Upload
others
View
4
Download
0
Embed Size (px)
Citation preview
Elastic Cloud Server
Product Introduction
Issue 06
Date 2018-07-10
HUAWEI TECHNOLOGIES CO., LTD.
Copyright © Huawei Technologies Co., Ltd. 2018. All rights reserved.No part of this document may be reproduced or transmitted in any form or by any means without prior writtenconsent of Huawei Technologies Co., Ltd. Trademarks and Permissions
and other Huawei trademarks are trademarks of Huawei Technologies Co., Ltd.All other trademarks and trade names mentioned in this document are the property of their respectiveholders. NoticeThe purchased products, services and features are stipulated by the contract made between Huawei and thecustomer. All or part of the products, services and features described in this document may not be within thepurchase scope or the usage scope. Unless otherwise specified in the contract, all statements, information,and recommendations in this document are provided "AS IS" without warranties, guarantees orrepresentations of any kind, either express or implied.
The information in this document is subject to change without notice. Every effort has been made in thepreparation of this document to ensure accuracy of the contents, but all statements, information, andrecommendations in this document do not constitute a warranty of any kind, express or implied.
Huawei Technologies Co., Ltd.Address: Huawei Industrial Base
Bantian, LonggangShenzhen 518129People's Republic of China
Website: http://www.huawei.com
Email: [email protected]
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
i
Contents
1 ECS................................................................................................................................................... 1
2 Regions and AZs............................................................................................................................3
3 Storage............................................................................................................................................. 4
4 Network...........................................................................................................................................64.1 VPC................................................................................................................................................................................ 64.2 Network Bandwidth........................................................................................................................................................6
5 Security............................................................................................................................................ 75.1 User Encryption..............................................................................................................................................................75.2 Cloud-Init........................................................................................................................................................................95.3 Project...........................................................................................................................................................................10
6 IMS................................................................................................................................................. 11
7 Instances and Application Scenarios.......................................................................................137.1 General Computing ECSs.............................................................................................................................................147.2 General Computing-Plus ECSs.................................................................................................................................... 187.3 Memory-optimized ECSs............................................................................................................................................. 207.4 Disk-intensive ECSs..................................................................................................................................................... 237.5 Ultra-high I/O ECSs..................................................................................................................................................... 277.6 High-Performance Computing ECSs............................................................................................................................297.7 High-Performance Computing ECSs............................................................................................................................327.8 GPU-accelerated ECSs................................................................................................................................................. 347.8.1 Graphics-accelerated ECSs........................................................................................................................................357.8.2 Computing-accelerated ECSs.................................................................................................................................... 427.9 FPGA-accelerated ECSs...............................................................................................................................................47
8 Accessing and Using ECSs.........................................................................................................528.1 Accessing ECSs............................................................................................................................................................528.2 ECS Charging Standards.............................................................................................................................................. 528.3 ECSs and Other Services..............................................................................................................................................528.4 User Permissions.......................................................................................................................................................... 56
9 Change History............................................................................................................................ 57
Elastic Cloud ServerProduct Introduction Contents
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
ii
1 ECS
An ECS is a computing server consisting of CPUs, memory, images, and Elastic VolumeService (EVS) disks that allow on-demand allocation and elastic scaling. ECSs integrateVirtual Private Cloud (VPC), virtual firewalls, and multi-data-copy capabilities to create anefficient, reliable, and secure computing environment. This ensures stable and uninterruptedoperation of services. After creating an ECS, you can use it like using your local computer orphysical server.
ECSs support self-service creation, modification, and operation. You can create ECSs byspecifying the CPU, memory, image specifications, and login authentication mode. Aftercreating an ECS, you can modify its specifications as required.
System ArchitectureECS works with other public cloud products and services to provide computing, storage,network, and image installation functions.
l ECSs are deployed in multiple availability zones (AZs) connected with each otherthrough an internal network. If an AZ becomes faulty, other AZs in the same region willnot be affected.
l With the Virtual Private Cloud (VPC) service, you can build a dedicated network, set thesubnet and security group, and allow the VPC to communicate with the external networkthrough an EIP (bandwidth support required).
l With the Image Management Service (IMS), you can install images on ECSs, or createECSs using private images and deploy services quickly.
l The Elastic Volume Service (EVS) provides storage and Volume Backup Service (VBS)provides data backup and recovery functions.
Elastic Cloud ServerProduct Introduction 1 ECS
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
1
Figure 1-1 ECS architecture
Elastic Cloud ServerProduct Introduction 1 ECS
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
2
2 Regions and AZs
A region is a geographic area where resources used by your ECSs are located.
ECSs in the same region can communicate with each other over an intranet, but ECSs indifferent regions cannot.
Public cloud data centers are deployed worldwide in places, such as North America, Europe,and Asia. Creating ECSs in different regions can better suit certain user requirements. Forexample, applications can be designed to meet user requirements in specific regions orcomply with local laws or regulations. ECS pricing also changes based on region.
Each region contains many AZs where power and networks are physically isolated. AZs in thesame region can communicate with each other over an intranet. Each AZ provides cost-effective and low-latency network connections that are unaffected by faults that may occur inother AZs.
Elastic Cloud ServerProduct Introduction 2 Regions and AZs
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
3
3 Storage
EVS Disk Types
ECSs support the following types of EVS disks for storing data:
l Common I/O: EVS disks of this type deliver a maximum of 1,000 IOPS. This disk typeis suitable for application scenarios that require large capacity, a medium read/write rate,and fewer transactions, such as enterprises' office applications and small-scale test.
l High I/O: EVS disks of this type deliver a maximum of 3,000 IOPS and a minimum of 1ms read/write latency. This disk type is designed to meet the needs of mainstream high-performance, high-reliability application scenarios, such as enterprise applications, large-scale development and test, and web server logs.
l Ultra-high I/O: EVS disks of this type deliver a maximum of 20,000 IOPS and aminimum of 1 ms read/write latency. This disk type is excellent for ultra-high I/O, ultra-high bandwidth, and intensive read/write application scenarios. For example, thedistributed file systems in the HPC scenarios or NoSQL and relational databases in I/O-intensive scenarios.
l Ultra-high I/O (Latency optimized): EVS disks of this type deliver a maximum of 1GB/s throughput and a minimum of 1 ms latency. They can be used for enterprises' keyservices, such as SAP HANA and Oracle RAC.
NOTE
Currently, ultra-high I/O (latency optimization) EVS disks can be attached to SAP HANA ECSsonly.
EVS disks with different I/O capacities provide different features at different prices. ChooseEVS disks based on your requirements. For more information about EVS disk specificationsand performance, see Elastic Volume Service User Guide.
EVS Disk Device Types
EVS disks have two device types, Virtual Block Device (VBD) and Small Computer SystemInterface (SCSI).l VBD
When you create an EVS disk on the management console, Device Type of the EVS diskis VBD by default. VBD EVS disks support only simple SCSI read/write commands.
l SCSI
Elastic Cloud ServerProduct Introduction 3 Storage
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
4
You can create EVS disks for which Device Type is SCSI on the management console.These EVS disks support transparent SCSI command transmission, allowing ECS OS todirectly access underlying storage media. SCSI EVS disks support both basic andadvanced SCSI commands.
NOTE
For more information about how to use SCSI EVS disks, such as installing a driver for a SCSIEVS disk, see FAQs > Do I Need to Install a Driver for EVS Disks with Device Type SCSI? inElastic Volume Service User Guide.
Elastic Cloud ServerProduct Introduction 3 Storage
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
5
4 Network
4.1 VPCVirtual Private Cloud (VPC) allows you to create customized virtual networks in yourlogically isolated AZ on the public cloud. Such networks are dedicated zones that arelogically isolated for your ECSs. You can define security groups, virtual private networks(VPNs), IP address segments, and bandwidth for a VPC. This facilitates internal networkconfiguration and management as well as secure and convenient network modification. Youcan also customize the ECS access rules within a security group and between security groupsto strengthen ECS security protection. For more information, see Virtual Private Cloud UserGuide.
4.2 Network BandwidthThe intranet bandwidth and PPS of an ECS are determined based on ECS specifications.l Assured intranet bandwidth: indicates the assured ECS bandwidth.l Maximum intranet bandwidth: indicates the maximum ECS bandwidth.l Maximum intranet PPS: indicates the maximum ECS capabilities in transmitting and
receiving packets.
NOTE
l Network bandwidth has been released in South China, North China, and Hong Kong, and will bereleased in other regions.
l For instructions about how to test packet transmit and receive, see section How Can I Test NetworkPerformance?
l For instructions about how to enable the NIC multi-queue, see section Enabling NIC Multi-Queue.
Elastic Cloud ServerProduct Introduction 4 Network
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
6
5 Security
5.1 User EncryptionUser encryption allows you to use the encryption feature provided on the public cloudplatform to encrypt ECS resources, improving data security. User encryption includes imageencryption and EVS disk encryption.
Image Encryption
Key encryption supports encrypting private images. When creating an ECS, if you select anencrypted image, the system disk of the created ECS automatically has encryption enabled,implementing system disk encryption and improving data security.
Use either of the following methods to create an encrypted image:
l Create an encrypted image using an existing encrypted ECS.
l Create an encrypted image using an external image file.
For more information about image encryption, see Image Management Service User Guide.
EVS Disk Encryption
EVS disk encryption supports system disk encryption and data disk encryption.
l When creating an ECS, you can encrypt added data disks.
l System disk encryption relies on the image. When creating an ECS, if you select anencrypted image, the system disk of the created ECS automatically has encryptionenabled, and the encryption mode complies with the image encryption mode.
For more information about EVS disk encryption, see Elastic Volume Service User Guide.
Impact on AS
If you use an encrypted ECS to create an Auto Scaling (AS) configuration, the encryptionmode of the created AS configuration complies with the ECS encryption mode.
