Move to Google Cloud with speed and confidence using Kyndryl Cloud Migration Services

Moving to Google Cloud offers many benefits for companies, including improved efficiency and performance, greater security, more flexibility, slashed IT costs, and the potential for innovation and developing new capabilities.

However, moving a company’s IT also moves its business. For a successful journey, a company must rely on a proven migration method, planning, and execution by an experienced team of skilled professionals. As a Google Cloud partner, Kyndryl provides skills and experience to successfully move your applications and workloads to the Google Cloud Platform.

Kyndryl has developed a proven and scalable migration methodology, and has used it to complete hundreds of thousands of migrations successfully. When migrating to Google Cloud, Kyndryl’s migration methodology is aligned with the Google Cloud Architecture Framework.

Where are you in your cloud transformation journey? Do any of these use cases resonate?

• Rip and Replace: Rewriting existing application to run under Google Cloud. There usually needs to be strong motivation to embark on this journey but, the benefit would be to take full advantage of everything the cloud has to offer.
• Improve and Move: Make improvements or modernize an existing on-premises application, for example, containerization, standard OS images, updated developer pipeline, or database changes while migrating it to cloud.
• Lift and shift: Lifting to the cloud provides access to easy machine scaling, managed instance groups, and other cloud native features without having to rewrite the application. This typically allows for much better rightsizing and significant cost savings.

Kyndryl leverages Google Cloud’s migration strategy of four distinct phases: Discover & Assess, Plan & Foundation, Migration, and Optimize.

Discover & Assess

First, let’s discuss the Discover and Assess phase. Discovery is used to understand details of a on-premises environment to define a migration plan. Clients usually prefer to use discovery tools with some form of familiarity. In cases where a tool is unfamiliar, our clients look to one which is simple to use, easy to deploy, non-disruptive, comprehensive, and secure.  

Kyndryl’s goal is to understand what our clients have, and what workloads can be migrated. We gather inventory data from the environment using an automated tool. Once the information is gathered, Kyndryl works with application owners and subject matter experts to help assign meaning. 

Kyndryl leverages tools such as Google’s StratoZone, Flexera Cloudscape, IBM’s Analytics for Logical Dependency Mapping (ALDM), Device 42, as well as Kyndryl specific tools and other utilities based on the needs of our clients.

Our assessment of the discovery data defines a post migration solution architecture, pricing, and phased migration approach. This phase is complex with many knowledge intense design tradeoffs which determine a final solution architecture and pricing.

Kyndryl’s discovery process creates the following:

• Automated inventory and dependency groupings
• Better understanding of application architectures
• Technical landscape requirements

Plan & Foundation

Second, the Plan and Foundation phase is designed to plan and build the foundation, or landing zone, within the Google Cloud Platform. Kyndryl understands planning is not monolithic or that planning is necessarily linear. Kyndryl partners with Google Cloud to assist our clients through the construction of detail migration wave planning as well as the grouping of workloads into waves. 

Kyndryl takes advantage of Google Cloud’s Cloud Foundations toolkit which offers a set of Deployment Manager templates and supporting utilities which can be used to build Google Cloud Platform infrastructure. Leveraging Cloud Foundations significantly reduces the amount time to deploy the landing zone.

Kyndryl understands the need to build a strong foundation within a Google Cloud Platform deployment and those ‘pillars’ are defined as follows:

• Identity and Access: Determines how individuals authenticate to Google Cloud Platform, and what resources those individuals are authorized to access. 
• Networking: Keeping a simple network topology can be the best way to ensure a manageable, reliable, and well-understood architecture
• Instrumentation: Real time system metrics which provides visibility into performance, availability, and health over the infrastructure with Cloud Monitoring
• Cost Control: Use of Cost Management to provide cost analysis and rightsizing notifications to lower costs by way of Google’s Recommendations API

Migration

Next, during the Migration phase, our Kyndryl engineers utilize the migration plan for the first set of migration waves which were established during the previous phase and work closely with our stakeholders to detail what workloads would migrate into Google Cloud. Kyndryl uses Google’s Migrate for Compute Engine and configures automation for fast, efficient migrations.

Our Kyndryl Cloud Migration and Modernization Architects and Engineers, working in partnership with Google Cloud, adhere to the Migrate for Compute Engine workflow for a seamless migration effort. An overview of the workflow is detailed below:

• Model: Determine VM relationships and dependencies in the runbook
• Test: Run application clones in the cloud, without affecting production data/uptime
• Run-in-cloud: Move workload to the cloud, operational in minutes, while data is accessed from cache
• Migrate storage: Continuous production operations while data transfers in the background
• Prepare to detach: Write data from cache to native Google Cloud Platform disks
• Detach: Detach from Migrate from Compute Engine and run natively in cloud
• Move back: Ability to roll back or revert to a source workload during the ‘migrate storage’ or ‘prepare to detach’ workflow stage

The Kyndryl team continuously reviews and improves upon the migration sprint by focusing on results and successes. 

Key activities of the migration phase include:

• Create waves and runbooks
• Define configurations for special applications
• Migrate waves of applications
• Improvements to the migration sprint process

Optimize

Last, with the workloads migrated into Google Cloud, Kyndryl continues to work with its clients to take full advantage of the Google Cloud Platform. 

The use of Managed Instance Groups (MIG) provides a well-balanced approach for scaling optimization and release management. Our clients can benefit from Google Compute Engine, which can provide zero-downtime updates to instance groups, the ability to support testing, and rollback.

A common way to optimize applications within the Google Cloud Platform is to replace VM-based parts of the architecture with managed services. 

Kyndryl works closely with its clients to leverage these managed services:

• MySQL -> CloudSQL, Spanner
• HBase -> BigTable
• Kafka -> Cloud Pub/Sub
• Hadoop/Spark -> Dataproc

The benefits our clients realize are:

• Reduced administrative overhead
• Better availability and scalability
• Increased functionality

With Migrate for Anthos and GKE, Kyndryl can modernize and convert our clients’ applications into containers. The Google Cloud Platform offers automatic updates which keeps the applications secure and can run more applications on a given host for better resource utilization.

Google Cloud Tools

Kyndryl is knowledgeable and proficient in all the migration tools and services available to complete the migration. Our teams regularly receive Google Cloud training to keep up to date on the latest Google Cloud technologies.

Some of the Google Cloud tools we utilize include on migration to the Cloud journey:

• StratoZone
• Migrate for Compute Engine
• Database Migration Service
• Migrate for Anthos and GKE

However, Kyndryl is not limited only to the Google Cloud toolset. Depending on workload and target platform, we have additional options to meet requirements, such as:

• CloudScape
• Carbonite Migrate
• Zerto
• VMWare HCX
• Oracle Data Guard
• SQL Server Always On
• DB2 HADR

In conclusion, Kyndryl’s world-class migration and modernization approach combines Google’s migration process of Assess, Plan, Deploy, and Optimize with industry-leading experience, skills, and techniques. Using these Google Cloud tools, the Kyndryl team can complete simple and/or complex migrations for any number of workloads.

Find more information on Kyndryl cloud services for Google Cloud here: https://www.kyndryl.com/alliances/google-cloud

Authors: 

Michael Mitchell, Cloud Migration & Modernization Architect, Kyndryl

Andre Dantas, Cloud Migration & Modernization, Technical Solution Architect, Kyndryl