Policy-Based Scheduling: Improving Resources Provisioning In OpenStack May 12 th, 2014 Atlanta Khanh-Toan TRAN (R&D Engineer)

Slides:



Advertisements
Similar presentations
Agile Infrastructure built on OpenStack Building The Next Generation Data Center with OpenStack John Griffith, Senior Software Engineer,
Advertisements

Take your CMS to the cloud to lighten the load Brett Pollak Campus Web Office UC San Diego.
Open Stack Summit – Hong Kong OPENSTACK
Cloud Computing Imranul Hoque. Today’s Cloud Computing.
1 Security on OpenStack 11/7/2013 Brian Chong – Global Technology Strategist.
Oracle Enterprise Manager – Cloud Control 12c Simon Keys, The Small Ronnie Martin Lambert, The Large Ronnie.
Apache CloudStack Evolution Proposal Alex Huang Software Architect, Citrix Systems.
DataGrid is a project funded by the European Union 22 September 2003 – n° 1 EDG WP4 Fabric Management: Fabric Monitoring and Fault Tolerance
Zhipeng (Howard) Huang
© 2009 IBM Corporation RESEARCH Peeking into Cloud for better Application Manageability Sambit Sahu IBM Research.
FI-WARE – Future Internet Core Platform FI-WARE Cloud Hosting July 2011 High-level description.
Microsoft Virtual Server 2005 Product Overview Mikael Nyström – TrueSec AB MVP Windows Server – Setup/Deployment Mikael Nyström – TrueSec AB MVP Windows.
1 Doctor Fault Management 18 May 2015 Ryota Mibu, NEC.
Architecture overview 6/03/12 F. Desprez - ISC Cloud Context : Development of a toolbox for deploying application services providers with a hierarchical.
VMware vCenter Server Module 4.
© 2008 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice. HP Automates Infrastructure Outsourcing.
Web application architecture
Opensource for Cloud Deployments – Risk – Reward – Reality
Cloud Computing for the Enterprise November 18th, This work is licensed under a Creative Commons.

Module 12: Designing an AD LDS Implementation. AD LDS Usage AD LDS is most commonly used as a solution to the following requirements: Providing an LDAP-based.
Virtual Machine Hosting for Networked Clusters: Building the Foundations for “Autonomic” Orchestration Based on paper by Laura Grit, David Irwin, Aydan.
CLOUD COMPUTING  IT is a service provider which provides information.  IT allows the employees to work remotely  IT is a on demand network access.
+ CS 325: CS Hardware and Software Organization and Architecture Cloud Architectures.
INTRODUCTION TO CLOUD COMPUTING CS 595 LECTURE 2.
Ocean Observatories Initiative Common Execution Infrastructure (CEI) Overview Michael Meisinger September 29, 2009.
Cloud Computing & Amazon Web Services – EC2 Arpita Patel Software Engineer.
From Virtualization Management to Private Cloud with SCVMM 2012 Dan Stolts Sr. IT Pro Evangelist Microsoft Corporation
1 Week #10Business Continuity Backing Up Data Configuring Shadow Copies Providing Server and Service Availability.
Copyright © cs-tutorial.com. Overview Introduction Architecture Implementation Evaluation.
Chapter 2 Database System Concepts and Architecture Dr. Bernard Chen Ph.D. University of Central Arkansas.
SC2012 Infrastructure Components Management Justin Cook (Data # 3) Principal Consultant, Systems Management Noel Fairclough (Data # 3) Consultant, Systems.
Microsoft Virtual Academy. STANDARDIZATION SELF SERVICEAUTOMATION Give Customers of IT services the ability to identify, access and request services.
Visual Studio Windows Azure Portal Rest APIs / PS Cmdlets US-North Central Region FC TOR PDU Servers TOR PDU Servers TOR PDU Servers TOR PDU.
VMware vSphere Configuration and Management v6
What’s New with Windows Server 2012 and Microsoft System Center 2012 SP1 Vijay Tewari Principal Group Program Manager Microsoft Corporation.
Project Name Program Name Project Scope Title Project Code and Name Insert Project Branding Image Here.
20409A 7: Installing and Configuring System Center 2012 R2 Virtual Machine Manager Module 7 Installing and Configuring System Center 2012 R2 Virtual.
Chapter 8 – Cloud Computing
Vignesh Ravindran Sankarbala Manoharan. Infrastructure As A Service (IAAS) is a model that is used to deliver a platform virtualization environment with.
OpenStack overview of the project Belmiro Daniel Rodrigues Moreira CERN IT-PES-PS January 2011 Disclaimer: This presentation reflects the experience and.
noun ; Software Defined Enterprise/SDE/ The enterprise who leverages software to flank their traditional business offerings, or to create entirely new.
Cloud Computing – UNIT - II. VIRTUALIZATION Virtualization Hiding the reality The mantra of smart computing is to intelligently hide the reality Binary->
SYSTEM CENTER VIRTUAL MACHINE MANAGER 2012 Gorazd Šemrov Microsoft Consulting Services
Data Hosting and Security Overview January, 2011.
Platform & Engineering Services CERN IT Department CH-1211 Geneva 23 Switzerland t PES Agile Infrastructure Project Overview : Status and.
RED HAT ENTERPRISE VIRTUALIZATION 3.0 YOUR STRATEGIC VIRTUALIZATION ALTERNATIVE.
Ashiq Khan NTT DOCOMO Congress in NFV-based Mobile Cellular Network Fault Recovery Ryota Mibu NEC Masahito Muroi NTT Tomi Juvonen Nokia 28 April 2016OpenStack.
Ashiq Khan NTT DOCOMO Congress in NFV-based Mobile Cellular Network Fault Recovery Ryota Mibu NEC Masahito Muroi NTT Tomi Juvonen Nokia 28 April 2016OpenStack.
ETRIKS Platform for bioinformatics ISGC 17/03/15 Pengfei Liu, CC-IN2P3/CNRS.
EPAM Cloud Orchestration
Md Baitul Al Sadi, Isaac J. Cushman, Lei Chen, Rami J. Haddad
Unix Server Consolidation
Course: Cluster, grid and cloud computing systems Course author: Prof
EPAM Cloud Orchestration
Security on OpenStack 11/7/2013
Principles of Computer Security
StratusLab Final Periodic Review
StratusLab Final Periodic Review
Hybrid Cloud Architecture for Software-as-a-Service Provider to Achieve Higher Privacy and Decrease Securiity Concerns about Cloud Computing P. Reinhold.
Introduction to Data Management in EGI
VIDIZMO Deployment Options
OpenStack Ani Bicaku 18/04/ © (SG)² Konsortium.
GGF15 – Grids and Network Virtualization
Cloud Computing Dr. Sharad Saxena.
DPACC Management Aspects
Users Manage Terabytes of Data with Powerful and Agnostic Hosting from Azure Cloud Service Partner Logo “Given the challenges we face both in dealing with.
06 | SQL Server and the Cloud
Presentation transcript:

