Data Management Glossary
Observer (Komprise Observer)
The Komprise Observer is a virtual appliance running at the customer site that analyzes data across NAS silos, moves and replicates data by data management policy, and provides transparent file access to data that’s stored in the cloud.
Komprise handles any scale of data using an elastic scale-out architecture that has no central bottlenecks. Start with a Komprise Observer virtual machine and simply scale by adding more. Komprise automatically load balances and creates fault-tolerance across the elastic grid. There are no central databases or bottlenecks to limit scalability. Komprise scales to handle billions of files and tens of petabytes of data with a lightweight distributed architecture. There is no single point of failure. Komprise uses fault-tolerant architecture principles to create a resilient grid with Observer virtual instances that provide failover to one another in an active-active configuration without requiring any dedicated infrastructure.
What is an Observer in the Komprise architecture?
An Observer is a node in Komprise’s distributed, scale-out architecture. It’s part of a “grid of workers” that perform analysis, metadata gathering, and other tasks by bringing compute to the data rather than moving all the data centrally. Observers are stateless; if one fails, another can pick up and resume its job.
What roles do Komprise Observers play in data analysis and management?
Komprise Observers are responsible for:
- Discovering and indexing files and objects across storage silos into the Komprise Global Metadatabase (KMDB), enabling visibility across your data estate.
- Executing metadata and content scans (for example for tagging, classification) as part of Smart Data Workflows.
- Assisting in data migration, tiering, and other data operations by distributing the work load across multiple Observers to scale performance.
Are Komprise Observers stateful? What happens if one fails?
No. Observers are stateless. If an Observer fails, another Observer in the grid takes over. The work resumes from where the failed one left off.
How do Komprise Observers scale, and how many are needed?
Observers scale elastically. The system allows control over:
- The number of Observers in the grid
- The compute & memory size of each Observer node
- The number of threads per operation
This means you can increase or decrease Observers depending on the size of your data estate, performance demands, etc. The load is broken into tasks that are distributed among Observers.
How do Komprise Observers contribute to cost savings and operational benefits?
Komprise Observers enable:
- Fast discovery and classification to identify cold, unused data and allow tiering or migration, lowering storage, backup, and disaster recovery costs.
- More efficient migrations via parallelization (many Observers working together) to speed up data movement, reduce risk, and improve planning.
- Transparent tiering without vendor lock-in or “rehydration penalties.” Observers help implement policies and workflows to move cold data without affecting user/application access.
How do Komprise Observers relate to security, especially regarding ransomware exposure and sensitive data handling?
Komprise Observers are part of the mechanism that:
- Identifies inactive / cold unstructured data that might be a vulnerability, enabling it to be tiered to immutable storage or protected snapshots.
- Supports metadata enrichment and scanning (e.g. for PII) via Smart Data Workflows. This allows tagging or excluding sensitive data from less secure paths (e.g. AI ingestion).