Elastic Cloud ServerProduct Introduction 5 Security
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
7
About Keys
The key required for encryption relies on Data Encryption Workshop (DEW). DEW uses adata encryption key (DEK) to encrypt data and a customer master key (CMK) to encrypt theDEK.
Figure 5-1 Data encryption process
Table 5-1 describes the keys involved in the data encryption process.
Table 5-1 Keys
Name Description Function
DEK An encryption key that is used for encryptingdata.
Encrypts specific data.
CMK An encrypted key that is created using KMS.A CMK encrypts and protects DEKs.A CMK can encrypt multiple DEKs.
Supports CMK disablingand scheduled deletion.
Default CMK A type of CMK with suffix /default. Thedefault CMK is automatically generated bythe system when you use KMS forencryption for the first time.For example, evs/default
l Supports viewingdetails of the defaultCMK on the KMSconsole.
l Does not supportCMK disabling orscheduled deletion.
NOTE
After disabling a CMK or scheduling the deletion of a CMK takes effect, the EVS disk encrypted usingthis CMK can still be used until the disk is detached from and then attached to an ECS again. Duringthis process, the disk fails to be attached to the ECS because the CMK cannot be obtained. Therefore,the EVS disk becomes unavailable.
For details about DEW, see Data Encryption Workshop User Guide.
Elastic Cloud ServerProduct Introduction 5 Security
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
8
5.2 Cloud-InitCloud-Init is an open-source cloud initialization program, which initializes specifiedcustomized configurations, such as the host name, key pair, and user data, of a newly createdECS.
Using Cloud-Init to initialize your ECSs will affect your ECS, IMS, and AS services.
Impact on IMS
To ensure that ECSs created using private images support customized configurations, youmust install Cloud-Init or Cloudbase-Init before creating private images.
l For Windows OSs, download and install Cloudbase-Init.l For Linux OSs, download and install Cloud-Init.
After Cloud-Init or Cloudbase-Init is installed in an image, Cloud-Init or Cloudbase-Initautomatically configures initial ECS attributes when the ECS is created. For instructionsabout how to install Cloud-Init or Cloudbase-Init, see Image Management Service UserGuide.
Impact on ECSl When creating an ECS, if the selected image supports Cloud-Init, you can use user data
injection to inject customized configuration, such as ECS login password, forinitializing.
l After Cloud-Init is supported, ECS login modes will be affected.l After Cloud-Init is supported, you can view and use metadata to configure and manage
running ECSs.
Impact on ASl When creating an AS configuration, you can use user data injection to specify ECS
configurations for initialization. If the AS configuration has taken effect in an AS group,the ECSs newly created in the AS group will automatically initialize their configurations.
l For an existing AS configuration, if its private image has Cloud-Init or Cloudbase-Initnot installed, the login mode of the ECSs created in the AS group where the ASconfiguration takes effect will be influenced. To resolve this issue, see section "HowDoes Cloud-Init Influence the AS Service?" in Auto Scaling User Guide.
For more information about AS, see Auto Scaling User Guide.
Notesl When using Cloud-Init, enable DHCP in the VPC to which the ECS belongs.l When using Cloud-Init, ensure that security group rules in the outbound direction meet
the following requirements:– Protocol: TCP– Port Range: 80– Remote End: 169.254.0.0/16
Elastic Cloud ServerProduct Introduction 5 Security
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
9
NOTE
If you use the default security group rules in the outbound direction, the preceding requirementsare met, and the metadata can be accessed. Default security group rules in the outbound directionare as follows:
l Protocol: ANY
l Port Range: ANY
l Remote End: 0.0.0.0/16
5.3 ProjectA project groups and isolates OpenStack computing, storage, and network resources. Aproject can be a department or a team.
Multiple projects can be created for one account.
Elastic Cloud ServerProduct Introduction 5 Security
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
10
6 IMS
Introduction
Image Management Service (IMS) allows you to create ECSs using images. An image is anECS template that contains an OS and may also contain proprietary software and applicationsoftware, such as database software.
Images can be public, private, or shared. Public images are provided by the system by default,private images are manually created, and shared images are private images that are shared byanother user. You can use any type of image to create an ECS. You can also create a privateimage using an existing ECS. This provides you with a simple way to create ECSs thatcomply with your service requirements. For example, if you use web services, your image cancontain a web server, static configurations, and dynamic page code. When you use this imageto create an ECS, your web server and applications will be available for use immediately.
Image Types
Table 6-1 Image types
ImageType
Description
Public image A standard, widely used image. It contains an OS, comes with preinstalledpublic applications, and is available to all users.
Privateimage
An image that is available only to the user who created it based on an ECSor EVS backup (system disk backup). Such a private image contains an OS,preinstalled public applications, and the user's private applications.A private image can be a system image or a data image.l System image: contains an OS, comes with preinstalled application
software for running services. You can use a system image to createECSs and migrate your services to the cloud.
l Data image: contains only your service data. You can use a data imageto create EVS disks and migrate your service data to the cloud.
Sharedimage
A private image shared by another user.
Elastic Cloud ServerProduct Introduction 6 IMS
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
11
ImageType
Description
Marketplaceimage
A third-party image that has the OS, application environment, and softwarepre-installed. You can use the images to deploy websites and applicationdevelopment environments with a few clicks. No additional configurationoperation is required.
Conversion Between Images and ECSsYou can use images to create ECSs and you can convert ECS configurations into images.
Elastic Cloud ServerProduct Introduction 6 IMS
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
12
7 Instances and Application Scenarios
The public cloud provides various ECS types for different application scenarios.
l General computing
l General computing-plus
l Memory-optimized
l Disk-intensive
l Ultra-high I/O
l High-performance computing
l Ultra-high performance computing
l GPU-accelerated
l FPGA-accelerated
ECS Flavor Naming Rules
ECS flavors are named using the format "AB.C.D".
The format is defined as follows:
l A specifies the ECS type. For example, s indicates a general-purpose ECS, c acomputing ECS, and m a memory-optimized ECS.
l B specifies the type ID. For example, the 1 in s1 indicates a general-purpose first-generation ECS, and the 2 in s2 indicates a general-purpose second-generation ECS.
l C can be any of the following options:
– medium
– large
– xlarge
l D specifies the ratio of memory to vCPUs expressed in a digit. For example, value 4indicates that the ratio of memory to vCPUs is 4.
Obtaining Specifications When Creating an ECS
Specifications for the ECS being created are located in the specifications list.
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
13
Network Bandwidth
The intranet bandwidth and PPS of an ECS are determined based on ECS specifications.
l Assured intranet bandwidth: indicates the assured ECS bandwidth.
l Maximum intranet bandwidth: indicates the maximum ECS bandwidth.
l Maximum intranet PPS: indicates the maximum ECS capabilities in transmitting andreceiving packets.
NOTE
l Network bandwidth has been released in South China, North China, and Hong Kong, and will bereleased in other regions.
l For instructions about how to test packet transmit and receive, see section How Can I Test NetworkPerformance?
l For instructions about how to enable the NIC multi-queue, see section Enabling NIC Multi-Queue.
7.1 General Computing ECSs
Overview
General computing ECSs provide basic vCPU performance and a balance of computing,memory, and network resources. These ECSs are suitable for many applications, such as webservers, enterprise R&D, and small-scale databases.
Compared with S1 ECSs, S2 ECSs are based on the latest-generation virtualization platformand use non-uniform memory access (NUMA) binding to provide higher computingperformance.
Compared with S1 and S2 ECSs, S3 ECSs use latest-generation Intel Xeon SkyLake CPUs,which significantly improve the comprehensive performance. They provide basic vCPUperformance and a balance of computing, memory, and network resources. These ECSs aresuitable for many applications.
Compared with S3 ECSs, SN3 ECSs feature improved network performance. Working in 25Gbit/s networks, SN3 ECSs offer higher bandwidths, lower latency, more stable computingperformance for high PPS and cost-effectiveness.