Policy-Based Scheduling: Improving Resources Provisioning In OpenStack May 12 th, 2014 Atlanta Khanh-Toan TRAN (R&D Engineer)

Outline Motivation  Why Smart Placement ? Current Openstack Provisioning  Overview  Limitation Proposed Policy-Based Scheduler Usecases & Demo Meta-scheduling Motivation  Why Smart Placement ? Current Openstack Provisioning  Overview  Limitation Proposed Policy-Based Scheduler Usecases & Demo Meta-scheduling

MOTIVATIONMOTIVATION

Smart Placement Expectations For the clients perspective  Contract guarantee  Transparency  Better experience  More services, more offers  Attractive price For the cloud provider perspective  Flexible scheduling Scheduling per client Scheduling per resources Scheduling per context  Capable of realizing different admin objectives and operations  Efficient and simplified control and management by policies For the clients perspective  Contract guarantee  Transparency  Better experience  More services, more offers  Attractive price For the cloud provider perspective  Flexible scheduling Scheduling per client Scheduling per resources Scheduling per context  Capable of realizing different admin objectives and operations  Efficient and simplified control and management by policies

Smart Placement Big Picture Smart Placement Compute, storage and network nodes, etc … Datacenter 1Datacenter 2Datacenter n Cloud provider operation Clients Environmental info Client contract Regulation OpenStack High-level requirements Environmental info Infrastructure info Security policies

OVERVIEW OF THE EXISTING OPENSTACK PROVISIONING

Placement in Openstack: FilterScheduler Filter_Scheduler Nova-scheduler Filter Weigh Request Filter Weigh 1 VM 2VCPU, 2GB RAM Service=gold Filtering Weighing Compute 1 8VCPU; 8Go Compute 2 3VCPU; 4Go weight = 8 weight = 4 Compute 1 8VCPU; 8GB RAM Compute 2 3VCPU; 4GB RAM Compute 3 4VCPU; 1GB RAM Service=gold Compute 4 8VCPU; 8GB RAM Service=gold Compute 1 8VCPU; 8GB RAM Compute 2 3VCPU; 4GB RAM Compute 3 4VCPU; 1GB RAM Compute 4 8VCPU; 4GB RAM Openstack Nova:  Scheduling Module : Driver  Default driver: FilterScheduler 1.Filters out hosts that are not admissible to hosts the VMs (Filter). 2.Weighs the remaining hosts following their capacity (Weigher) Openstack Nova:  Scheduling Module : Driver  Default driver: FilterScheduler 1.Filters out hosts that are not admissible to hosts the VMs (Filter). 2.Weighs the remaining hosts following their capacity (Weigher)

