open stack meetup 2014 11-13 - 101 + high availability

29
OpenStack 101 Building a Highly Available Cloud Rick Ashford Sr. Technical Specialist [email protected]

Upload: rick-ashford

Post on 29-Jun-2015

237 views

Category:

Internet


2 download

DESCRIPTION

Introduction to OpenStack and Building Highly-Available clouds

TRANSCRIPT

Page 1: Open stack meetup 2014 11-13 - 101 + high availability

OpenStack 101

Building a Highly Available Cloud

Rick AshfordSr. Technical Specialist

[email protected]

Page 2: Open stack meetup 2014 11-13 - 101 + high availability

OpenStack 101

Page 3: Open stack meetup 2014 11-13 - 101 + high availability

3

Why an Open Source Cloud?

Drivers Concerns

Cost Savings

Open Standards

No Vendor Lock-In

Portability

Flexibility

Lack of Support

Security

Ecosystem

Integration

Maturity

Page 4: Open stack meetup 2014 11-13 - 101 + high availability

4

The OpenStack Transformation

• Leading open source project for building IaaS clouds

• Active developer community

• Rapid innovation: 2 integrated projects to 10

• Massive industry ecosystem

• Growing enterprise adoption

• OpenStack Foundation ensures long-term viability

Page 5: Open stack meetup 2014 11-13 - 101 + high availability

5

What is OpenStack?

Page 6: Open stack meetup 2014 11-13 - 101 + high availability

6

What is OpenStack?

You are not installing a single project. You are coordinating the installation of multiple inter-dependent projects

Page 7: Open stack meetup 2014 11-13 - 101 + high availability

7

Cheat SheetWhat do all these names mean?

Generic Project

Compute Nova

Identity Keystone

Image Glance

Block Storage Cinder

Network Neutron

Object Storage Swift (or Ceph)

Orchestration Heat

Telemetry Ceilometer

Dashboard Horizon

Database Trove

Page 8: Open stack meetup 2014 11-13 - 101 + high availability

8

What is OpenStack?

Not something you just drop off at the edge of the driveway and take off.

Page 9: Open stack meetup 2014 11-13 - 101 + high availability

Why SUSE?

Page 10: Open stack meetup 2014 11-13 - 101 + high availability

10

Linux is free! As long as your time is worth nothing to you.

Page 11: Open stack meetup 2014 11-13 - 101 + high availability

11

http://blog.adamspiers.org/2014/11/05/ruler-of-the-stack/

Page 12: Open stack meetup 2014 11-13 - 101 + high availability

12

OpenStack Distribution

Billing VM Mgmt Image Tool Portal App Monitor Sec & Perf

Cloud

Management

Orchestration(Heat)

Dashboard(Horizon)

Cloud APIs(OpenStack and

EC2)

Identity(Keystone)

Images(Glance)

Hypervisor

Compute(Nova)

Operating System

Physical Infrastructure: x86-64, Switches, Storage

OpenStack Icehouse

Object(Swift)

Network(Neutron)

Adapters Adapters

Telemetry(Ceilometer)

Block(Cinder)

SUSE Cloud Adds

RequiredServicesRabbitMQPostgresql

Ins

tall

Fra

me

wo

rk(C

row

ba

r, C

he

f, T

FT

P, D

NS

, D

HC

P)

SUSEManager

SUSEStudio

HypervisorXen, KVM

SUSE Linux Enterprise Server 11 SP3

SUSE Product

Rados

RBD

RadosGW

Ceph

Adapters

Highly Available Services

SUSE® Cloud 4

Physical Infrastructure: x86-64, Switches, Storage

Billing Portal App Monitor Sec & Perf

Adapters Adapters VMware, Hyper-V

Partner Solutions

Page 13: Open stack meetup 2014 11-13 - 101 + high availability

High-Availability Considerations

Page 14: Open stack meetup 2014 11-13 - 101 + high availability

14

High Availability for OpenStack

● First question: what are we trying to protect?