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
14
Specifications
Table 7-1 S1 ECS specifications
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth
MaximumPPS
VirtualizationType
s1.medium 1 4 Low Low XEN
s1.large 2 8 Low Low XEN
s1.xlarge 4 16 Medium Medium XEN
s1.2xlarge 8 32 Medium Medium XEN
s1.4xlarge 16 64 Medium Medium XEN
s1.8xlarge 32 128 Medium Medium XEN
Table 7-2 C1 ECS specifications
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth
MaximumPPS
VirtualizationType
c1.medium 1 1 Low Low XEN
c1.large 2 2 Low Low XEN
c1.xlarge 4 4 Medium Medium XEN
c1.2xlarge 8 8 Medium Medium XEN
c1.4xlarge 16 16 Medium Medium XEN
c1.8xlarge 32 32 Medium Medium XEN
Table 7-3 C2 ECS specifications
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth
MaximumPPS
VirtualizationType
c2.medium 1 2 Low Low XEN
c2.large 2 4 Low Low XEN
c2.xlarge 4 8 Medium Medium XEN
c2.2xlarge 8 16 Medium Medium XEN
c2.4xlarge 16 32 Medium Medium XEN
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
15
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth
MaximumPPS
VirtualizationType
c2.8xlarge 32 64 Medium Medium XEN
Table 7-4 S2 ECS specifications
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth(Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
Virtualization Type
s2.small.1 1 1 0.5/0.1 5 1 KVM
s2.medium.2
1 2 0.5/0.1 5 1 KVM
s2.large.2 2 4 0.8/0.2 10 1 KVM
s2.xlarge.2 4 8 1.5/0.4 15 1 KVM
s2.2xlarge.2 8 16 3/0.8 20 2 KVM
s2.4xlarge.2 16 32 4/1.5 30 4 KVM
s2.8xlarge.2 32 64 6/3 50 8 KVM
s2.medium.4
1 4 0.5/0.1 5 1 KVM
s2.large.4 2 8 0.8/0.2 10 1 KVM
s2.xlarge.4 4 16 1.5/0.4 15 1 KVM
s2.2xlarge.4 8 32 3/0.8 20 2 KVM
s2.4xlarge.4 16 64 4/1.5 30 4 KVM
s2.8xlarge.4 32 128 6/3 50 8 KVM
Table 7-5 S3 ECS specifications
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth(Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
Virtualization Type
s3.small.1 1 1 0.5/0.1 5 1 KVM
s3.medium.2
1 2 0.5/0.1 5 1 KVM
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
16
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth(Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
Virtualization Type
s3.large.2 2 4 0.8/0.2 10 1 KVM
s3.xlarge.2 4 8 1.5/0.4 15 1 KVM
s3.2xlarge.2 8 16 3/0.8 20 2 KVM
s3.4xlarge.2 16 32 4/1.5 30 4 KVM
s3.medium.4
1 4 0.5/0.1 5 1 KVM
s3.large.4 2 8 0.8/0.2 10 1 KVM
s3.xlarge.4 4 16 1.5/0.4 15 1 KVM
s3.2xlarge.4 8 32 3/0.8 20 2 KVM
s3.4xlarge.4 16 64 4/1.5 30 4 KVM
Table 7-6 SN3 ECS specifications
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth(Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
Virtualization Type
sn3.small.1 1 1 0.8/0.2 10 1 KVM
sn3.medium.2
1 2 0.8/0.2 10 1 KVM
sn3.large.2 2 4 1.5/0.35 15 1 KVM
sn3.xlarge.2 4 8 2/0.7 25 1 KVM
sn3.2xlarge.2
8 16 3/1.3 50 2 KVM
sn3.4xlarge.2
16 32 6/2.5 100 4 KVM
sn3.medium.4
1 4 0.8/0.2 10 1 KVM
sn3.large.4 2 8 1.5/0.35 15 1 KVM
sn3.xlarge.4 4 16 2/0.7 25 1 KVM
sn3.2xlarge.4
8 32 3/1.3 50 2 KVM
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
17
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth(Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
Virtualization Type
sn3.4xlarge.4
16 64 6/2.5 100 4 KVM
Scenariosl Web servers, small-load applications, and development and test environments
l Small-and medium-sized databases, cache servers, and search clusters
7.2 General Computing-Plus ECSs
Overview
Compared with general computing ECSs, the general computing-plus ECSs provide thecombinations of vCPUs and memory with larger specifications, offering more options for youto select. In addition, the ECSs use latest-generation network acceleration engines and DPDKrapid packet processing mechanism to provide higher network performance, meetingrequirements in different scenarios.
l C3 ECSs are newly released on the public cloud platform. They use latest-generationIntel Xeon SkyLake CPUs and feature high and stable computing performance. Workingin high-performance networks, the C3 ECSs provide higher performance and stability,meeting enterprise-class application requirements.
l C3ne ECSs provide higher computing and network forwarding capabilities than C3ECSs. Using latest-generation Intel Xeon Skylake CPUs and high-speed smart Hi1822NICs, the C3ne ECSs provide 50GE and 15 million PPS for enterprise-class applicationswith high network performance requirements.
Specifications
Table 7-7 C3 ECS specifications
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth (Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
Virtualization Type
c3.large.2 2 4 1.5/0.6 30 2 KVM
c3.xlarge.2 4 8 3/1 50 2 KVM
c3.2xlarge.2 8 16 5/2 90 4 KVM
c3.3xlarge.2 12 24 7/3 110 4 KVM
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
18
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth (Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
Virtualization Type
c3.4xlarge.2 16 32 10/4 130 4 KVM
c3.6xlarge.2 24 48 12/6 200 8 KVM
c3.8xlarge.2 32 64 15/8 260 8 KVM
c3.15xlarge.2
60 128 17/16 500 16 KVM
c3.large.4 2 8 1.5/0.6 30 2 KVM
c3.xlarge.4 4 16 3/1 50 2 KVM
c3.2xlarge.4 8 32 5/2 90 4 KVM
c3.3xlarge.4 12 48 7/3 110 4 KVM
c3.4xlarge.4 16 64 10/4 130 4 KVM
c3.6xlarge.4 24 96 12/6 200 8 KVM
c3.8xlarge.4 32 128 15/8 260 8 KVM
c3.15xlarge.4
60 256 17/16 500 16 KVM
Table 7-8 C3ne ECS specifications
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth(Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
Virtualization Type
c3ne.large.2 2 4 5/1.3 40 2 KVM
c3ne.xlarge.2
4 8 10/2.5 80 2 KVM
c3ne.2xlarge.2
8 16 15/5 150 4 KVM
c3ne.4xlarge.2
16 32 20/10 280 8 KVM
c3ne.8xlarge.2
32 64 30/20 550 16 KVM
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
19
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth(Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
Virtualization Type
c3ne.15xlarge.2
60 128 40/40 1000 32 KVM
c3ne.large.4 2 8 5/1.3 40 2 KVM
c3ne.xlarge.4
4 16 10/2.5 80 2 KVM
c3ne.2xlarge.4
8 32 15/5 150 4 KVM
c3ne.4xlarge.4
16 64 20/10 280 8 KVM
c3ne.8xlarge.4
32 128 30/20 550 16 KVM
c3ne.15xlarge.4
60 256 40/40 1000 32 KVM
Scenariosl C3 ECSs
Small- and medium-scale databases, cache servers, and search clusters with highrequirements on stability; enterprise-class applications of diverse types and in variousscales
l C3ne ECSs
Network data forwarding, databases, cache servers, and search clusters with highrequirements on computing and network performance; enterprise-class applications ofdiverse types and in various scales
7.3 Memory-optimized ECSs
Overviewl M1 ECSs have a large memory size and provide high memory performance. They are
designed for memory-intensive applications that process a large amount of data, such asprecision advertising, e-commerce big data analysis, and IoV big data analysis.
l M2 ECSs use Intel Xeon E5-2690 v4 CPUs and are designed for memory-optimizedapplications.
l M3 ECSs are developed based on the KVM virtualization platform and designed forprocessing large-scale data sets in the memory. They use latest-generation Intel XeonSkyLake CPUs, network acceleration engines, and DPDK rapid packet processing
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
20
mechanism to provide higher network performance, providing a maximum memory sizeof 512 GB based on DDR4 for high-memory computing applications.
l M3ne ECSs are suited for large-memory datasets with high network performancerequirements. Using latest-generation Intel Xeon SkyLake CPUs and high-speed smartHi1822 NICs, the M3ne ECSs provide a maximum memory size of 512 GB based onDDR4 for large-memory applications with high requirements on network performance.
Specifications
Table 7-9 M1 ECS specifications
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth
MaximumPPS
VirtualizationType
m1.medium 1 8 Low Low XEN
m1.large 2 16 Low Low XEN
m1.xlarge 4 32 Medium Medium XEN
m1.2xlarge 8 64 Medium Medium XEN
m1.4xlarge 16 128 Medium Medium XEN
Table 7-10 M2 ECS specifications
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth (Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
Virtualization Type
m2.large.8 2 16 1.5/0.5 10 1 KVM
m2.xlarge.8 4 32 3/1 15 1 KVM
m2.2xlarge.8
8 64 5/2 30 2 KVM
m2.4xlarge.8
16 128 8/4 40 4 KVM
m2.8xlarge.8
32 256 13/8 60 8 KVM
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
21
Table 7-11 M3 ECS specifications
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth (Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
Virtualization Type
m3.large.8 2 16 1.5/0.6 30 2 KVM
m3.xlarge.8 4 32 3/1.1 50 2 KVM
m3.2xlarge.8
8 64 5/2 90 4 KVM
m3.3xlarge.8
12 96 8/3.5 110 4 KVM
m3.4xlarge.8
16 128 10/4.5 130 4 KVM
m3.6xlarge.8
24 192 12/6.5 200 8 KVM
m3.8xlarge.8
32 256 15/9 260 8 KVM
m3.15xlarge.8
60 512 17/17 500 16 KVM
Table 7-12 M3ne ECS specifications
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth(Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
VirtualizationType
m3ne.large.8 2 16 5/1.3 40 2 KVM
m3ne.xlarge.8 4 32 10/2.5 80 2 KVM
m3ne.2xlarge.8 8 64 15/5 150 4 KVM
m3ne.4xlarge.8 16 128 20/10 280 8 KVM
m3ne.8xlarge.8 32 256 30/20 550 16 KVM
m3ne.15xlarge.8
60 512 40/40 1000 32 KVM
Scenariosl High-performance relational (MySQL) and NoSQL (MongoDB and Cassandra)
databases
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
22
l Distributed web scale cache stores that provide in-memory caching of key-value typedata (Memcached and Redis)
l Applications processing big unstructured data in real time (financial services, Hadoop/Spark clusters)
l High-performance computing (HPC) and electronic design automation (EDA)
Notes on Using M2 ECSsl M2 ECSs support all rollout OSs.
l M2 ECSs do not have InfiniBand or SSD NICs configured.
l M2 ECSs support modifying specifications if the source and target ECSs are of the sametype.
Notes on Using M3 ECSsl M3 ECSs support all rollout OSs.
l M3 ECSs do not have InfiniBand or SSD NICs configured.
l M3 ECSs support modifying specifications if the source and target ECSs are of the sametype.
7.4 Disk-intensive ECSs
Overview
D1 ECSs are developed based on the XEN virtualization platform, use local storage, andprovide high network performance. They are designed for applications requiring sequentialread/write on ultra-large datasets in local storage (such as distributed Hadoop computing) aswell as large-scale parallel data processing and log processing.
