Se está descargando tu SlideShare. ×

Not just for Developers: Cloud Foundry for Ops! (VMworld 2014)

•

08 de sep de 2014

• •

Anuncio

Anuncio

Anuncio

Anuncio

Anuncio

Anuncio

Anuncio

Anuncio

Anuncio

Anuncio

Anuncio

Anuncio

Not just for Developers:
Cloud Foundry for Ops!
VMWorld 2014
Cornelia Davis  @cdavisafc  cdavis@pivotal.io
Director,...

Demo
© Copyright 2013 Pivotal. All rights reserved. 2

Pivotal CF: Developer and Operator Productivity
Built-in and Ecosystem
Services
• Applications
• Rabbit MQ
• MySQL HA...

PIVOTAL CF
Four Layers of HA
© Copyright 2013 Pivotal. All rights reserved. 4

Application Instances and Availability Zones
Router
App Ops
Zone 1 Zone 2 Application instances
DEA
DEA
DEA
are eve...

Failed Application Instances Replaced
Router
Blobstore
Cloud
Controller
Desired State Actual State
Health Manager
M...

ERS Processes are Monitored
PaaS Ops
Health Manager
AGENT
DEA
AGENT
Cloud Controller
AGENT
Message Bus
Pivotal CF...

ERS Processes are Monitored
PaaS Ops
Health Manager
AGENT
DEA
AGENT
Cloud Controller
AGENT
Message Bus
Pivotal CF...

ERS Processes are Monitored
PaaS Ops
Health Manager
AGENT
DEA
AGENT
Cloud Controller
AGENT
Message Bus
Pivotal CF...

VMs are Monitored
PaaS Ops
Health Manager
AGENT
DEA
AGENT
Cloud Controller
AGENT
BOSH Director
Message Bus
Pivot...

VMs are Monitored
PaaS Ops
Health Manager
AGENT
DEA
AGENT
Cloud Controller
AGENT
BOSH Director
Message Bus
Pivot...

VMs are Monitored
PaaS Ops
Health Manager
AGENT
DEA
AGENT
Cloud Controller
AGENT
BOSH Director
Message Bus
CPI
...

Four levels of HA in PCF
Elastic Runtime (ERS):
 Distribution across availability zones
 Application health managemen...

Thank You
Pivotal Confidential––Internal Use Only 15

BUILT FOR THE SPEED OF BUSINESS

Siguientes SlideShares

Próximo SlideShare

Four Levels of High Availability in Cloud Foundry (Cloud Foundry Summit 2014)

Four Levels of High Availability in Cloud Foundry (Cloud Foundry Summit 2014)

Cargando en…3

×

Eche un vistazo a continuación

Quantifying the Benefits of Cloud Foundry

Building a University Community PaaS Using Cloud Foundry (Cloud Foundry Summ...

How to Scale Operations for a Multi-Cloud Platform using PCF

Case Study: Orange Labs & Cloud Foundry

Guillaume Berche

Dissecting The PaaS Landscape

Rishidot Research

Cloud foundry architecture and deep dive

Using Pivotal Cloud Foundry with Google’s BigQuery and Cloud Vision API

Practical PaaS presentation

Not just for Developers: Cloud Foundry for Ops! (VMworld 2014)

08 de sep de 2014

• •

Descargar para leer sin conexión

Presented by: Cornelia Davis - Platform Engineer, Cloud Foundry, Pivotal

If you believe everything you’ve read about Platform as a Service (PaaS) you probably think it’s all about the developer. If we told you that a Pivotal CF could auto scale your applications based on current load and provide consolidated logs and monitoring across all app instances, would your operators be happy? If they learned that four levels of high availability would cut down on the middle of the night pages, would they rejoice?

We’ll show the wealth of operational benefits realized with use of Pivotal CF, powered by Cloud Foundry. Learn how Pivotal CF can free your IT Operations staff from fire fighting duties, allowing them to innovate instead.

Presented by: Cornelia Davis - Platform Engineer, Cloud Foundry, Pivotal

If you believe everything you’ve read about Platform as a Service (PaaS) you probably think it’s all about the developer. If we told you that a Pivotal CF could auto scale your applications based on current load and provide consolidated logs and monitoring across all app instances, would your operators be happy? If they learned that four levels of high availability would cut down on the middle of the night pages, would they rejoice?

