused by viruses or online attacks.
- Professional anti-DDoS technologies are used to refine anti-DDoS capabilities for Internet applications.
- Working with the internationally renowned digital certificate service institute, Huawei provides one-stop lifecycle management of Secure Sockets Layer (SSL) certificates and Transport Layer Security (TLS) certificates to offer reliable identity authentication and secure data transmission for websites.
Application Scenarios
Run virtually any business-critical applications or compute tasks on Huawei Cloud
Requirements
No special requirements for CPU, memory, and disk space
Moderate bandwidth
Strong security and reliability
Minimal initial investment and low maintenance
Example Scenarios
Corporate websites | Website development and testing | Small databases
Recommended Instance Type
General computing ECSs, which provide a balance of computing, memory, and network resources. This ECS type is ideal for light- and medium-load applications.
Requirements
Large memory
High-volume data processing
Rapid data transfer and processing
Example Scenarios
Precision marketing | E-Commerce | Mobile apps
Recommended Instance Type
Memory-optimized ECSs, which have a large amount of memory, support ultra-high I/O EVS disks, and can be configured with flexible bandwidths. This ECS type is ideal for applications that process large volumes of data.
Requirements
Large data volume processing
High I/O performance and rapid data transfer and processing
Example Scenarios
MapReduce | Hadoop
Recommended Instance Type
Disk-intensive ECSs, which are designed for applications requiring sequential read/write on ultra-large datasets in local storage, as well as large-scale parallel data processing and log processing. Disk-intensive ECSs are HDD-compatible, and feature 10 Gigabit Ethernet network bandwidth, high throughput, and low network latency.
Requirements
High computing performance
High throughput
Example Scenarios
Scientific computing | Genetic engineering | Games and animation | Biopharmaceuticals and storage
Recommended Instance Type
High-performance computing ECSs, which provide a large number of parallel computing resources required by high-performance infrastructure services and applications in addition to meeting storage and rendering needs.
New Features
Documentation
Learn more about ECS