Compared with D1 ECSs, D2 ECSs are developed based on KVM virtualization, use localstorage, and provide high storage performance and intranet bandwidth. They are designed fordistributed Hadoop computing, large data warehouse, distributed file system, data processing,and log processing.
NOTE
D1 ECSs have been discontinued. D2 ECSs that feature higher performance are available for new ordersand capacity expansion for D1 ECSs.
Specifications
Table 7-13 D1 ECS specifications
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth
MaximumPPS
LocalDisks(GB)
Virtualization Type
d1.xlarge 4 32 Medium Medium 3 x1800
XEN
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
23
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth
MaximumPPS
LocalDisks(GB)
Virtualization Type
d1.2xlarge 8 64 Medium Medium 6 x1800
XEN
d1.4xlarge 16 128 Medium Medium 12 x1800
XEN
d1.9xlarge 36 256 Medium Medium 24 x1800
XEN
Table 7-14 D2 ECS specifications
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth (Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
LocalDisks(GB)
VirtualizationType
d2.xlarge.8 4 32 3/1 15 2 2 x1800
KVM
d2.2xlarge.8
8 64 5/2 30 2 4 x1800
KVM
d2.4xlarge.8
16 128 8/4 40 4 8 x1800
KVM
d2.6xlarge.8
24 192 10/6 50 6 12 x1800
KVM
d2.8xlarge.8
32 256 13/8 60 8 16 x1800
KVM
d2.12xlarge.8
48 384 12/12 90 8 24 x1800
KVM
Scenariosl Applications
Disk-intensive ECSs are suitable for applications that require large volumes of data toprocess, high I/O performance, and rapid data switching and processing.
l Application scenariosBig data computing, network file systems, data processing, MapReduce, Hadoop, anddata-intensive computing
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
24
Features of D2 ECSsl D2 ECSs use local disks to provide high sequential read/write performance and low
latency, improving file read/write performance.l D2 ECSs provide powerful and stable computing capabilities, ensuring efficient data
processing.l D2 ECSs with a CPU/memory ratio of 1:8 process large volumes of data.l D2 ECSs provide high intranet performance, including high intranet bandwidth and
packet per second (pps), meeting requirements for data exchange between ECSs duringpeak hours.
Notes on Using D1 ECSsl D1 ECSs do not support NIC hot swapping.l The primary and extension NICs of a D1 ECS have specified application scenarios. For
details, see Table 7-15.
Table 7-15 Application scenarios of the NICs of a D1 ECS
NIC Type Application Scenario Remarks
Primary NIC Applies to vertical layer 3communication.
N/A
Extension NIC Applies to horizontal layer2 communication.
To improve network performance, youcan set the MTU of the extension NICto 8888.
l D1 ECSs do not support modifying specifications.l D1 ECSs do not support OS reinstallation or change.l D1 ECSs support the following OSs:
– CentOS 7.2 64bit– CentOS 7.3 64bit– CentOS 6.8 64bit– SUSE Enterprise Linux Server 11 SP3 64bit– SUSE Enterprise Linux Server 11 SP4 64bit– SUSE Enterprise Linux Server 12 SP1 64bit– SUSE Enterprise Linux Server 12 SP2 64bit– Red Hat Enterprise Linux 6.8 64bit– Red Hat Enterprise Linux 7.3 64bit
l D1 ECSs can use both local disks and EVS disks to store data. Use restrictions on thetwo types of storage media are as follows:– Only an EVS disk, not a local disk, can be used as the system disk of a D1 ECS.– Both an EVS disk and a local disk can be used as the data disk of a D1 ECS.– A D1 ECS can be attached with up to 24 disks, and the total number of attached
disks and NICs is suggested to be at most 25. Otherwise, disks may fail to beattached to the ECS.
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
25
l The basic resources, including vCPUs, memory, and image of a stopped D1 ECS are stillcharged. To stop charging such an ECS, delete it.
Notes on Using D2 ECSsl D2 ECSs support the following OSs:
– CentOS 6.7/6.8/7.2/7.3/7.4 64bit– SUSE Enterprise Linux Server 11 SP3/SP4 64bit– SUSE Enterprise Linux Server 12 SP1/SP2 64bit– Red Hat Enterprise Linux 6.8/7.3 64bit– Windows Server 2008 R2 Enterprise 64bit– Windows Server 2012 R2 Standard 64bit– Windows Server 2016 Standard 64bit– Debian 8.7/9/9.0.0 64bit– EulerOS 2.2 64bit– Fedora 25/26 64bit– OpenSUSE 42.2/42.3 64bit
l When the physical host where a D2 ECS is deployed becomes faulty, the ECS cannot bemigrated.
l To improve network performance, you can set the NIC MTU to 8888.l D2 ECSs do not support modifying specifications.l D2 ECSs do not support local disk snapshot or backup.l D2 ECSs do not support OS reinstallation or change.l D2 ECSs do not support automatic recovery.l D2 ECSs can use both local disks and EVS disks to store data. In addition, they can have
EVS disks attached to provide a larger storage size. Use restrictions on the two types ofstorage media are as follows:– Only an EVS disk, not a local disk, can be used as the system disk of a D2 ECS.– Both an EVS disk and a local disk can be used as the data disk of a D2 ECS.– A D2 ECS can be attached with up to 24 disks (not including local disks).– You are advised to use World Wide Names (WWNs), but not drive letters, in
applications to perform operations on local disks to prevent drive letter drift (lowprobability) on Linux. Take local disk attachment as an example:If the local disk WWN is wwn-0x50014ee2b14249f6, run the mount /dev/disk/by-id/wwn-0x50014ee2b14249f6 command.
NOTE
How can I view the local disk WWN?1. Log in to the ECS.2. Run the following command to view the WWN:
ll /dev/disk/by-id
l The basic resources, including vCPUs, memory, and image of a stopped D2 ECS are stillcharged. To stop charging such an ECS, delete it.
l The local disk data of a D2 ECS may be lost due to some reasons, such as host machinebreakdown or local disk damage. If your application does not use the data reliabilityarchitecture, you are strongly advised to use EVS disks to build your ECS.
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
26
l When you delete a D2 ECS, the data on the local disk is automatically deleted.l Do not store service data for a long time in local disks. Instead, use EVS disks to store
the data. In addition, back up data in a timely manner and use a high availabilityarchitecture.
l You are not allowed to buy additional local disks. The quantity and capacity of yourlocal disks are determined according to the specifications of your ECS. For D2 ECSs, ifadditional local disks are required, buy them when creating the ECSs.
7.5 Ultra-high I/O ECSs
OverviewUltra-high I/O ECSs use high-performance local NVMe SSDs to provide high storage IOPSand low read/write latency. The ratio of vCPU to memory is 1:8. You can create such ECSswith high-performance NVMe SSDs attached on the management console.
Ultra-high I/O ECSs can be used for high-performance relational databases, NoSQL databases(such as Cassandra and MongoDB), and ElasticSearch search.
Specifications
Table 7-16 Ultra-high I/O I3 ECS specifications
Flavor vCPUs Memory (GB)
Maximum/AssuredBandwidth(Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
LocalDisks
VirtualizationType
i3.2xlarge.8
8 64 8/3.5 100 4 1 x1600GBNVMe
KVM
i3.4xlarge.8
16 128 15/7 160 4 2 x1600GBNVMe
KVM
i3.8xlarge.8
32 256 20/14 280 8 4 x1600GBNVMe
KVM
i3.12xlarge.8
48 384 25/20 420 8 6 x1600GBNVMe
KVM
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
27
Flavor vCPUs Memory (GB)
Maximum/AssuredBandwidth(Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
LocalDisks
VirtualizationType
i3.15xlarge.8
60 512 25/25 500 16 7 x1600GBNVMe
KVM
FeaturesTable 7-17 lists the IOPS performance of I3 ECSs.
Table 7-17 IOPS performance
Flavor Maximum IOPS for Random 4 KBRead
i3.2xlarge.8 750000
i3.4xlarge.8 1500000
i3.8xlarge.8 3000000
i3.12xlarge.8 4500000
i3.15xlarge.8 5250000
Notesl I3 ECSs do not support specifications modification.l After an I3 ECS is deleted, the data on the local NVMe SSD is automatically deleted.
Back up the data before deleting it.l The data reliability of local disks depends on the reliability of physical servers and hard
disks, which are SPOF-prone. Therefore, you are advised to perform data redundancy atthe application layer to ensure data availability. Use EVS disks to store long-term servicedata.
l I3 ECSs support the following OSs:o EulerOS 2.2o CentOS 7.2o CentOS 7.3o Ubuntu Server 16.04o SUSE Linux Enterprise Server 12 SP2o Fedora 25 64bito OpenSUSE 42.2 64bit
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
28
NOTE
You are advised to use EulerOS 2.2 or Ubuntu Server 16.04 for optimal performance.
l If you use an I3 ECS created using a private image, make sure that the NVMe driver hasbeen installed during private image creation. If the driver has not been installed, install itafter creating the I3 ECS. For details, see section "(Optional) Installing SpecialDrivers" in Image Management Service User Guide.
l An I3 ECS can be attached with up to 24 VBD disks and 30 SCSI disks.l The basic resources, including vCPUs, memory, and image of an I3 ECS are still
charged. To stop charging such an ECS, delete it.
7.6 High-Performance Computing ECSs
OverviewH1 ECSs provide a large number of CPU cores, large memory size, and high throughput.These ECSs are suitable for high-performance processor applications restricted by computingperformance.