We’ll show the wealth of operational benefits realized with use of Pivotal CF, powered by Cloud Foundry. Learn how Pivotal CF can free your IT Operations staff from fire fighting duties, allowing them to innovate instead.

Más Contenido Relacionado

Quantifying the Benefits of Cloud Foundry

Building a University Community PaaS Using Cloud Foundry (Cloud Foundry Summ...

How to Scale Operations for a Multi-Cloud Platform using PCF

Case Study: Orange Labs & Cloud Foundry

Guillaume Berche

Dissecting The PaaS Landscape

Rishidot Research

Cloud foundry architecture and deep dive

Using Pivotal Cloud Foundry with Google’s BigQuery and Cloud Vision API

Practical PaaS presentation

PCF Cloud-Native Workshop Slides

What's new in Pivotal Cloud Foundry 1.6

Deploy your Multi-tier Application in Cloud Foundry

PaaS Solutions Comparison

GlobalLogic Ukraine

Enterprise PaaS Golden Pitch

Devops: Enabled Through a Recasting of Operational Roles

Adopting Azure, Cloud Foundry and Microservice Architecture at Merrill Corpor...

Unlock your VMWare Investment with Pivotal Cloud Foundry (VMworld 2014)

Cloud Foundry - An Open Innovation Platform

All Things Open

Introduction to Platform-as-a-Service and Cloud Foundry

Manuel Silveyra

Hack for Good and Profit (Cloud Foundry Summit 2014)

Part 4: Custom Buildpacks and Data Services (Pivotal Cloud Platform Roadshow)

Quantifying the Benefits of Cloud Foundry

23 diapositivas

Building a University Community PaaS Using Cloud Foundry (Cloud Foundry Summ...

34 diapositivas

How to Scale Operations for a Multi-Cloud Platform using PCF

20 diapositivas

Case Study: Orange Labs & Cloud Foundry

Guillaume Berche

32 diapositivas

Dissecting The PaaS Landscape

Rishidot Research

34 diapositivas

Cloud foundry architecture and deep dive

14 diapositivas

Ahli jawatankuasa

Elecciones locales DF 2015

Akcja Zdrowo i sportowo - rok szkolny 2015/2016

Tennis&Management

Report Hi Bo Maxi Ch

Eddie Robb from Make it Social Panel/Presentation at The Travel Marketing Sum...

Ejemplo de producto con Código QR

Retrospectiva y tendencias futuras: Quince años de publicidad online y market...

Tu mejor amigo será virtual

José María Puerta González

Litterae calcatae a paideia, setembre 2011

El gato agazapado - Cuento

Super Realistic

Co.incidir 28 junio 2016 (1)

Trastorno de ansiedad por separación

Presentacion Mobile Marketing - Ivan Orbegozo NEO Consulting

Salman Alfarizi

Verso il "Social Customer Care"

Sebastian Aka Beaver

Ahli jawatankuasa

14 diapositivas

Elecciones locales DF 2015

21 diapositivas

Akcja Zdrowo i sportowo - rok szkolny 2015/2016

Tennis&Management

11 diapositivas

Report Hi Bo Maxi Ch

50 diapositivas

Pivotal cf for_devops_mkim_20141209

Development on cloud_paa_s_sddc_mkim_20141216_final

Development on Cloud,PaaS and SDDC

VMworld_PivotalCF_And_Containers

Pivotal CF 소개

Declarative Infrastructure with Cloud Foundry BOSH

Part 2: Architecture and the Operator Experience (Pivotal Cloud Platform Road...

