You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When migrating from Elasticsearch to OpenSearch, users need a reliable way to compare and validate behavioral and performance differences in real-time. Inflight validations should be performed at the replayer level to ensure that any discrepancies between source and target clusters are quickly identified. Currently, while Migration Assistant provides support for comparing status codes and drilling down into individual requests, this feature would focus on a more comprehensive validation to surface behavioral differences between the clusters during the migration.
What solution would you like?
Implement a feature within OpenSearch Migrations that performs real-time inflight validation at the replayer level. This feature should compare key metrics and behaviors between the source (Elasticsearch) and target (OpenSearch) clusters during the migration process. It should provide automated alerts and dashboards highlighting discrepancies such as differences in query responses, status codes, latency, and resource utilization between the two clusters.
What alternatives have you considered?
Offline post-migration validation – This approach is insufficient for catching real-time discrepancies and can lead to delayed identification and resolution of critical issues.
No comparison validation – This is a major obstacle for users, as they often need confirmation that the behavior of the OpenSearch target cluster aligns with their original Elasticsearch cluster.
Do you have any additional context?
This feature would offer users the confidence to execute large-scale migrations by providing clear insights into behavioral consistency between Elasticsearch and OpenSearch. By focusing on real-time validations at the replayer level, this feature can help users address migration risks proactively. Integration with existing monitoring tools like AWS CloudWatch or OpenSearch Dashboards would be ideal to offer a unified view of migration metrics and alerts.
The text was updated successfully, but these errors were encountered:
sumobrian
changed the title
[FEATURE] Implement In-Flight Behavioral Validation at Replayer Level for Elasticsearch to OpenSearch Migrations
In-Flight Behavioral and Performance Validation
Oct 23, 2024
Is your feature request related to a problem?
When migrating from Elasticsearch to OpenSearch, users need a reliable way to compare and validate behavioral and performance differences in real-time. Inflight validations should be performed at the replayer level to ensure that any discrepancies between source and target clusters are quickly identified. Currently, while Migration Assistant provides support for comparing status codes and drilling down into individual requests, this feature would focus on a more comprehensive validation to surface behavioral differences between the clusters during the migration.
What solution would you like?
Implement a feature within OpenSearch Migrations that performs real-time inflight validation at the replayer level. This feature should compare key metrics and behaviors between the source (Elasticsearch) and target (OpenSearch) clusters during the migration process. It should provide automated alerts and dashboards highlighting discrepancies such as differences in query responses, status codes, latency, and resource utilization between the two clusters.
What alternatives have you considered?
Do you have any additional context?
This feature would offer users the confidence to execute large-scale migrations by providing clear insights into behavioral consistency between Elasticsearch and OpenSearch. By focusing on real-time validations at the replayer level, this feature can help users address migration risks proactively. Integration with existing monitoring tools like AWS CloudWatch or OpenSearch Dashboards would be ideal to offer a unified view of migration metrics and alerts.
The text was updated successfully, but these errors were encountered: