In parts 1 and 2 of this blog series, we introduced the key elements of a storage reclamation program and dove deeply into 5 key capabilities that would be needed. In this last blog, we pull all of the guidance together and wrap up the discussion.
Putting It All Together
With the five key capabilities addressed in the earlier blogs, any organization can be prepared to put together a comprehensive file cleanup workflow. While no two organizations are alike, a cleanup campaign should follow the same basic outline. The steps below outline an approach to clean up a single file server. Based on the size and scale of an organization, the approach can be adjusted to address multiple file servers simultaneously.
Step 1 – Initial Discovery and Planning
During this phase, an initial discovery is performed on all file servers within the organization. The expected outcomes of this discovery should be:
- A listing of all file servers by type (e.g. Windows, NetApp, SharePoint Online, Box, Dropbox, etc.)
- Amount of used/available storage on each system (broken out by volume)
- Number of shared folders on each system
- A detailed plan including the goal of the cleanup and a process flow for the approach
Step 2 – Deep Dive
With a plan formed and a general awareness of what file servers are out there, it is time to go deeper into the file servers and gather additional data that can be used to focus the cleanup campaign.
By running a full inventory of the files and folders on a file server, the following information should be captured:
- The number and the total size of files over a certain age threshold (e.g. all files that have not been modified/accessed in over 5 years)
- The number and the total size of files by file extension. This can be used to identify how much space is consumed by media files, office files, personal files, and database files.
- A listing of orphaned files where the file owner is no longer with the company or in the same role.
- A listing of all files with sensitive data and the classification level of all files on the system.
- Security of the files, including identifying any unsecured locations such as folder structures with open access to all users. These are at high risk for a data breach and should be addressed immediately.
Gathering this information can provide a good understanding of how much data may be cleaned up and where the highest risk data exists that may need immediate attention.
Step 3 – Monitor
To minimize the impact on end users, it is beneficial to monitor activity on the file server to understand access patterns. Doing so will allow any active files to be left alone, and only unused files will be archived. Last Modified and Last Accessed timestamps can be unreliable for various reasons, and activity monitoring provides the most actionable data prior to a cleanup.
In addition, knowing the most active users enables targeted engagements with those users prior to making any decision on which files will be archived.
Step 4 – Outreach
Once all of the information has been gathered as to what data is the best candidate for cleanup, the business owners and data users should get involved. The active users and managers should be notified of the cleanup campaign and provided details on what to do if their data is inadvertently moved or goes missing. Moreover, the owners should be surveyed and allowed to attest to the files that are being moved, as well as provide feedback on whether those files are needed or not. By knowing which files are sensitive and which files are actively used, a much more filtered list of files can be provided to the end user so they can focus only on the important ones.
Step 5 – Move
Now that feedback has been gathered, it is time to move the data to its archive location. Typically this will be a separate, low-cost file server. The files that have met the policies identified during planning should be moved.
During the move, it is important to maintain the folder hierarchy of the file on the archive system so it can easily be restored if needed. More importantly, stub files should be used to ensure users can still locate and access these files even if they are not located on the same file server. There are various approaches to stub files, but using file shortcuts is ideal.
Once the files are on the archive file server, they should be monitored for activity. If no users are following the stub files and accessing the archive files, they can be permanently deleted, but only at the end of any relevant retention policy. If users are actively using the files, they can be restored to primary storage.
Step 6 – Automate
To keep file systems clean, the newly defined policies should be automated and repeated. Data can regularly be archived if it meets the archive policies. Owners can periodically recertify the files that they own and make sure old and unnecessary files are deleted. This will ensure file servers are kept clean and only necessary files are managed going forward.
Performing file server cleanups provides numerous benefits for an organization. Not only does it provide better security and improved ease-of-use for end users, but it also offers significant savings through reducing storage and management costs. By following the steps in this paper, any organization can achieve file cleanups with minimal impact to end users.
If you’d like to learn more about our storage reclamation/file cleanup capability, check out our storage reclamation web page: https://www.stealthbits.com/storage-reclamation-solution and to see how much money you could be saving using a product like StealthAUDIT, check out our storage reclamation ROI calculator: https://www.stealthbits.com/storage-reclamation-calculator
As the VP of Product Marketing, Darin is responsible for product messaging and positioning as well as generating industry and market awareness for Stealthbits products. He is an experienced leader who has worked in software for over 21 years.
Prior to joining Stealthbits, he was VP of Marketing for Quorum and SecureAuth, and has held positions in product management & product marketing at Oracle, and Quest Software.