INSIGHTS
Rethinking how enterprise
information is managed.
Perspectives on Data Lifecycle Management, Active Archive, information preservation, optical storage, AI-era data strategy, and long-term enterprise information governance.
LATEST THINKING
Better questions lead
to better data strategy.
Why do we manage data based on its birthday?
Age is easy to measure. Value, risk, retention, security, and future usefulness are what should determine lifecycle decisions.
Why optical storage matters again in the AI era.
AI increases both the value and vulnerability of retained information, making durable and physically isolated preservation increasingly important.
Technology changes. Information persists.
Applications, infrastructure, and media change. Information context, provenance, relationships, and governance must persist.
Compliance Retention: The Right Information, for the Right Time
Retention is a policy decision. Storage should execute the policy—without keeping every retained record on infrastructure built for active workloads.
Public Records That Outlive the System That Created Them
A deed recorded today may be requested in fifty years. The record has to survive every system refresh in between, and still be found on deadline.
When the Recovery Copy Has to Be Out of Reach
Defense programs and grid operations both depend on information that must survive an attack and outlast the hardware it lives on. Isolation helps, but only inside a disciplined security boundary.
Preserve the Data Behind the Model
When someone asks what a model was trained on, the answer should not depend on whether the data happened to survive the last storage cleanup.
Stop Powering Data Nobody Is Reading
Most stored data is rarely read, yet it sits on infrastructure that draws power around the clock and must be rebuilt every few years. Energy reporting is making that cost visible.
Media Archives: Find and Reuse What You Already Own
A clip may exist in the archive and still be unusable: the master is hard to locate, its provenance is unclear, or the license no longer covers the planned reuse. Preservation starts with the package and the workflow, not just the storage medium.
Remote Offices: Recovery Without the WAN
When a branch link is unreliable, backups that depend on routine transfers to headquarters can fall behind. A useful local recovery plan needs more than a local copy: it needs an outage-ready manifest, a clear catalog authority, and a tested restore path.
Enterprise Backup: Test the Copy, Not the Job
A green backup-job status does not prove that a business can recover. Define exactly which data gets a second copy, then test its integrity and the complete path back into service.
Offline Preservation: Isolate and Test the Restore
An offline copy can be beyond ordinary network reach and still fail when responders cannot locate, mount, verify, or restore it. Isolation matters; so does practicing the way back.
Scientific Data Must Outlive the Project
A grant can end while its data becomes more valuable. Preserve the original observations and the context future researchers need to validate, compare, and responsibly reuse them.
Long-Term Digital Retention with ELS100
Preserve completed projects, records, research, and intellectual property on a compact optical library after routine access declines—while keeping records governance, access controls, and migration planning in place.
Keep Imaging Records Beyond the PACS Refresh Cycle
A PACS storage decision is also a clinical retrieval and records-governance decision. Separate the performance tier from preservation without losing the path back to a prior study.
Ransomware Recovery for Credit Unions: The Offline Copy
Network-connected backups may be exposed during a ransomware incident. A prepared, verified offline copy can add another recovery path, but only as part of a rehearsed recovery program.
Performance and preservation should work together.
Important information should remain available on the right performance tier while being preserved on durable optical storage.
The Evidence Must Outlive the Storage System
Servers, disk arrays, and applications change. Evidence still has to be identified, verified, retrievable, and understandable years later.
A file turning 90 days old isn't a business event.
Infrastructure-driven lifecycle rules often fail to reflect the actual business state of information.
Old data isn't necessarily cold data.
Information can become relevant again as business, regulatory, analytical, and AI requirements change.
SAVARTUS PERSPECTIVE
Information should move
when its requirements change.
Not because a clock expired. Not because a file reached a certain age. Because the information itself changed.