Migrating to the cloud is smart but it’s not always easy. It can be a complicated process that requires a substantial investment of time and resources. Yet the benefits make the journey well worthwhile. Many of those who have completed the process confirm that they are realizing improved read performance and reliability as well as reduced stress on staff and lower total cost of ownership.
If your company is gearing up to make the big move, you’re likely asking: How will our applications and databases perform in the cloud? How can we improve the predictability of that performance and the resources it will consume? Is a loss of control inevitable? How can we keep our key stakeholders informed and in synch throughout the migration process?
The stakes are high as you move into the brave new world of cloud hosting and it’s easy to feel intimidated but the general issue is really one you’ve probably faced before – how to move workload from one data platform to another. To understand what a successful migration looks like is to understand what success means for your workload. We can break this down into how workload performs against how resources are used to execute that workload.
Before you begin the migration, you should conduct a baseline inventory of your current on-premise architecture and determine the scope of your migration. When you have a full grasp of your applications, your hosts, and their architecture, you reduce the possibility of missing dependencies during your migration. Start by surveying all your databases and stack ranking them in terms of usage and potential risk. This will allow you to take a phased approach, starting with the database with the lowest usage and lower risk in case of failure. As you migrate each successive database – learning from any mistakes and confirming best practices – your team will gain expertise and confidence that will be helpful when the time comes to migrate the databases that are more heavily used and pose a greater risk.
To start the migration process itself, the first thing you need is visibility. Planning, managing and validating the success of your workload migration to the cloud will require you to monitor the performance of your applications and databases every step of the way. You must be able to quickly, easily track:
- Infrastructure workload including CPU, Memory, Disk and Network. Each can be considered in the context of utilization, understanding that each has a cost and, as you approach saturation, performance regressions may be introduced.
- Query workload requires a deeper dive. You need high-granularity insights into query behavior and how that workload impacts database internals (contention for concurrent running queries) as well as system resources such as CPU, Memory and IO.
Before, during and after you migrate to a cloud environment, you must be able to see and understand your workload behavior in terms of Throughput, Concurrency, Latency and Errors. Visibility here means that you can detect changes in any one of those metrics and dissect the workload to the point where you can view the sets of tasks or hosts which are diverging from the expected norm.
For years, industry leaders have gotten the complete visibility that they need in locally hosted environments by using two complementary SaaS platforms. Gaining broader visibility is especially important for companies that want to go beyond database-only migration and move everything to a cloud environment. If your team is planning to lift and shift your entire tech stack to the cloud, you’ll need to know how both databases and applications are behaving with:
- Database Performance Monitoring (DPM) from VividCortex offers quick, easy drill-down views into everything the database does, from query execution to database internals and system resources. Lightweight agents in flexible configurations deliver in-depth views of query behavior and database workload in 1-second granularity across thousands of servers in any environment – cloud, local or hybrid – with virtually no overhead.
- Application Performance Monitoring (APM) provides a complete view of your apps and operating environment from a single screen. Auto-instrumentation shows what’s happening in any application environment. Quickly find root causes and fix issues fast, thanks to in-depth transaction details that show exact method calls with line numbers, including external dependencies for apps of any size and complexity.
Now you can use that same winning DPM-APM combination as you migrate to a cloud hosting environment such AWS Cloud. For nine years in a row, Gartner Research rated AWS as the leader in cloud Infrastructure as a Service (IaaS) with the highest score in both axes of measurement, Ability to Execute and Completeness of Vision1. VividCortex is a member of the AWS Partner Network (APN) and is ready to work closely with your team each step of the way throughout your migration process. For example, we can provide the multi-dimensional monitoring that you’ll need to:
- Evaluate traffic patterns in your locally hosted databases and analyze how that traffic would be handled in the AWS Cloud instances that you’ve chosen
- Replicate data from the first locally hosted database to the AWS Cloud instance and gradually turn up read transactions to while closely monitoring performance
- Validate your results to date to see if you’re achieving the level of predictability that you need to proceed confidently with your migration
- Begin writing directly to your first AWS Cloud instance and replicating back to your locally hosted database while measuring performance on both sides.
- Continue monitoring performance once the first instance is in production... and repeat this process for each of your databases.
- Establish and verify everyday monitoring capabilities to ensure continued strong performance in your new cloud environment and position you to scale up as you grow.
These are just a few of the steps in a typical cloud migration. While we won’t dive deeper into the details here, we hope you can see that having full visibility into the performance of your applications and databases is absolutely essential to navigate this journey as quickly and smoothly as possible.
To learn more about how VividCortex and APM can work together to help you migrate to AWS Cloud, please come see us at AWS re:Invent in Las Vegas, NV. You can stop by the VividCortex booth #3611 in the AWS re:Invent Expo (Venetian Hall) anytime. VividCortex will also be doing a special Lightning Talk on this topic:
Date/Time: 10:30 AM, Tuesday December 3
Location: New Relic booth #401 (Venetian)
We look forward to seeing you there! For more information, visit www.vividcortex.com.