FilterScheduler Limitations Static scheduling  Difficult to answer to admin’s various objectives  Ineffective in management in different situations (maintenance, failure, breakout, catastrophe…) Lack of client context  No consideration for each client requirements  Client contract is not taken into account Lack of consideration for infrastructure  Unable to provide fine-grain scheduling Static scheduling  Difficult to answer to admin’s various objectives  Ineffective in management in different situations (maintenance, failure, breakout, catastrophe…) Lack of client context  No consideration for each client requirements  Client contract is not taken into account Lack of consideration for infrastructure  Unable to provide fine-grain scheduling Example: A cloud service provider wishes to apply the global Load Balancing policy to the whole datacenter and a local Consolidation policy to a cluster of hosts. Example: A cloud service provider wishes to apply the global Load Balancing policy to the whole datacenter and a local Consolidation policy to a cluster of hosts. Unfeasible

Two-step approach for the situation improvement Short term  Policy-based scheduling solution (incremental): Enhancing the existing Nova-centric architecture. Mid / Long term  Meta-scheduling solution (disruptive): independent scheduling service for OpenStack on top of Nova, Cinder and Neutron. Short term  Policy-based scheduling solution (incremental): Enhancing the existing Nova-centric architecture. Mid / Long term  Meta-scheduling solution (disruptive): independent scheduling service for OpenStack on top of Nova, Cinder and Neutron.

POLICY-BASED SCHEDULING (SHORT TERM) POLICY-BASED SCHEDULING (SHORT TERM)

Policy-Based Scheduling Module Decoupling the scheduling logic and the execution domain The scheduling logic is represented as Rule: Target – Effect – Condition with correspondent Plugins. Rules are stored in Policy Repository (PR) Policy-Based Scheduling Engine (PBSE) consults the Policy Repository to find and execute the suitable rules Policy-Based Scheduler (PBS) is the proxy between Nova- scheduler and PBSE Decoupling the scheduling logic and the execution domain The scheduling logic is represented as Rule: Target – Effect – Condition with correspondent Plugins. Rules are stored in Policy Repository (PR) Policy-Based Scheduling Engine (PBSE) consults the Policy Repository to find and execute the suitable rules Policy-Based Scheduler (PBS) is the proxy between Nova- scheduler and PBSE Nova-scheduler Filter Weigh Client request Filter Weigh Policy-Based Scheduler (PBS) Policy-Based Scheduling Engine (PBSE) Policy Repository Plugin

Policy-Based Scheduling Nova-scheduler Filter Weigh Client request Filter Weigh Policy-Based Scheduler (PBS) Policy-Based Scheduling Engine (PBSE) Policy Repository Plugin TargetRules « all »{ «effect »: {« LoadBalancing»: «Ram » }, «condition » : «all» } « aggr: aggr-1 » { « effect »:{« Consolidation»:«Ram»}, «condition » : « all » } Service_Class_Plugin LoadBalancing_Plugin Consolidation_Plugin

Policy-Based Scheduling Availability-Zone-1 Aggregate-1 Server- 01 Server- 02 Availability-Zone-2 Aggr-2Aggr-3 Load-Balancing Consolidation Server- 03 Server- 04

USE CASES & DEMO

Usecases and Demo Use Cases 1: Respecting the regulationRespecting the regulation  A French medical company wants that all their VMs and data are hosted in France as requested by French authority  All VMs are initiated in France zone regardless of the choice of the users (e.g. employees) Use Case 2: Enforcing the client contractEnforcing the client contract  A client signs a gold contract  All his VMs are initiated in the gold zone with high-end equipments (e.g. high freq. CPU, high-mem, SSD storage…) Use Case 3: Reducing software licencesReducing software licences  A cloud provider signs a contract with an ISV which charges the license fee based on the number of hardware  Admin wants to apply 2 policies: A global Load-Balancing policy in all his datacenter to favor the load sharing and limiting the overload A local Consolidation policy in the zone dedicated to the software to reduce the number of charged physical servers Use Cases 1: Respecting the regulationRespecting the regulation  A French medical company wants that all their VMs and data are hosted in France as requested by French authority  All VMs are initiated in France zone regardless of the choice of the users (e.g. employees) Use Case 2: Enforcing the client contractEnforcing the client contract  A client signs a gold contract  All his VMs are initiated in the gold zone with high-end equipments (e.g. high freq. CPU, high-mem, SSD storage…) Use Case 3: Reducing software licencesReducing software licences  A cloud provider signs a contract with an ISV which charges the license fee based on the number of hardware  Admin wants to apply 2 policies: A global Load-Balancing policy in all his datacenter to favor the load sharing and limiting the overload A local Consolidation policy in the zone dedicated to the software to reduce the number of charged physical servers

META-SCHEDULING (MID / LONG TERM) META-SCHEDULING

Meta-Scheduling Policy-Based Scheduler blueprint:

Meta-Scheduling

Related works Scheduler service: Gantt   run-instance run-instance  Remove DB between scheduler and compute nodes  Instance Group API  extension extension Support for Multiple Active Schedulers  SolverScheduler  Scheduler service: Gantt   run-instance run-instance  Remove DB between scheduler and compute nodes  Instance Group API  extension extension Support for Multiple Active Schedulers  SolverScheduler 

Thanks for your attention Policy-Based Scheduler blueprint: