Cloud Migration Manager Technology

Sanovi Cloud Migration Manager Technology Overview

 
Migration to cloud – Changing Landscape
contact-expertAs medium to large organizations are migrating their workloads at scale to cloud, there is a need for a new approach towards migration into cloud. Unlike moving the Dev/Test environment to the cloud, enterprises’ business applications are complex, scale of migration is much larger than before and with stringent SLAs.
 
Large Scale migration into Cloud needs a new Paradigm
Traditional approaches to migrating workloads to cloud are focused on migrating individual virtual machines and relies on experts to perform the analysis, planning and coordination to move the complete application. In addition, there is a lack of visibility and difficulty in tracking these migration projects due to the large scale. These expert-driven operations in a large scale migration project often leads to manual errors and inconsistency. Public cloud providers have already adopted a life-cycle approach towards migration to cloud, and are training their global partners to skill-up on this approach, for better customer experience and completeness of migration. All these factors suggest that there is a need for a new paradigm for large scale migration to cloud – a life-cycle based automation approach.
 
Challenges in Workload Migration to Cloud:
System Integrators and Managed Service Providers are facing huge challenges in large scale workload migrations to clouds due to the technical complexities involved in discovery, planning and testing. Enterprise applications are complex with multi-tiered architecture and have to be migrated with very little downtime. Moreover, the benefits of cloud can be achieved post migration only through rightful planning, architecture transformation and configuration of application infrastructure during cloud migration. Few of the challenges are:

  • Multiple tools are being used to achieve end-to-end workload migration – Separate tools for discovery, resource planning and migration, leading to sub-optimal resource allocation, higher cost and complexity
  • The focus is on individual virtual machines, rather than application – This approach leads to manual errors, leading to migration delays, inconsistent and erratic application behavior
  • There is little or no visibility on Plans & Status – This leads to gaps in monitoring, missed schedules and non-availability of reports

 
Sanovi Cloud Migration Manager Approach
Sanovi Cloud Migration Manager addresses the challenges in the traditional migration methodology by providing an application centric, lifecycle based phased approach – from discovery & planning to final cutover, with automation and visibility across all phases. This reduces the manual dependency, inconsistency and errors and also provides clear visibility to track the progress of the migration.
 
Sanovi Cloud Migration Manager Architecture
Sanovi Cloud Migration Manager deployment architecture consists of Sanovi Cloud Migration Manager Server and Sanovi Replicator Appliance (SRA), which are installed in the On Premise data center which interfaces with the VMware vSphere infrastructure. A proxy migrator server and the SRA is deployed in the VPC of customer’s AWS account and interfaces with AWS.
 
Cloud Migration Manager Technology
 
The migrator server connects to vSphere for discovering the applications and capture the application’s performance characteristics like CPU, memory and disk I/O. Once the application’s characteristics is captured, the intelligent resource optimizer powered by Sanovi’s Application Defined Continuity platform analyzes the application and recommends the optimum resources that needs to the provisioned on the cloud based on the chosen migration plan. It also generates a Bill of Materials (BOM) which can be used to compute the costs of the resources in the cloud. These recommendations can be overridden manually for finer control.
 
Once the assessment and planning is finalized, the virtual machines of the application will be replicated using a secure channel to the cloud where they will reside waiting to be activated for testing. After the initial synchronization of the virtual machine is completed, incremental changes that continues to happen in on premise data center are replicated and synchronized to the cloud periodically.
 
When it is time for testing the migrated application on the cloud, instances of the migrated virtual machines are provisioned, network is created, security rules are attached and the whole stack is started up automatically without any manual intervention. The complete task of provisioning the cloud resources like compute, network, storage etc. are completely automated using Sanovi’s powerful workflow orchestrator and Recovery Action Library(RAL). Business users can then test the application and verify it by running their tests on the cloud and IT can monitor the performance. Based on the tests, any tuning to the resources if necessary can be done by updating the migration plan. This testing will be performed on the cloud without impacting the production application that is running in on premise.
 
To cutover the application from on premise to cloud, the application is shut down and the final incremental data is securely transmitted to the cloud and the instances are brought up and open for live transactions.