● Control Plane● Guests

Page 15: Open stack meetup 2014 11-13 - 101 + high availability

15

High Availability Directions

• Control Plane‒ Hot standby for Control Nodes

‒ Ensures that cloud continues to operate

• Guest vs. Server‒ Cloud 101 – Plan for infrastructure failure, or

‒ Enterprise 101 – build a reliable infrastructure

• High availability guests‒ Use high-availability tools in VMs

‒ Failover VM is physically separated

‒ Application does not need to be changed

• High availability compute nodes‒ Use high-availability tools on physical nodes

‒ Failover machine is in same availability zone, but could be geographically different

‒ All workloads on server are backed up

Page 16: Open stack meetup 2014 11-13 - 101 + high availability

16

OpenStack Distribution Components

Page 17: Open stack meetup 2014 11-13 - 101 + high availability

17

Component failure impact

• Admin server‒ New cloud nodes require manual addition and configuration

‒ Currently no ability to rediscover existing nodes on restart

‒ No impact on currently operating cloud

• Control node‒ Cannot start or stop guest instances

‒ No ability to rediscover existing nodes or guest VMs on restart

‒ No impact on currently deployed instances

• Compute Node‒ Loss of VMs on that node

‒ Recovery is by restart and re-provisioning of physical server

‒ Can be mitigated through application design

Page 18: Open stack meetup 2014 11-13 - 101 + high availability

18

Component failure assessment

• Control Node‒ Highest priority

‒ Recovery realistically requires complete cloud restart

• Compute Node‒ Application level recovery is normal practice for existing clouds

‒ Not existing “enterprise” expectation, but workaround exists for new workloads

• Admin Server‒ Least impact on deployed system

‒ Operation can continue with no impact on end users

Page 19: Open stack meetup 2014 11-13 - 101 + high availability

19

Cloud Structure

Cloud Orchestration

HostServer

HostServer

HostServer

HostServer

HostServer

HostServer

...

VM VM VMVMVM VM VMVM ...Control Node

Page 20: Open stack meetup 2014 11-13 - 101 + high availability

20

Zone BZone AControlCluster

Cloud Structure – HA Control Cluster

Cloud Orchestration

HostServer

HostServer

HostServer

HostServer

HostServer

HostServer

...

VM VM VMVMVM VM VMVM ...Control Node

Control Node

Page 21: Open stack meetup 2014 11-13 - 101 + high availability

What about the workloads?

Page 22: Open stack meetup 2014 11-13 - 101 + high availability

22

Zone BZone A

Cloud Structure – Availability Zones

Cloud Orchestration

HostServer

HostServer

HostServer

HostServer

HostServer

HostServer

...

VM VM VMVMVM VM VMVM ...Control Node

Page 23: Open stack meetup 2014 11-13 - 101 + high availability

23

Zone BZone A High Availability ClusterControlCluster

Cloud Structure – Host Cluster

Cloud Orchestration

HostServer

HostServer

HostServer

HostServer

HostServer

HostServer

...

VM VM VMVMVM VM VMVM ...Control Node

Control Node

Page 24: Open stack meetup 2014 11-13 - 101 + high availability

24

Zone BZone A

High Availability Cluster

ControlCluster

Cloud Structure – VM Cluster

Cloud Orchestration

HostServer

HostServer

HostServer

HostServer

HostServer

HostServer

...

VM VM VMVMVM VM VMVM ...Control Node

Control Node

Page 25: Open stack meetup 2014 11-13 - 101 + high availability

Cluster Architecture

Page 26: Open stack meetup 2014 11-13 - 101 + high availability

26

3 Node Cluster Overview

Host

VM1

ApacheIP

Host Host

Corosync + openAIS

Pacemaker

DLM

cLVM2+OCFS2

VM2

Network Links

Clients

Storage

Page 27: Open stack meetup 2014 11-13 - 101 + high availability
Page 28: Open stack meetup 2014 11-13 - 101 + high availability

Unpublished Work of SUSE. All Rights Reserved.This work is an unpublished work and contains confidential, proprietary and trade secret information of SUSE. Access to this work is restricted to SUSE employees who have a need to know to perform tasks within the scope of their assignments. No part of this work may be practiced, performed, copied, distributed, revised, modified, translated, abridged, condensed, expanded, collected, or adapted without the prior written consent of SUSE. Any use or exploitation of this work without authorization could subject the perpetrator to criminal and civil liability.

General DisclaimerThis document is not to be construed as a promise by any participating company to develop, deliver, or market a product. It is not a commitment to deliver any material, code, or functionality, and should not be relied upon in making purchasing decisions. SUSE makes no representations or warranties with respect to the contents of this document, and specifically disclaims any express or implied warranties of merchantability or fitness for any particular purpose. The development, release, and timing of features or functionality described for SUSE products remains at the sole discretion of SUSE. Further, SUSE reserves the right to revise this document and to make changes to its content, at any time, without obligation to notify any person or entity of such revisions or changes. All SUSE marks referenced in this presentation are trademarks or registered trademarks of Novell, Inc. in the United States and other countries. All third-party trademarks are the property of their respective owners.

Page 29: Open stack meetup 2014 11-13 - 101 + high availability

Thank you.

33

Call to action line oneand call to action line twowww.calltoaction.com