Sydney cloud foundry meetup - Service Brokers

Lawrence Crowther

Cloud Roundtable | Pivoltal: Agile platform

Journey to the Modern App with Containers, Microservices and Big Data

Pivotal Big Data Suite: A Technical Overview

Complete Visibility into Docker Containers with AppDynamics

Agile Tour Pune 2015: Dev-ops- niche or mainstream: Bhaskar Venugopalan

India Scrum Enthusiasts Community

Supercharge Your Application Delivery: The Journey to Enterprise PaaS

Cloud Foundry - Second Generation Code (CCNG). Technical Overview

Supercharge Your Application Delivery

Pivotal Cloud Foundry: A Technical Overview

Pivotal Cloud Foundry: A Technical Overview

It’s a Mobile First World: Faster Mobile Apps with Pivotal and VMware

Cloud Foundry Diego, Lattice, Docker and more

Pivotal cf for_devops_mkim_20141209

44 diapositivas

Development on cloud_paa_s_sddc_mkim_20141216_final

52 diapositivas

Development on Cloud,PaaS and SDDC

52 diapositivas

VMworld_PivotalCF_And_Containers

11 diapositivas

Pivotal CF 소개

49 diapositivas

Declarative Infrastructure with Cloud Foundry BOSH

31 diapositivas

Why you shouldn't worry about scaling your Java webapp.pdf

Let's keep it simple and streaming.pdf

Introduction to Testcontainers

Running Untrusted Code in Spring with WebAssembly

Declarative Clients in Spring

Product Development Journey Markers

Tanzu Developer Connect Workshop - Philadelphia

Level Up Your Integration Testing With Testcontainers

Who Should Use Kubernetes

Getting Started with Spring for GraphQL

Responsible Microservices

Getting Started with Kubernetes

A New CLI for Spring Developer Productivity

Observability: Beyond the Three Pillars with Spring

Application Modernization: Migrating mainframe apps to the cloud using Spring

More Maintainable, Testable Applications with Dependency Injection in Spring

The Unicorn Project and the Five Ideals.pdf

Tracy Miranda_DevOps Loop, May 2022.pdf

Stacy Cashmore_What has speaking done for us.pptx.pdf

Sam Newman_Designing For Habitability.pdf

Why you shouldn't worry about scaling your Java webapp.pdf

Let's keep it simple and streaming.pdf

38 diapositivas

Introduction to Testcontainers

24 diapositivas

Running Untrusted Code in Spring with WebAssembly

19 diapositivas

Declarative Clients in Spring

61 diapositivas

Product Development Journey Markers

50 Steps to Improving Your Academic W... (z-lib.org).pdf

Topic 4 Chapter 12.pdf

Neo4j GraphSummit London - The Path To Success With Graph Database and Data S...

RDT Analysis.pptx

UN BUSINESS TRANSFORMATION ORACLE EOI.pdf

Top 10 Gaming Laptops under 50000

ankitrana744793

Automated Provisioning, Management & Cost Control for Kubernetes Clusters

🌺 Women in Automation Program: Become a Citizen Developer with UiPath StudioX...

RFP-PROVISION OF CONSULTANCY SERVICES TO IPSAS.pdf

EY + Neo4j GraphSummit London 2023.pptx

Unit 4 COA.pptx

AkshatBhatnagar30

overview of DRILLING OPERATION.pptx

Accenture-Cloud-Data-Migration-POV-Final.pdf

Building Modern Applications on AWS.pptx

DDGK: Learning Graph Representations for Deep Divergence Graph Kernels

5 Web Development Trends for 2023 Predicted by Experts

FORCES - EBS Upgrade Compared to SaaS Cloud.pdf

HOLIDAY ASSIGNMENT IT ,pavan.pptx

akashpatil146883

i18n was the missing piece_ Let 70%+ of the users in the world access your ap...

50 Steps to Improving Your Academic W... (z-lib.org).pdf

271 diapositivas

Topic 4 Chapter 12.pdf

23 diapositivas

Neo4j GraphSummit London - The Path To Success With Graph Database and Data S...

