CHT Project
Progress Report
11/04 Simon
CHT Project
Develop a resource management sche duling algorithm for CHT datacente r.
◦Two types of jobs, interactive/latenc y-sensitive and batch/computation-int ensive.
◦Minimize SLA violation with limited r esources.
# of servers
◦Based on Red Hat Openshift
Current Plan
Two components
◦Scheduler
Deploy container/pod to server
Inside Kubernetes
◦Container Scaler
Decide the number of container/pod for eac h service.
Output the results in JSON
As an input of the scheduler
Container Scaler
Decide the number of container/pod for each service according to the monitoring data.
◦Ex: add one container if the average CPU utilization is higher than 85%.
Add/adjust rules to make better de
cisions.
Adjust According to Monito r
If the QoS of an application can b e measured from the monitor.
◦Add container according to the QoS.
◦Ex: if the average response time is ov er 80% of the SLA, add new container.
If not, need to “derive” from th e monitoring data.
◦Ex: if the CPU utilization over 75% an d the network I/O is larger than 8000b ps, add new container.
Improving the Rules
Apply data analysis techniques to improve the derivation.
◦SVM(Support Vector Machine)
Find the hyper-plane that separate the dat a points into “safe” and “violation”.
Add containers for those applications being clas sified into “violation”.
Find another hyper-plane that separate the data points into “safe” and “over-provi sion”.
Remove containers for those applications being c lassified into “over-provision”.
Implementation
Working on container scaler with b asic rules.
◦No data analysis techniques are invol ved in current state.
◦Will coordinate the data format with Prof. Lin’s group.
XML or JSON.
Next
