Context aware workload management using fingerprints

Registered by tmetsch on 2016-04-06

This blueprint has been superseded. See the newer blueprint "Workload Fingerprint learning modules" for updated plans.

Understanding application behavior has been a key source of insight for striking a balance between reducing operational cost while improving the workload performance. Contentiously observing the workloads in the data center allows for getting a good understanding of how workloads performs in a given data center/context.

By fingerprinting especially the network and the flows between VMs we want to be able to make better placement and re-balancing decisions. The fingerprints including information of how entities are connected can hence be analyzed and resulting models/heuristics describing e.g. better deployment options (based on e.g. hardware features) or migrations suggestions etc. The heuristics can hence influence the strategy for doing optimal service delivery with a special focus on network.

Blueprint information

Status:
Complete
Approver:
Antoine Cabot
Priority:
Medium
Drafter:
tmetsch
Direction:
Approved
Assignee:
None
Definition:
Superseded
Series goal:
None
Implementation:
Unknown
Milestone target:
None
Completed by
Antoine Cabot on 2017-02-22

Related branches

Sprints

Whiteboard

(?)

Work Items

Dependency tree

* Blueprints in grey have been implemented.

This blueprint contains Public information 
Everyone can see this information.