42 diapositivas

RDT Analysis.pptx

UN BUSINESS TRANSFORMATION ORACLE EOI.pdf

17 diapositivas

Top 10 Gaming Laptops under 50000

ankitrana744793

12 diapositivas

Not just for Developers: Cloud Foundry for Ops! (VMworld 2014)

1. Not just for Developers: Cloud Foundry for Ops! VMWorld 2014 Cornelia Davis  @cdavisafc  cdavis@pivotal.io Director, Platform Engineering, Cloud Foundry Pivotal © Copyright 2013 Pivotal. All rights reserved. 1
2. Demo © Copyright 2013 Pivotal. All rights reserved. 2
3. Pivotal CF: Developer and Operator Productivity Built-in and Ecosystem Services • Applications • Rabbit MQ • MySQL HA • Cassandra • Elasticsearch • Jenkins (CI) • Memcached • MongoDB • Neo4j • Redis • Riak CS • Data/Analytics • Elastic Hadoop • HAWQ • GemFire XD • Mobile • Push Notification • Data Synch • API Gateway Easy to add and customize Simple, Developer Friendly Commands & API • Auto-detect frameworks • “Push and it works” • Simple service binding • Agile Microservices Easy to add and customize Operational Benefits for Every Application • Instant dynamic routing • Log stream aggregation • Access controls & policies • Built-in Containerization • APM & Operational metrics • 4 Layers of High Availability • App-Instance • Availability Zone • Process • Virtual Machine Deploy, Operate, Update & Scale with minimal downtime on IaaS ….and more • .WAR • Dockerfile • .NET © Copyright 2013 Pivotal. All rights reserved. 3
4. PIVOTAL CF Four Layers of HA © Copyright 2013 Pivotal. All rights reserved. 4
5. Application Instances and Availability Zones Router App Ops Zone 1 Zone 2 Application instances DEA DEA DEA are evenly distributed over two availability zones. Loosing an AZ keeps instances running and available. Pivotal CF Elastic Runtime DEA DEA DEA © Copyright 2013 Pivotal. All rights reserved. 5
6. Failed Application Instances Replaced Router Blobstore Cloud Controller Desired State Actual State Health Manager Messaging (NATS) DEA DEA DEA App Ops Pivotal CF Elastic Runtime © Copyright 2013 Pivotal. All rights reserved. 6
7. ERS Processes are Monitored PaaS Ops Health Manager AGENT DEA AGENT Cloud Controller AGENT Message Bus Pivotal CF Operations Manager IaaS Health Monitor Responses: pager email monitoring … © Copyright 2013 Pivotal. All rights reserved. 8
8. ERS Processes are Monitored PaaS Ops Health Manager AGENT DEA AGENT Cloud Controller AGENT Message Bus Pivotal CF Operations Manager IaaS Health Monitor Responses: pager email monitoring … © Copyright 2013 Pivotal. All rights reserved. 9
9. ERS Processes are Monitored PaaS Ops Health Manager AGENT DEA AGENT Cloud Controller AGENT Message Bus Pivotal CF Operations Manager IaaS Health Monitor Responses: pager email monitoring … © Copyright 2013 Pivotal. All rights reserved. 10
10. VMs are Monitored PaaS Ops Health Manager AGENT DEA AGENT Cloud Controller AGENT BOSH Director Message Bus Pivotal CF Operations Manager IaaS Desired State Actual State Health Monitor Responses: pager email monitoring ressurector … © Copyright 2013 Pivotal. All rights reserved. 11
11. VMs are Monitored PaaS Ops Health Manager AGENT DEA AGENT Cloud Controller AGENT BOSH Director Message Bus Pivotal CF Operations Manager IaaS Desired State Actual State Health Monitor Responses: pager email monitoring ressurector … © Copyright 2013 Pivotal. All rights reserved. 12
12. VMs are Monitored PaaS Ops Health Manager AGENT DEA AGENT Cloud Controller AGENT BOSH Director Message Bus CPI Pivotal CF Operations Manager IaaS Desired State Actual State Health Monitor Responses: pager email monitoring ressurector … © Copyright 2013 Pivotal. All rights reserved. 13
13. Four levels of HA in PCF Elastic Runtime (ERS):  Distribution across availability zones  Application health management and recovery Operations Manager (cluster management):  Process monitoring, recovery and alerting  Virtual machine health monitoring, recovery and alerting http://blog.gopivotal.com/cloud-foundry-pivotal/products/the-four-levels-of-ha-in-pivotal-cf © Copyright 2013 Pivotal. All rights reserved. 14
14. Thank You Pivotal Confidential––Internal Use Only 15
15. BUILT FOR THE SPEED OF BUSINESS

