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

Slides:



Advertisements
Similar presentations
Starfish: A Self-tuning System for Big Data Analytics.
Advertisements

Energy-efficient Task Scheduling in Heterogeneous Environment 2013/10/25.
Multilayer Perceptrons 1. Overview  Recap of neural network theory  The multi-layered perceptron  Back-propagation  Introduction to training  Uses.
QoS Management l Many distributed applications have real- time requirements that coexist with tradition non-real-time applications. l Resources must be.
ProActive Task Manager Component for SEGL Parameter Sweeping Natalia Currle-Linde and Wasseim Alzouabi High Performance Computing Center Stuttgart (HLRS),
Two main requirements: 1. Implementation Inspection policies (scheduling algorithms) that will extand the current AutoSched software : Taking to account.
SWiM Panel on Engine Implementation Jennifer Widom.
5: CPU-Scheduling1 Jerry Breecher OPERATING SYSTEMS SCHEDULING.
MCTS Guide to Microsoft Windows Server 2008 Network Infrastructure Configuration Chapter 8 Introduction to Printers in a Windows Server 2008 Network.
What is adaptive web technology?  There is an increasingly large demand for software systems which are able to operate effectively in dynamic environments.
Milestone #1 – Project Design Chris Thomas CS 2310 – Software Engineering.
©Ian Sommerville 2004Software Engineering, 7th edition. Chapter 15 Slide 1 Real-time Systems 1.
Rensselaer Polytechnic Institute CSCI-4210 – Operating Systems David Goldschmidt, Ph.D.
Chapter 11: Artificial Intelligence
Connecting OurGrid & GridSAM A Short Overview. Content Goals OurGrid: architecture overview OurGrid: short overview GridSAM: short overview GridSAM: example.
OPERATING SYSTEMS CPU SCHEDULING.  Introduction to CPU scheduling Introduction to CPU scheduling  Dispatcher Dispatcher  Terms used in CPU scheduling.
EMBEDDED SYSTEMS 9 April 2013 William W. McMillan.
Cloud Computing Energy efficient cloud computing Keke Chen.
© 2005 IBM Corporation IBM Printing Systems IBM OUTPUT ENVIRONMENT IPPD How workflow techniques can by implemented using IPPD Simon Jones 14 th September.
Chapter 1 Introduction. Goal to learn about computers and programming to compile and run your first Java program to recognize compile-time and run-time.
Semantic Interoperability Berlin, 25 March 2008 Semantically Enhanced Resource Allocator Marc de Palol Jorge Ejarque, Iñigo Goiri, Ferran Julià, Jordi.
Cloud Resource Scheduling for Online and Batch Applications Kick-off meeting.
Module 4: Systems Development Chapter 12: (IS) Project Management.
MindMajix.com TIBCO BUSINESSWORKS Introducing TIBCO ActiveMatrix BusinessWorks.
Artificial Intelligence Techniques Multilayer Perceptrons.
Cloud Resource Scheduling for Online and Batch Applications Q1 review 9/3.
The Owner Share scheduler for a distributed system 2009 International Conference on Parallel Processing Workshops Reporter: 李長霖.
Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley Chapter 11: Artificial Intelligence Computer Science: An Overview Tenth Edition.
Procedures for managing workflow components Workflow components: A workflow can usually be described using formal or informal flow diagramming techniques,
TOP 5 PRIORITIES OF MANAGEMENT INFORMATION SYSTEM Mr. Jerome Estuita Presented by:
CE Operating Systems Lecture 3 Overview of OS functions and structure.
Ph No: Mob: , plot No-27, NGGO's Colony, Pattabhi reddy gardens, Visakhapatnam-07 Oracle.
Dependable software deployment Wouter Swierstra 13 October 2006.
1 MRP: Based on Dependent Demand u This chapter explores: –Master production schedule –Material Requirements Planning (MRP) »System structure »Example.
Virtual Memory The memory space of a process is normally divided into blocks that are either pages or segments. Virtual memory management takes.
Advanced Operating Systems - Spring 2009 Lecture 14 – February 25, 2009 Dan C. Marinescu Office: HEC 439 B. Office.
1 Project: Page Replacement Algorithms Lubomir Bic.
Virtualization and Databases Ashraf Aboulnaga University of Waterloo.
DISTRIBUTED COMPUTING
Progress Report 9/9. CHT Project Develop a resource management scheduling algorithm for CHT datacenter. ◦ Two types of jobs, interactive/latency- sensitive.
Functions of Management
Cloud Resource Scheduling for Online and Batch Applications Midterm report 12/16.
Miron Livny Computer Sciences Department University of Wisconsin-Madison Condor and (the) Grid (one of.
Introduction Contain two or more CPU share common memory and peripherals. Provide greater system throughput. Multiple processor executing simultaneous.
Progress Report 8/26. CHT Project Develop a resource management scheduling algorithm for CHT datacenter. ◦ Meet the SLA of jobs.  Two types of jobs,
Software Engineering Algorithms, Compilers, & Lifecycle.
Chapter I: Introduction to Computer Science. Computer: is a machine that accepts input data, processes the data and creates output data. This is a specific-purpose.
Progress Report 07/06 Simon.
Event Based Simulation of The Backfilling Algorithm OOP tirgul No
TensorFlow– A system for large-scale machine learning
Lecture 12: Real-Time Scheduling
Kubernetes Modifications for GPUs
CHT Project Progress Report
Running Multiple Schedulers in Kubernetes
Chapter 11: Artificial Intelligence
Computing Resource Allocation and Scheduling in A Data Center
湖南大学-信息科学与工程学院-计算机与科学系
Apollo Weize Sun Feb.17th, 2017.
Operating systems Process scheduling.
CPU SCHEDULING.
Horizontally Partitioned Hybrid Main Memory with PCM
Presented By: Darlene Banta
Progress Report 08/31 Simon.
David Botzer and Opher Etzion
Processing Devices.
Managing and monitoring SQL Server on Linux from the command line
Cloud Resource Scheduling for Online and Batch Applications
Progress Report 04/27 Simon.
Presentation transcript:

CHT Project Progress Report 10/07 Simon

CHT Project Develop 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

Current Plan Two 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

Scheduler Implement 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.

Rule Engine Decide 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.

Alternative Way Build 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.

Possible future extension Apply machine learning to add/adjust the rules.(CHT) Apply machine learning to minimize the size of a container/pod.(Prof. Lin)

Next Study the current predicates and priority functions inside Kubernetes. Keep working.