Bayllocator: A Proactive System to Predict Server Utilization and Dynamically Allocate Memory Resources Using Bayesian Networks and Ballooning.
Vangelis TasoulasHårek HaugerudKyrre M. BegnumPublished in: LISA (2012)
Keyphrases
- bayesian networks
- resource allocation
- allocate resources
- network bandwidth
- state information
- resource consumption
- client server
- learning bayesian networks
- computing resources
- probabilistic inference
- resource management
- resource utilization
- web server
- secondary storage
- structure learning
- database
- graphical models
- probabilistic reasoning
- storage devices
- memory space
- dynamically created
- memory usage
- conditional independence
- context aware
- resource constraints
- limited resources
- incomplete data
- information resources
- application level
- expert knowledge
- main memory
- inference in bayesian networks
- website
- neural network