The processor and memory ratio of an HC2 ECS is 1:2 or 1:4. Each vCPU corresponds to thehyperthreading of an Intel Xeon processor core. HC2 ECSs can be used for high-performancecomputing services. They provide a large number of parallel computing resources and high-performance infrastructure services to meet the requirements of high-performance computingand massive storage and ensure rendering efficiency. HC2 ECSs are frequently used in thefollowing scenarios:l Computing and storage systems for genetic engineering, games, animations,
biopharmaceuticals, and scientific computingl Public rendering platforms for renderfarms and animation and film bases; other
rendering platforms for movies and videosl High-performance frontend clusters, web servers, high-performance science and
engineering applications, advertisements, video coding, and distributed analysis
H3 ECSs use high-performance Intel Xeon SkyLake CPUs. Each vCPU corresponds to thehyper-thread of an Intel Xeon processor core, providing stable computing capabilities. H3ECSs are suitable for high-performance computing services. In addition, the ECSs use latest-generation network acceleration engines and DPDK rapid packet processing mechanism toprovide high network performance.
NOTE
H1 ECSs have been discontinued. HC2 ECSs that feature higher performance are available for neworders and capacity expansion for H1 ECSs.
Specifications
Table 7-18 H1 ECS specifications
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth
MaximumPPS
VirtualizationType
h1.xlarge 4 8 Medium Medium XEN
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
29
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth
MaximumPPS
VirtualizationType
h1.2xlarge 8 16 Medium Medium XEN
h1.4xlarge 20 40 Medium Medium XEN
h1.8xlarge 44 88 Medium Medium XEN
Table 7-19 HC2 ECS specifications
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth (Gbit/s)
MaximumPPS(10,000)
NICMulti-Queue
Virtualization Type
hc2.large.2
2 4 1.5/0.5 10 1 KVM
hc2.xlarge.2
4 8 3/1 15 1 KVM
hc2.2xlarge.2
8 16 5/2 30 2 KVM
hc2.4xlarge.2
16 32 8/4 40 4 KVM
hc2.8xlarge.2
32 64 13/8 60 8 KVM
hc2.large.4
2 8 1.5/0.5 10 1 KVM
hc2.xlarge.4
4 16 3/1 15 1 KVM
hc2.2xlarge.4
8 32 5/2 30 2 KVM
hc2.4xlarge.4
16 64 8/4 40 4 KVM
hc2.8xlarge.4
32 128 13/8 60 8 KVM
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
30
Table 7-20 H3 ECS specifications
Flavor vCPUs Memory (GB)
Maximum/AssuredBandwidth (Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
Virtualization Type
h3.large.2 2 4 2/1 30 2 KVM
h3.xlarge.2 4 8 4/2 60 2 KVM
h3.2large.2 8 16 6/3.5 120 4 KVM
h3.3xlarge.2
12 24 6/5.5 160 4 KVM
h3.4large.2 16 32 12/7.5 200 8 KVM
h3.6xlarge.2
24 48 15/11 300 8 KVM
h3.8large.2 32 64 17/15 400 16 KVM
h3.large.4 2 8 2/1 30 2 KVM
h3.xlarge.4 4 16 4/2 60 2 KVM
h3.2large.4 8 32 6/3.5 120 4 KVM
h3.3xlarge.4
12 48 6/5.5 160 4 KVM
h3.4large.4 16 64 12/7.5 200 8 KVM
h3.6xlarge.4
24 96 15/11 300 8 KVM
h3.8large.4 32 128 17/15 400 16 KVM
Scenariosl Computing and storage systems for genetic engineering, games, animations, and
biopharmaceuticalsl Public rendering platforms for renderfarms and animation and film bases; other
rendering platforms for movies and videosl High-performance frontend clusters, web servers, high-performance science and
engineering applications, advertisements, video coding, and distributed analysisl Batch-processed workload and high-performance computing (HPC)l Computing-intensive services, such as large-scale multiplayer online (MMO) gaming
Notes on Using H1 ECSsl H1 ECSs do not support NIC hot swapping.l H1 ECSs support modifying specifications only if the source and target ECSs are of the
same type.
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
31
l H1 ECSs support the following OSs:– CentOS 6.8 64bit– CentOS 7.2 64bit– CentOS 7.3 64bit– Windows Server 2008– Windows Server 2012– Windows Server 2016– SUSE Enterprise Linux Server 11 SP3 64bit– SUSE Enterprise Linux Server 11 SP4 64bit– SUSE Enterprise Linux Server 12 SP1 64bit– SUSE Enterprise Linux Server 12 SP2 64bit– Red Hat Enterprise Linux 6.8 64bit– Red Hat Enterprise Linux 7.3 64bit
l The primary and extension NICs of an H1 ECS have specified application scenarios. Fordetails, see Table 7-21.
Table 7-21 Application scenarios of the NICs of an H1 ECS
NIC Type Application Scenario Remarks
Primary NIC Applies to vertical layer 3communication.
N/A
Extension NIC Applies to horizontal layer2 communication.
To improve network performance,you can set the MTU of theextension NIC to 8888.
7.7 High-Performance Computing ECSs
OverviewHigh-performance computing ECSs are designed to meet high-end computational needs, suchas industrial simulation, molecular modeling, and computational fluid dynamics. In additionto the substantial CPU power, the high-performance computing ECSs offer diverse options forlow-latency RDMA networking using EDR InfiniBand NICs to support memory-intensivecomputational requirements.
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
32
Specifications
Table 7-22 H2 ECS specifications
Flavor vCPUs
Memory (GB)
Maximum/AssuredBandwidth(Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
LocalDisks(TB)
VirtualizationType
h2.4xlarge.8
16 128 8/8 90 8 1 x 3.2 KVM
h2.4xlarge.16
16 256 8/8 90 8 1 x 3.2 KVM
Table 7-23 HI3 ECS specifications
Flavor vCPUs Memory (GB)
Maximum/AssuredBandwidth(Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
LocalDisks(TB)
VirtualizationType
hi3.9xlarge.4
36 144 13/13 90 8 1 x 1.6 KVM
Featuresl Large memory capacity and more processor cores than other types of ECSsl InfiniBand NICs providing a bandwidth of 100 Gbit/sl 100 Gbit/s EDR InfiniBand network
Notes on Using H2 ECSsl H2 ECSs do not support OS reinstallation or change.l H2 ECSs do not support specifications modification.l H2 ECSs do not support cold migration, live migration, or HA.l H2 ECSs support the following OSs:
– For public images:n CentOS 7.3 64bitn SUSE Linux Enterprise Server 11 SP4 64bitn SUSE Linux Enterprise Server 12 SP2 64bit
– For private images:n CentOS 6.5 64bit
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
33
n CentOS 7.2 64bitn CentOS 7.3 64bitn SUSE Linux Enterprise Server 11 SP4 64bitn SUSE Linux Enterprise Server 12 SP2 64bitn Red Hat Enterprise Linux 7.2 64bitn Red Hat Enterprise Linux 7.3 64bit
l H2 ECSs use InfiniBand NICs that provide a bandwidth of 100 Gbit/s.l Each H2 ECS uses one PCIe 3.2 TB SSD card for temporary local storage.l If an H2 ECS is created using a private image, install an InfiniBand NIC driver on the
ECS after the ECS creation following the instructions provided by Mellanox. Downloadthe required version (4.2-1.0.0.0) of InfiniBand NIC driver from the official Mellanoxwebsite and install the driver by following the instructions provided by Mellanox.– InfiniBand NIC type: Mellanox Technologies ConnectX-4 Infiniband HBA
(MCX455A-ECAT)– Mellanox official website: http://www.mellanox.com/– NIC driver download path: http://www.mellanox.com/page/products_dyn?
product_family=26&mtag=linux_sw_driversl If Internet Protocol over Infiniband (IPoIB) is required on an H2 ECS running a SUSE
OS, you must manually allocate an IP address to the InfiniBand NIC after installing theInfiniBand NIC driver.
l After you delete an H2 ECS, the data stored in SSD disks is automatically cleared.Therefore, do not store persistence data into SSD disks during ECS running.
l An H2 ECS is charged even if it is stopped. To stop charging such an ECS, delete it.
7.8 GPU-accelerated ECSsGPU-accelerated ECSs provide outstanding floating-point computing capabilities. They aresuitable for scenarios that require real-time, highly concurrent massive computing.
GPU-accelerated ECSs are classified as graphics-accelerated (G series) and computing-accelerated (P series) ECSs.l P series ECSs are designed for deep learning, scientific computing, and CAE.l G series ECSs are suitable for 3D animation rendering and CAD.
Scenariosl Applications
GPU-accelerated ECSs are suitable for applications with high requirements on real-timeperformance and massive concurrent computing, such as deep learning, scientificcomputing, CAE, 3D animation rendering, and CAD.
l Application scenarios– P1 artificial intelligence: Each GPU contains thousands of computing units,
providing outstanding parallel computing capabilities. P1 ECSs have beenoptimized for deep learning, supporting massive computing within a short period oftime.
– P1 scientific computing: Scientific computing has strict requirements on double-precision computing. During computing emulation, a large number of computing
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
34
resources are used, and large volumes of data are generated. Therefore, scientificcomputing also has strict requirements on storage bandwidth and latency. P1 ECSsperfectly resolve this issue.
– G1 graphic workstation: G1 ECSs provide outstanding computing capabilities forprofessional CAD, video rendering, and graphics processing.
7.8.1 Graphics-accelerated ECSs
Overviewl G1 ECSs are based on NVIDIA GRID virtual GPUs and provide economical graphics
acceleration. They use NVIDIA Tesla M60 GPUs and support DirectX and OpenGL.The ECSs have a maximum of 8 GB GPU memory and 4,096 x 2,160 resolution, and areused for applications that require high performance in graphics rendering.
l G3 ECSs are based on PCI passthrough and exclusively use GPUs for professionalgraphics acceleration. In addition, G3 ECSs use NVIDIA Tesla M60 GPUs and supportDirectX and OpenGL with up to 16 GB GPU memory and a maximum of 4,096 x 2,160resolution. They are ideal for professional graphics workstations.
Select your required GPU-accelerated ECS type and specifications.
