8
CHT Project Progress Report 10/07 Simon

CHT Project Progress Report 10/07 Simon. CHT Project Develop a resource management scheduling algorithm for CHT datacenter. ◦ Two types of jobs, interactive/latency-

Embed Size (px)

Citation preview

Page 1: CHT Project Progress Report 10/07 Simon. CHT Project Develop a resource management scheduling algorithm for CHT datacenter. ◦ Two types of jobs, interactive/latency-

CHT ProjectProgress Report10/07Simon

Page 2: CHT Project Progress Report 10/07 Simon. CHT Project Develop a resource management scheduling algorithm for CHT datacenter. ◦ Two types of jobs, interactive/latency-

CHT ProjectDevelop a resource management

scheduling algorithm for CHT datacenter.◦Two types of jobs, interactive/latency-sensitive and batch/computation-intensive.

◦Minimize SLA violation with limited resources. # of servers

◦Based on Red Hat Openshift

Page 3: CHT Project Progress Report 10/07 Simon. CHT Project Develop a resource management scheduling algorithm for CHT datacenter. ◦ Two types of jobs, interactive/latency-

Current PlanTwo components

◦Scheduler Deploy container/pod to server Inside Kubernetes

◦Rule Engine Decide the number of container/pod for

each service. Output the results in JSON

Input of the scheduler

Page 4: CHT Project Progress Report 10/07 Simon. CHT Project Develop a resource management scheduling algorithm for CHT datacenter. ◦ Two types of jobs, interactive/latency-

SchedulerImplement a new scheduler as

plug-in and replace the original one.Use the original scheduler, but change the policy.◦“Provide a JSON file that specifies

the predicates and priority functions to configure the scheduler”.

◦Change the weight of the (built-in) priority function to meet our score function.

Page 5: CHT Project Progress Report 10/07 Simon. CHT Project Develop a resource management scheduling algorithm for CHT datacenter. ◦ Two types of jobs, interactive/latency-

Rule EngineDecide the number of

container/pod for each service according to the monitoring data.

Add/adjust rules to make better decisions.

However,◦Rule Engine is another Red Hat

product.◦Creating and adjusting new rules

requires experiences.

Page 6: CHT Project Progress Report 10/07 Simon. CHT Project Develop a resource management scheduling algorithm for CHT datacenter. ◦ Two types of jobs, interactive/latency-

Alternative WayBuild our own Resource Allocator.

◦Decides the number of container/pod for each service according to the monitoring data.

◦Basic rules Rules about the critical resources such as

CPU, memory …etc.

Page 7: CHT Project Progress Report 10/07 Simon. CHT Project Develop a resource management scheduling algorithm for CHT datacenter. ◦ Two types of jobs, interactive/latency-

Possible future extensionApply machine learning to

add/adjust the rules.(CHT)Apply machine learning to

minimize the size of a container/pod.(Prof. Lin)

Page 8: CHT Project Progress Report 10/07 Simon. CHT Project Develop a resource management scheduling algorithm for CHT datacenter. ◦ Two types of jobs, interactive/latency-

NextStudy the current predicates and

priority functions inside Kubernetes.

Keep working.