Industry Trends

Hybrid Cloud Disaster Recovery and Backup Development Trends

Hybrid Cloud Disaster Recovery and Backup Development Trends

With the rapid development of the cloud computing industry and ever-increasing customer needs, the IT infrastructure at organizations of all sizes are placing more and more services into a cloud framework. This move presents enterprises with some daunting challenges in ensuring service continuity.


Enterprises are choosing cloud services because they are cheaper, able to eliminate environment bottlenecks, and able to fully gear into ever-changing developments to suit the growth strategies of the enterprises. For service systems, security and stability are always the top priority. Traditional geo-redundant DR models often comprising two sites and three data centers in a single cloud ensure rapid restoration for services during downed events at one or even multiple locations. However, these types of models cannot be used to handle service interruptions caused by system errors. Therefore, the multi-cloud backup and on-cloud DR models will become the focus in the cloud computing context to ensure zero service interruptions.


Enterprises use multi-cloud solutions for various reasons. The data requirements as well as data protection and management strategies of an enterprise determine whether multi-cloud and cross-cloud DR capabilities are required. For example, if the enterprise data management strategy emphasizes multiple backups, then the multi-cloud strategy is the correct choice. According to research from Enterprise Management Associates (EMA), 61% of the enterprises with services running in clouds use at least two clouds.

Business Challenges

  • High Costs in DR Center Builds and O&M

    To build a traditional DR center, enterprises need to build or rent equipment rooms and invest lots of manpower in maintenance and testing. Accomplishing a full DR mode incurs high CAPEX and OPEX as the arms of both or multiple centers must reach the same scale and achieve near real-time transmission.

    To build a traditional DR center, enterprises need to build or rent equipment rooms and invest lots of manpower in maintenance and testing. Accomplishing a full DR mode incurs high CAPEX and OPEX as the arms of both or multiple centers must reach the same scale and achieve near real-time transmission.

  • Many Restrictions on DR Drills

    If a DR drill is performed in a traditional DR center, equipment rooms and cables need to be scheduled, and some cold standby devices need to be powered on, which causes a high fault rate. Switchovers may delay or even fail at the critical moment in spite of the drills, leaving the enterprise vulnerable with weak service recovery capabilities.

    If a DR drill is performed in a traditional DR center, equipment rooms and cables need to be scheduled, and some cold standby devices need to be powered on, which causes a high fault rate. Switchovers may delay or even fail at the critical moment in spite of the drills, leaving the enterprise vulnerable with weak service recovery capabilities.

  • Complicated Data Synchronization Policies

    The synchronization policies for hosts, databases, and storage devices are complicated and restricted by the transmission distance and data volume. In addition, the data encryption/decryption mechanism complicates deployments and maintenance. As a result, some data becomes invalid, or the system cannot meet the designed RPO.

    The synchronization policies for hosts, databases, and storage devices are complicated and restricted by the transmission distance and data volume. In addition, the data encryption/decryption mechanism complicates deployments and maintenance. As a result, some data becomes invalid, or the system cannot meet the designed RPO.

  • Architecture Difficult to Implement

    Core services require multi-center architectures to ensure service continuity, requiring large amounts of investment and extensive consideration in design as to how all the parts of the architecture will collaborate. Implementing this architecture with a traditional DR model equates to high costs and the benefits are limited.

    Core services require multi-center architectures to ensure service continuity, requiring large amounts of investment and extensive consideration in design as to how all the parts of the architecture will collaborate. Implementing this architecture with a traditional DR model equates to high costs and the benefits are limited.

Typical Scenarios

Introduction

Cost-Effective and Secure Enterprise Backup Solution

Benefits

  • Economical and Efficient Backup

    Storage resources are billed by usage. Upfront investments are reduced by 20% to 80%. Data is transmitted after deduplication and compression, saving you big on storage costs while shortening the backup window often by as much as 75%.

  • Secure and Reliable Data Transmission and Backup

    Data is encrypted throughout the entire process: at the source end, during transmission, and in OBS. Data durability reaches 99.99999%.

Related Services

Introduction

HA Public Cloud DR Center

Benefits

  • New IoT Connection Processing Module

  • Selects the optimal replication technology for the layers and other particulars in your profile:

    Application layer: multi-cloud architecture and multi-cloud connections switchover

    Virtualization layer: data synchronization supported by the ecosystem cooperation solution

    Database layer: replication software of mainstream databases, such as Oracle and MySQL

    Storage layer: asynchronous replication of Huawei storage arrays to HUAWEI CLOUD Dedicated Enterprise Storage Service (DESS)

Introduction

Intra-City DR Capability with RPO = 0

Benefits

  • Enterprise-level DR-as-a-Service

  • High reliability

    Provides VM-level DR protection across AZs (RPO = 0) and hour-level RTO to meet tier-5 DR standards. Quickly recover services if the production site goes down, minimizing service interruptions and helping eliminate data loss.