Specifications
Table 7-24 G1 ECS specifications
Flavor vCPUs Memory(GB)
Maximum/AssuredBandwidth
Maximum PPS
GPUs GPUMemory (GB)
VirtualizationType
g1.xlarge 4 8 Medium Medium 1 xM60-1Q
1 XEN
g1.xlarge.4
4 16 Medium Medium 1 xM60-1Q
1 XEN
g1.2xlarge 8 16 Medium Medium 1 xM60-2Q
2 XEN
g1.2xlarge.8
8 64 Medium Medium Passthrough
8 XEN
g1.4xlarge 16 32 Medium Medium 1 xM60-4Q
4 XEN
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
35
NOTE
In M60-xQ, x can be 1, 2, 4, or 8, indicating that M60 GPUs are virtualized to vGPUs with differentspecifications and models using GRID. x specifies the vGPU memory, and Q indicates that the vGPU ofthis type is designed to work in workstations and desktop scenarios. For more details about GPUvirtualization using GRID, see GRID VIRTUAL GPU User Guide.
Table 7-25 G3 ECS specifications
Flavor vCPUs
Memory (GB)
Maximum/AssuredBandwidth(Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
GPUs
GPUMemory (GB)
VirtualizationType
g3.4xlarge.4
16 64 8/2.5 50 2 1 xM60
1 x 8 KVM
g3.8xlarge.4
32 128 10/5 100 4 2 xM60
2 x 8 KVM
Featuresl G1 ECSs have the following features:
– NVIDIA M60 GPUs– Graphics acceleration applications– GPU hardware virtualization (vGPUs)– Application flow identical to common ECSs– Automatic scheduling of G1 ECSs to AZs where NVIDIA M60 GPUs are used– A maximum specification of 4 GB GPU memory and 4,096 x 2,160 resolution for
processing graphics and videosl G3 ECSs have the following features:
– Professional graphics acceleration APIs– NVIDIA M60 GPUs– Graphics acceleration applications– GPU hardware virtualization (vGPUs)– Application flow identical to common ECSs– Automatic scheduling of G3 ECSs to AZs where NVIDIA M60 GPUs are used– A maximum specification of 16 GB GPU memory and 4,096 x 2,160 resolution for
processing graphics and videos
Notes on Using G1 ECSsl G1 ECSs do not support specifications modification.l G1 ECSs support the following OSs:
– Windows Server 2008 R2 Enterprise SP1 64bit– Windows Server 2012 R2 Standard 64bit
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
36
– Windows Server 2016 DataCenterl g1.2xlarge.8 G1 ECSs do not support the remote login function provided by the public
cloud platform. To remotely log in to such an ECS, use MSTSC to log in to it and installVNC on the ECS.Non-g1.2xlarge.8 G1 ECSs support the remote login function provided by the publiccloud platform. For details, see section Logging In to an ECS Using VNC.
l If a G1 ECS is created using a private image, install a GPU driver on the ECS after theECS creation. To download the driver, log in at http://www.nvidia.com/grid-eval, setthe NVIDIA GRID version to 4.1, and select the GRID for UVP software package. Theoperations are as follows:
a. Check whether NVIDIA is used for the first time:
i. If yes, go to step b.ii. If no, go to step d.
b. Obtain the Product Activation Key (PAK) from the email indicating successfulregistration with NVIDIA.
Figure 7-1 PAK
c. Enter the PAK obtained in step b on the Redeem Product Activation Keys pageand click Redeem.
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
37
Figure 7-2 Redeem Product Activation Keys
d. Specify Username and Password and click LOGIN.
Figure 7-3 Logging in to the official NVIDIA website
e. Log in at the official NVIDIA website as prompted and choose Software &Services > Product Information.
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
38
f. Click the Archived Versions tab.
g. Click NVIDIA GRID of version 4.1.
h. On the Product Download page, click GRID for UVP.
l If you log in to a G1 ECS using a RDP-based remote login tool, such as WindowsMSTSC, GPU acceleration will fail. This is because MSTSC replaces the WDDM GPUdriver with a non-accelerated remote desktop display driver. Therefore, you must useother methods to log in to the ECS, such as VNC.
If the remote login function available on the management console fails to meet yourservice requirements, you must install a suitable remote login tool on the ECS.
Notes on Using G3 ECSsl G3 ECSs do not support specifications modification.
l G3 ECSs support the following OSs:
– Windows Server 2008 R2 Enterprise SP1 64bit
– Windows Server 2012 R2 Standard 64bit
l G3 ECSs do not support the remote login function provided by the public cloudplatform. To remotely log in to such an ECS, use MSTSC to log in to it and install VNCon the ECS.
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
39
l If a G3 ECS is created using a private image, install a GPU driver on the ECS after theECS creation. To download the driver, log in at http://www.nvidia.com/grid-eval, setthe NVIDIA GRID version to 4.1, and select the GRID for UVP software package. Theoperations are as follows:
a. Check whether NVIDIA is used for the first time:
i. If yes, go to step b.ii. If no, go to step f.
b. Obtain the Product Activation Key (PAK) from the email indicating successfulregistration with NVIDIA.
Figure 7-4 PAK
c. Enter the PAK obtained in step b on the Redeem Product Activation Keys pageand click Redeem.
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
40
Figure 7-5 Redeem Product Activation Keys
d. Specify Username and Password and click LOGIN.
Figure 7-6 Logging in to the official NVIDIA website
e. Log in at the official NVIDIA website as prompted and choose Software &Services > Product Information.
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
41
f. Click the Archived Versions tab.
g. Click NVIDIA GRID of version 4.1.
h. On the Product Download page, click GRID for UVP.
l If you log in to a G3 ECS using a RDP-based remote login tool, such as WindowsMSTSC, GPU acceleration will fail. This is because the remote login tool replaces theWDDM GPU driver with a non-accelerated remote desktop display driver. In such anevent, you must use other methods to log in to the ECS, such as VNC.
If the remote login function available on the management console fails to meet yourservice requirements, for example, a blank screen is displayed, you must install a suitableremote login tool on the ECS.
7.8.2 Computing-accelerated ECSs
Overview
P1 ECSs use industry-leading NVIDIA Tesla P100 GPUs and provide flexibility, highperformance, and cost-effectiveness. These ECSs support GPU Direct for directcommunication between GPUs, improving data transmission efficiency. P1 ECSs provideoutstanding universal computing capabilities and have strengths in deep learning, graphicdatabases, high-performance databases, Computational Fluid Dynamics (CFD), computing
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
42
finance, seismic analysis, molecular modeling, and genomics. They are designed for scientificcomputing.
Compared with P1 ECSs, P2 ECSs use the most advanced Tesla V100 GPUs, which haveimproved both single- and double-precision computing capabilities by 50% and offering 112TFLOPS of deep learning.
PI1 ECSs use NVIDIA Tesla P4 GPUs dedicated for AI inference. Working with P4 INT8calculators, PI1 ECSs have shortened the inference latency by 15 times. Working withhardware decoding engines, PI1 ECSs concurrently support real-time 35-channel HD videotranscoding and inference.
Specifications
Table 7-26 P1 ECS specifications
Flavor vCPUs
Memory(GB)
Max./AssuredBandwidth(Gbit/s)
MaximumPPS(10,000)
NICMulti-Queue
GPUs GPUMemory(GB)
LocalDisks
VirtualizationType
p1.2xlarge.8
8 64 5/1.6 35 2 1 xP100
1 x 16 1 x800GBNVMe
KVM
p1.4xlarge.8
16 128 8/3.2 70 4 2 xP100
2 x 16 2 x800GBNVMe
KVM
p1.8xlarge.8
32 256 10/6.5 140 8 4 xP100
4 x 16 4 x800GBNVMe
KVM
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
43
Table 7-27 P2 ECS specifications
Flavor vCPUs
Memory(GB)
Max./AssuredBandwidth(Gbit/s)
MaximumPPS(10,000)
NICMulti-Queue
GPUs GPUMemory(GB)
LocalDisks
VirtualizationType
p2.2xlarge.8
8 64 5/1.6 35 2 1 xV100
1 x 16 1 x800GBNVMe
KVM
Table 7-28 PI1 ECS specifications
Flavor vCPUs
Memory(GB)
Max./AssuredBandwidth(Gbit/s)
MaximumPPS(10,000)
NICMulti-Queue
GPUs
GPUMemory(GB)
LocalDisks
VirtualizationType
pi1.2xlarge.4
8 32 5/1.6 40 2 1 xP4
1 x 8GB
N/A KVM
pi1.4xlarge.4
16 64 8/3.2 70 4 1 xP4
1 x 8GB
N/A KVM
pi1.8xlarge.4
32 128 10/6.5 140 8 4 xP4
4 x 8GB
N/A KVM
pi1.6xlarge.4
24 96 10/5 100 4 1 xP4
1 x 8 1 x800GBNVMe
KVM
pi1.12xlarge.4
48 192 10/10 200 8 2 xP4
2 x 8 2 x800GBNVMe
KVM
Featuresl P1 ECSs have the following features:
– NVIDIA Tesla P100 GPUs
– 9.3 TFLOPS for single precision and 4.7 TFLOPS for double precision
– Maximum bandwidth of 10 Gbit/s
– 16 GB HBM2 GPU memory with a bandwidth of 732 Gbit/s
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
44
– 800 GB NVMe SSD disks for temporary local storage– Comprehensive basic capabilities
Networks are user-defined, subnets can be divided, and network access policies canbe configured as needed. Mass storage is used, elastic capacity expansion as well asbackup and restoration are supported to make data more secure. Auto Scalingallows you to add or reduce the number of ECSs quickly.
– FlexibilitySimilar to other types of ECSs, P1 ECSs can be provisioned in a few minutes. Youcan configure specifications as needed. P1 ECSs with two, four, and eight GPUswill be supported later.
