Everybody is doing it: migrating data to the cloud. Whether you call your project “data re-platform”, “data modernization”, “cloud data warehouse adoption,” “cloud data migration,” or any of the other hot buzz phrases, the idea is the same: move data from multiple on-prem and SaaS-based systems and data storage into a centralized location where you can use that data to spend less money or make more money. In other words, the goal of consolidating into a cloud data warehouse is, at its core, to save company money on cost of goods sold or grow company revenue streams.
What’s not to love? But one of the tradeoffs is that, as you go through the process, you end up losing the full visibility and control you had over data in your on-prem systems.
We see this concern about visibility and control come up over and over, at every stage of the cloud data migration journey – from CIOS, CISOs and CDOs who are accountable for making sure data stays secure, to the leaders who must address this in the overall project within the given budget, until it finally lands on the Data Engineers and DBAs who must decide how they’ll fix this and pick the actual solution.
Keep Cloud Data Migration Projects on Track
Putting your cloud data migration project on rails out of the gate means you can go more quickly, more securely. Even if you’re going 100 MPH with data, you can be sure you’re not going to:
- Fly off the path
- Put your data or company at risk
- Damage your reputation
How do you do that? No matter if you’re using an ETL or Snowpipe to transfer data, choosing a cloud-hosted BI (Business Intelligence) tool or an on-prem solution to run analytics, doing it all with visibility, control, and protection in place from the very moment you begin will mean that your continuous data path from on-prem to the cloud will be safe, speedy and secure.
And those pesky questions about visibility and control that pop up at each stage won’t slow you down or derail the project:
Selecting a Cloud Data Warehouse:
Whether it’s Snowflake, Amazon Redshift, Google, you’ll need to answer these to get your cloud data migration project off the ground…
- How do I give users access to the data?
- What level of logging do I have?
- How can I integrate this with my existing security stack (SSO, MFA, Splunk)?
- If data cannot be in clear text, is encryption okay or will I need tokenization?
Selecting a BI tool:
Again, whether you’re looking at Tableau, Qlik, PowerBI or other, similar operational questions will come up…
- Can I give different users using a shared service account varying levels of access?
- If I wanted to do tokenization, where in the stack is the best place to implement it? Should it be in the BI tool or elsewhere?
- How can I see what each user is doing through a single service account?
- How do I know what sensitive data is being accessed in the reports? How can I control that data?
Choosing an ETL:
When moving the actual data from data sources to the new cloud data warehouse, if these questions don’t come up, they should…
- When sensitive data is moving from on prem to the cloud, how can I protect it at rest?
- When I protect all my data, how can I keep it useful? How do I keep reports from breaking?
- When I move data into the cloud, how can I be sure it won’t be stolen?
Implementing a Data Catalog:
Data “brains”, like Collibra, OneTrust, and Allation, have all the information about users, data itself, data classifications and the policies that should be placed on user data access. What they don’t have is a way to operationalize that policy.
- Manually approving and applying access is a time sink. Is there anything to automate this?
- I have no way of controlling the admins of these systems. Do they have to have this much access to my data?
- I must apply masking policies manually with support tickets. Why?
If you can be prepared to answer these questions from the start, your cloud data migration project will go more smoothly, and the good news will travel back up the line very, very quickly to the executives who need to ensure that data visibility, control and protection are covered.
ALTR is here to help you solve the problem at a very low cost, with a low-friction implementation, and a short time to value. You can start with ALTR’s free plan today.