  • Low cost

    Reduces the enterprise DR TCO and simplifies the DR process.

Recommended Configurations

VM On-Cloud DR
Service VMs 100
Data Quantity 50 TB
Product
Specifications
E-waybill recognition
10 million/year package
--
--
Certificate recognition
100,000/year package
--
--
Image recognition
200,000/year package
--
--
ECS
4U8G | 4 units
--
--
OBS
10 TB/year
--
--
General-purpose enhanced MySQL
2U4G | 2 units
--
--
App On-Cloud DR
Service VMs 100
Data Quantity 50 TB
Product
Specifications
ECS (DR VM)
Same as computing resources of production VMs| 100
--
--
ECS (DR mgmt. system)
8 vCPUs | 16 GB memory | 500 GB Ultra-High I/O disk | 12
--
--
EVS
Same as storage resources of production VMs| 50 TB
--
--
Direct Connect
200 Mbit/s
--
--
VPN (optional)
Pay per usage (time-based)
--
--
Anti-DDoS
Free
--
--
WAF
Pay per bandwidth unit and usage (time-based)
--
--
App On-Cloud DR (DeC)
Service VMs 200
Data Quantity 100 TB
Product
Specifications
BMS
Determined by physical hosts in the production center| 8
--
--
DCC
Determined by the CPU/memory ratio of the production center | 12
--
--
DSS
Ultra High I/O | 1
--
--
DSS
High I/O| 1
--
--
Direct Connect (optional)
400 Mbit/s
--
--
VPN (optional)
Pay per usage (time-based)
--
--
Anti-DDoS
Free
--
--
WAF
Pay per bandwidth unit and usage (time-based)
--
--
On-Cloud DR
Service Hosts 100
Data Quantity 50 TB
Product
Specifications
SDRS
Determined by service requirements
--
--
ECS (DR VM)
Same as computing resources of production VMs | 100
--
--
EVS
Same as storage resources of production VMs | 50 TB
--
--
VPN (optional)
Pay per usage (time-based)
--
--
Anti-DDoS
Free
--
--
WAF
Pay per bandwidth unit and usage (time-based)
--
--
On-Cloud DR (DeC)
Service Hosts 200
Data Quantity 100 TB
Product
Specifications
SDRS
Pay per number of DR VMs
--
--
BMS
Pay per number of BMSs
--
--
DCC
Determined by the CPU/memory ratio of the production center | 12
--
--
DSS
Ultra High I/O | 2
--
--
VPN (optional)
Pay per usage (time-based)
--
--
Anti-DDoS
Free
--
--
WAF
Pay per bandwidth unit and usage (time-based)
--
--
CBH
4 vCPUs | 8 GB memory | 400 GB hard disk | 2
--
--
VM
VM On-Cloud DR
Service VMs 100
Data Quantity 50 TB
E-waybill recognition
10 million/year package
Certificate recognition
100,000/year package
Image recognition
200,000/year package
ECS
4U8G | 4 units
OBS
10 TB/year
General-purpose enhanced MySQL
2U4G | 2 units
App
App On-Cloud DR
Service VMs 100
Data Quantity 50 TB
ECS (DR VM)
Same as computing resources of production VMs| 100
ECS (DR mgmt. system)
8 vCPUs | 16 GB memory | 500 GB Ultra-High I/O disk | 12
EVS
Same as storage resources of production VMs| 50 TB
Direct Connect
200 Mbit/s
VPN (optional)
Pay per usage (time-based)
Anti-DDoS
Free
WAF
Pay per bandwidth unit and usage (time-based)
App (DeC)
App On-Cloud DR (DeC)
Service VMs 200
Data Quantity 100 TB
BMS
Determined by physical hosts in the production center| 8
DCC
Determined by the CPU/memory ratio of the production center | 12
DSS
Ultra High I/O | 1
DSS
High I/O| 1
Direct Connect (optional)
400 Mbit/s
VPN (optional)
Pay per usage (time-based)
Anti-DDoS
Free
WAF
Pay per bandwidth unit and usage (time-based)
On
Service Hosts 100
Data Quantity 50 TB
SDRS
Determined by service requirements
ECS (DR VM)
Same as computing resources of production VMs | 100
EVS
Same as storage resources of production VMs | 50 TB
VPN (optional)
Pay per usage (time-based)
Anti-DDoS
Free
WAF
Pay per bandwidth unit and usage (time-based)
On (DeC)
Service Hosts 200
Data Quantity 100 TB
SDRS
Pay per number of DR VMs
BMS
Pay per number of BMSs
DCC
Determined by the CPU/memory ratio of the production center | 12
DSS
Ultra High I/O | 2
VPN (optional)
Pay per usage (time-based)
Anti-DDoS
Free
WAF
Pay per bandwidth unit and usage (time-based)
CBH
4 vCPUs | 8 GB memory | 400 GB hard disk | 2

Partners