– Excellent supercomputing ecosystemThe supercomputing ecosystem allows you to build up a flexible, high-performance, cost-effective computing platform. A large number of HPCapplications and deep-learning frameworks can run on P1 ECSs.
l P2 ECSs have the following features:– NVIDIA Tesla V100 GPUs– 14 TFLOPS of single-precision computing, 7 TFLOPS of double-precision
computing, and 112 TFLOPS of deep learning– Maximum bandwidth of 12 Gbit/s– 16 GB HBM2 GPU memory with a bandwidth of 900 Gbit/s– 800 GB NVMe SSD disks for temporary local storage– Comprehensive basic capabilities
Networks are user-defined, subnets can be divided, and network access policies canbe configured as needed. Mass storage is used, elastic capacity expansion as well asbackup and restoration are supported to make data more secure. Auto Scalingallows you to add or reduce the number of ECSs quickly.
– FlexibilitySimilar to other types of ECSs, P2 ECSs can be provisioned in a few minutes.
– Excellent supercomputing ecosystemThe supercomputing ecosystem allows you to build up a flexible, high-performance, cost-effective computing platform. A large number of HPCapplications and deep-learning frameworks can run on P2 ECSs.
l PI1 ECSs have the following features:– NVIDIA Tesla P4 GPUs– Up to 5.5 TFLOPS by a single GPU for single precision– Up to 22 TOPS INT8 by a single GPU– 8 GB ECC GPU memory with a bandwidth of 192 Gbit/s by a single GPU– Hardware video encoding and decoding engines embedded in GPUs for concurrent
real-time 35-channel HD video transcoding and inference
Notes on Using P1 ECSsl The system disk size of a P1 ECS must be greater than or equal to 15 GB. It is
recommended that the system disk size be greater than 40 GB.l If a P1 ECS is created using a private image, make sure that the NVIDIA driver has been
installed during the private image creation. If not, install the driver after the P1 ECS is
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
45
created for computing acceleration. For instructions about how to install the NVIDIAdriver, see section Installing the NVIDIA Driver on a P1 ECS.
NOTE
For instructions about how create a private image, see Image Management Service User Guide.
l P1 ECSs do not support specifications modification.l P1 ECSs support the following OSs:
– Debian 8.0 64bit– Ubuntu Server 16.04 64bit– CentOS 7.3 64bit– EulerOS 2.2 64bit
l After you delete a P1 ECS, the data stored in local NVMe SSD disks is automaticallycleared. Therefore, you are advised to store only temporary data in local NVMe SSDdisks. If you store important data in the disks, back up the data.
Notes on Using P2 ECSsl If a P2 ECS is created using a private image, make sure that the NVIDIA driver has been
installed during the private image creation. If not, install the driver after the P2 ECS iscreated for computing acceleration. For instructions about how to install the NVIDIAdriver, see section Installing the NVIDIA GPU Driver and CUDA Toolkit on a P2ECS.
l P2 ECSs do not support specifications modification.l P2 ECSs support the following OS:
– Ubuntu Server 16.04 64bit– EulerOS 2.2 64bit
l After you delete a P2 ECS, the data stored in local NVMe SSD disks is automaticallycleared. Therefore, you are advised to store only temporary data in local NVMe SSDdisks. If you store important data in the disks, back up the data.
Notes on Using PI1 ECSsl If a PI1 ECS is created using a private image, make sure that a GPU driver has been
installed during the private image creation. If not, install the driver after the PI1 ECS iscreated for inference acceleration. To download the driver, log in at http://www.nvidia.com/Download/Find.aspx?lang=en. Figure 7-7 shows parameter settings.
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
46
Figure 7-7 NVIDIA Driver Downloads
l PI1 ECSs support the following OSs:– Ubuntu Server 14.04 64bit– CentOS 7.3 64bit
l PI1 ECSs do not support specifications modification.l After you delete a PI1 ECS, the data stored in local NVMe SSD disks is automatically
cleared. Therefore, you are advised to store only temporary data in local NVMe SSDdisks. If you store important data in the disks, back up the data.
7.9 FPGA-accelerated ECSs
Overview
An FPGA-accelerated ECS provides a tool and environment for developing and using FPGA.With it, you can easily develop FPGA accelerators, deploy FPGA-based services, and provideeasy-to-use, cost-effective, agile, and secure FPGA cloud services.
FFPGA-accelerated ECSs are classified as the following types:l ECSs using a high-performance architecture
The ECSs of this type are based on high-performance DPDK interaction framework andsupport the flow calculation model and data flow concurrency. They are designed forRTL development and meet the requirements of high bandwidth and low latency.
l ECSs using a universal architectureThe ECSs of this type are based on the general-purpose SDAccel interaction frameworkand support the block calculation module and Xilinx SGDMA data transmission. Theyare designed for advanced language development or existing algorithm transplantationand meet the requirements of fast service rollout.
NOTE
FPGA-accelerated ECSs are in the beta test phase.
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
47
Specifications
Table 7-29 OBT specifications for ECSs with high-performance architecture
Flavor vCPUs Memory (GB)
Maximum/AssuredBandwidth (Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
FPGAs
VirtualizationType
fp1.2xlarge.11.dev
8 88 5/1.3 20 2 1 xVU9P
KVM
fp1.4xlarge.11.dev
16 176 8/2.5 40 4 1 xVU9P
KVM
fp1.8xlarge.11.dev
32 352 10/5 60 8 4×VU9P
KVM
fp1.16xlarge.11.dev
64 704 10/10 100 8 8×VU9P
KVM
NOTE
Network: 10GbE
Table 7-30 OBT specifications for ECSs with general-computing architecture
Flavor vCPUs Memory (GB)
Maximum/AssuredBandwidth (Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
FPGAs
VirtualizationType
fp1c.2xlarge.11.dev
8 88 5/1.3 20 2 1 xVU9P
KVM
fp1c.4xlarge.11.dev
16 176 8/2.5 40 4 1 xVU9P
KVM
fp1c.8xlarge.11.dev
32 352 10/5 60 8 4 xVU9P
KVM
fp1c.16xlarge.11.dev
64 704 10/10 100 8 8 xVU9P
KVM
NOTE
Network: 10GbE
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
48
Table 7-31 FPGA-accelerated ECS specifications for commercial use
Flavor vCPUs Memory (GB)
Maximum/AssuredBandwidth (Gbit/s)
Maximum PPS(10,000)
NICMulti-Queue
FPGAs
VirtualizationType
fp1.2xlarge.11
8 88 5/1.3 20 2 1 xVU9P
KVM
fp1.8xlarge.8
32 224 10/5 60 8 1 xVU9P
KVM
fp1c.2xlarge.11
8 88 5/1.3 20 2 1 xVU9P
KVM
fp1c.8xlarge.8
32 224 10/5 60 8 1 xVU9P
KVM
NOTE
Network: 10GbE
Scenariosl Applications
FPGA-accelerated ECSs are suitable for applications that require dense computing, highconcurrency, and high bandwidth, such as video processing, machine learning, genomics,and financial risk analysis.
l Application scenarios– Video processing: Video applications, such as automatic image recognition and
classification, image search, video transcoding, real-time rendering, Internet-basedlive programs, and AR/VR, require high real-time computing performance, whichcannot be provided by common ECSs. FPGA ECSs offer cost-effective videosolutions, which are ideal for video scenarios.
– Machine learning: Multi-layer neural networks in machine learning require a largenumber of computing resources. The training process involves handling massivedata, while the inference process requires ultra-low latency. In addition, machinelearning algorithms are being continuously optimized. FPGA ECSs meet thepreceding requirements due to their high parallel computing, programmablehardware, low power consumption, and low latency. FPGA dynamically providesthe optimal hardware circuit design for different machine learning algorithms,meeting strict requirements on massive computing and ultra-low latency. Therefore,FPGA ECSs meet hardware requirements for machine learning.
– Genomics research: Precision medicine can be implemented through genesequencing and analysis as well as rapid analysis on massive biological and medicaldata. Many fields, such as pharmaceutical development and molecular breeding,also require the processing of massive data. These fields require hardwareacceleration to resolve performance bottleneck problems for biologicalcomputation. FPGA ECSs meet such requirements due to their outstandingprogrammable hardware computing performance.
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
49
– Financial risk analysis: The financial industry has strict requirements on computingcapabilities and real-time performance based on ultra-low latency and highthroughput for services, such as pricing tree model-based financial computing,high-frequency trading, fund or securities trading algorithms, financial risk analysisand decision-making, and transaction security assurance. Using programmablehardware acceleration, FPGA ECSs offer an optimal hardware acceleration solutionfor various scenarios. In certain scenarios, FPGA ECS performance has beenimproved by thousands of times when compared with the performance of stand-alone software.
Featuresl FPGA development kit
FPGA cloud services provide the hardware development kit (HDK). The HDK containsaccelerator examples, the encoding environment, simulation platform, automaticcompilation tool, and code encrypting and commissioning tool package. You can quicklydevelop and test your FPGA hardware accelerator by referring to the applicationexamples and development guide.
l Software development kitFPGA cloud services provide the software development kit (SDK). The SDK containsapplication examples, hardware abstract interfaces, accelerator abstract interfaces,accelerator driver and runtime, and the version management tool. Accelerator abstractinterfaces allow your applications to invoke the hardware accelerator in the same way asthe software function library, helping you easily develop high-performance applicationsbased on the hardware accelerator.
l FPGA hardware configurationEach FPGA ECS provides up to eight FPGAs and each of them contains 2.5 millionlogic units. The PCIe 3.0 x 16 interface is supported, and the throughput reaches 100Gbit/s. FPGAs communicate with each other through a mesh optical network with abandwidth of 300 Gbit/s. Each FPGA provides 64 GB DDR4 memory and an interfacerate of 2133 MHz. FPGA frees your applications from hardware restrictions.
l Hardware acceleration resource poolFPGA cloud services have a pool of hardware acceleration resources. You can allocatedesirable FPGA resources in a cost-effective way like allocating CPU resources. Withthe FPGA virtualization, isolation, and distributed technology, you can share FPGAresources on a node, which is transparent to your resources and can meet hardwareacceleration requirements of your business to the largest extent.
Notesl FPGA-accelerated ECSs support only the CentOS 7.3 64bit OS.l FPGA-accelerated ECSs do not support modifying specifications.l FPGA-accelerated ECSs cannot be migrated.l FPGA-accelerated ECSs do not support automatic recovery.l FPGA-accelerated ECSs are in the beta test phase. If such an ECS becomes faulty, you
will have to delete it and create another one. In this process, data in the original ECS willbe lost.
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
50
Follow-up ProcedureAfter the ECS is created, you can develop and use accelerated engine images (AEIs) throughthe hardware development kit (HDK) and application development kit (SDK) provided by theECS. For more details, see FPGA-accelerated Cloud Server User Guide.
Related LinksFACS User Guide
Elastic Cloud ServerProduct Introduction 7 Instances and Application Scenarios
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
51
8 Accessing and Using ECSs
8.1 Accessing ECSsThe public cloud provides a web-based service management platform. You can access ECSsthrough HTTPS-compliant application programming interfaces (APIs) or the managementconsole. These two access modes differ as follows:
l Accessing ECSs through APIsUse this mode if you are required to integrate the ECSs on the public cloud into a third-party system for secondary development. For detailed operations, see Elastic CloudServer API Reference.
l Accessing ECSs through the management consoleUse this mode if you are not required to integrate ECSs with a third-party system. Afterregistering on the public cloud, log in to the management console and click ElasticCloud Server under Computing on the homepage. If you have not registered withHUAWEI CLOUD, see Register on HUAWEI CLOUD Management Console.
8.2 ECS Charging StandardsECSs can be charged in Yearly/Monthly or Metered mode. You can choose one chargingmode based on your requirements.
8.3 ECSs and Other ServicesFigure 8-1 shows the relationships between ECS and other services.
Elastic Cloud ServerProduct Introduction 8 Accessing and Using ECSs
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
52
Figure 8-1 Relationships between ECS and other services
l Auto Scaling (AS)
Automatically adjusts ECS service resources based on the configured AS policies. Thisimproves resource usage and reduces resource costs.
l Elastic Load Balance (ELB)
Automatically distributes traffic to multiple ECSs. This enhances system service andfault tolerance capabilities.
l Elastic Volume Service (EVS)
Enables you to attach EVS disks to an ECS and expand their capacity.
l Virtual Private Cloud (VPC)
Enables you to configure internal networks and change network configurations bycustomizing security groups, VPNs, IP address segments, and bandwidth. This simplifiesnetwork management. You can also customize the ECS access rules within a securitygroup and between security groups to strengthen ECS security protection.
l Image Management Service (IMS)
Enables you to create ECSs using images. This improves the efficiency of ECS creation.
l Dedicated Computing Cluster (DCC)
To physically isolate your ECS, apply for a DCC before creating the ECS. After youobtain the DCC and set a region for it, your ECS is automatically allocated to the DCC.
Elastic Cloud ServerProduct Introduction 8 Accessing and Using ECSs
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
53
l Cloud EyeAllows you to check the status of monitored service objects after you have obtained anECS. This can be done without requiring additional plug-ins be installed. Table 8-1 liststhe ECS metrics supported by Cloud Eye.
Table 8-1 ECS monitoring metrics
Metric Description Formula Remarks
CPU Usage Indicates the CPU usage(%) of an ECS.
CPU usage of anECS/Number ofCPU cores in theECS
N/A
MemoryUsage
Indicates the memoryusage (%) of an ECS.
Used memory of anECS/Total memoryof the ECS
This metric isunavailable ifthe image hasno vmtoolsinstalled.
Disk Usage Indicates the disk usage(%) of an ECS.
Used capacity of anECS disk/Totalcapacity of the ECSdisk
This metric isunavailable ifthe image hasno vmtoolsinstalled.
Disks ReadRate
Indicates the number ofbytes read from an ECSper second.
Total number ofbytes read from anECS disk/Monitoringperiod
byte_out =(rd_bytes -last_rd_bytes)/Timedifference
Disks WriteRate
Indicates the number ofbytes written to an ECSper second.
Total number ofbytes written to anECS disk/Monitoringperiod
N/A
Disks ReadRequests
Indicates the number ofread requests sent to anECS per second.
Total number of readrequests sent to anECS disk/Monitoringperiod
req_out =(rd_req -last_rd_req)/Timedifference
Disks WriteRequests
Indicates the number ofwrite requests sent to anECS per second.
Total number ofwrite requests sent toan ECS disk/Monitoring period
req_in =(wr_req -last_wr_req)/Timedifference
InbandIncoming Rate
Indicates the number ofincoming bytes on an ECSper second.
Total number ofinband incomingbytes on an ECS/Monitoring period
N/A
Elastic Cloud ServerProduct Introduction 8 Accessing and Using ECSs
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
54
Metric Description Formula Remarks
InbandOutgoing Rate
Indicates the number ofoutgoing bytes on an ECSper second.
Total number ofinband outgoingbytes on an ECS/Monitoring period
N/A
OutbandIncoming Rate
Indicates the number ofincoming bytes on an ECSper second at thevirtualization layer.
Total number ofoutband incomingbytes on an ECS/Monitoring period
This metric isunavailable ifSR-IOV isenabled.
OutbandOutgoing Rate
Indicates the number ofoutgoing bytes on an ECSper second at thevirtualization layer.
Total number ofoutband outgoingbytes on an ECS/Monitoring period
This metric isunavailable ifSR-IOV isenabled.
System StatusCheck Failed
Check the status of thecloud platform for runningECSs. The check result is0 or 1.l 0: The system is
running properly. Allcheck items arenormal.
l 1: The system is notrunning properly. Atleast one check itemfailed.
The systemperiodically checksthe status and returnscheck results usingvalue 0 or 1.l If the detected
fault does notaffect ECSfunctions, certainmanagementoperationsperformed on theECS, such asstarting, stopping,or specificationsmodifications,may be affected.
l If the detectedfault affects ECSfunctions, such asa host powersupply failure, thesystem willrecover the ECSas soon aspossible.
When thepower sourceof the physicalhost fails orthe hardware/softwarebecomesfaulty, thecheck result is1.
l Data Encryption Workshop (DEW)The encryption feature relies on DEW. You can use an encrypted image or EVS diskswhen creating an ECS. In such a case, you are required to use the key provided by DEWto improve data security.
l Cloud Trace Service (CTS)Allows you to record ECS-related operations for later query, audit, and backtrack.
l Cloud Server Backup Service (CSBS)
Elastic Cloud ServerProduct Introduction 8 Accessing and Using ECSs
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
55
Protects ECS backups. CSBS backs up all EVS disks of an ECS, including the systemdisk and data disks, and uses the backup to restore the ECS.
8.4 User PermissionsThe public cloud system provides two types of permissions by default: user management andresource management.
l User management refers to the management of users, user groups, and user group rights.l Resource management refers to the control operations that can be performed by users on
cloud service resources.
For further details, see Permission Description.
Elastic Cloud ServerProduct Introduction 8 Accessing and Using ECSs
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
56
9 Change History
Released On Description
2018-07-10 This issue is the sixth official release, which incorporates thefollowing changes:Modified the following content:l Added OBT flavors fp1.16xlarge.11.dev and fp1c.16xlarge.
11.dev in section 7.9 FPGA-accelerated ECSs.l Added flavors c3.3xlarge.2, c3.6xlarge.2, c3.3xlarge.4, and
c3.6xlarge.4 in section 7.2 General Computing-Plus ECSs.l Added flavors m3.3xlarge.8 and m3.6xlarge.8 in section 7.3
Memory-optimized ECSs.l Added flavors h3.3xlarge.2, h3.6xlarge.2, h3.3xlarge.4, and
h3.6xlarge.4 in section 7.6 High-Performance ComputingECSs.
2018-07-02 This issue is the fifth official release, which incorporates thefollowing changes:Added the following content:l Added network QoS in section 4.2 Network Bandwidth.Modified the following content:l Officially released I3 ECSs for commercial use in section 7.5
Ultra-high I/O ECSs.l Added three types of PI1 flavors in section 7.8.2 Computing-
accelerated ECSs.l Added network QoS in section 7 Instances and Application
Scenarios.
Elastic Cloud ServerProduct Introduction 9 Change History
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
57
Released On Description
2018-06-11 This issue is the fourth official release, which incorporates thefollowing changes:Modified the following content:l Modified description in sections 7.4 Disk-intensive ECSs and
7.5 Ultra-high I/O ECSs, allowing I3 and D2 ECSs to beattached with up to 24 VBD disks and 30 SCSI disks.
l Added notes on using D2 ECSs in section 7.4 Disk-intensiveECSs.
l Released C3ne, M3ne, SN3, and HI3 ECSs.
2018-05-25 This issue is the third official release, which incorporates thefollowing changes:Modified the following content:l Added the g1.2xlarge.8 flavor in section 7.8 GPU-accelerated
ECSs.
2018-04-05 This issue is the second official release, which incorporates thefollowing changes:Added the following content:l Added the virtualization type for each ECS flavor.Modified the following content:l 7.8 GPU-accelerated ECSsl 7.9 FPGA-accelerated ECSs
2018-01-26 This issue is the first official release.
Elastic Cloud ServerProduct Introduction 9 Change History
Issue 06 (2018-07-10) Huawei Proprietary and ConfidentialCopyright © Huawei Technologies Co., Ltd.
58