Notas del editor

Edgar
Building the Pivotal CF story via animation
Developer experience - where we started from and keep improving
Operation benefits - When we started selling to enterprise customer beyond a cloud service, both at the Cloud Ops and Application Ops level
3 groups of build-in services (Pivotal and ecosystem) – Apps, Data/Analytics & Mobile. All Services are both deployed in a consistent / cloud in depended way by Cloud Ops and exposed to App Developer for easy binding to applications via the ERS marketplace
All 3 pillars (Dev, Ops, Services) enjoy a common operational, continuous delivery and scale on any IaaS without app/service code of config changes
Edgar
Zone could = rack, for example
The elastic runtime will keep the number of instances you’ve requested running by:
DEAs constantly reporting their state
Health manager constantly updating actual state model across all DEAs
HM periodically requests desired state from the cloud controller
When a difference is found, HM advises CC
CC initiates deployment of a new instance
A static version of the previous slide.
The processes running on the virtual machines (i.e. the health manager) are started with monit, a nice little utility that keeps an eye on processes and will respond when one dies. If a process dies then, monit will do two things: first, it will try to restart the process and second, whether the restart is successful or not, it will tell the BOSH agent about the failure. Recall that the Bosh agent is there to communicate with Operations Manager and in this case it will relay this failure information to the Operations Manager Health Monitor (not to be confused with the Health Manager of the elastic runtime that was discussed above) – we’ll abbreviate it OMHM. The OMHM will take this alert and pass it through a list of responders that can be configured to do things like send emails, page administrators and display alerts in operations dashboards. There’s a good chance that monit will already have recovered the process, but we also want there to be an opportunity for a human to respond.
The processes running on the virtual machines (i.e. the health manager) are started with monit, a nice little utility that keeps an eye on processes and will respond when one dies. If a process dies then, monit will do two things: first, it will try to restart the process and second, whether the restart is successful or not, it will tell the BOSH agent about the failure. Recall that the Bosh agent is there to communicate with Operations Manager and in this case it will relay this failure information to the Operations Manager Health Monitor (not to be confused with the Health Manager of the elastic runtime that was discussed above) – we’ll abbreviate it OMHM. The OMHM will take this alert and pass it through a list of responders that can be configured to do things like send emails, page administrators and display alerts in operations dashboards. There’s a good chance that monit will already have recovered the process, but we also want there to be an opportunity for a human to respond.
The processes running on the virtual machines (i.e. the health manager) are started with monit, a nice little utility that keeps an eye on processes and will respond when one dies. If a process dies then, monit will do two things: first, it will try to restart the process and second, whether the restart is successful or not, it will tell the BOSH agent about the failure. Recall that the Bosh agent is there to communicate with Operations Manager and in this case it will relay this failure information to the Operations Manager Health Monitor (not to be confused with the Health Manager of the elastic runtime that was discussed above) – we’ll abbreviate it OMHM. The OMHM will take this alert and pass it through a list of responders that can be configured to do things like send emails, page administrators and display alerts in operations dashboards. There’s a good chance that monit will already have recovered the process, but we also want there to be an opportunity for a human to respond.
Of course, the BOSH agent on a VM can only communicate back to the Operations Manager if the VM is there, so let’s talk about what happens when a VM disappears. First thing to understand is that by “disappear” I mean that the BOSH agent is not functional; the VM could be there, but Ops Manager no longer knows what it is up to so for all intents and purposes it’s “gone”. How does Ops Manager know? One of the things that a BOSH agent is responsible for is sending out heartbeat messages and by default it does so every 60 seconds. The OMHM is constantly listening for those heartbeats and when it finds that one is missing it will itself produce and alert and pass that through the list of responders. Just as described above, this could result in emails, pages and operations dashboard alerts, but in this case there is one more responder that kicks in – the “resurector”. The resurector will communicate with the IaaS over which PCF is running and will ask that the failed VM be replaced. Of course it will be replaced with a VM running the appropriate part of the elastic runtime – i.e. a health manager or DEA, etc. That’s right, Operations Manager will restart failed cluster components.
Of course, the BOSH agent on a VM can only communicate back to the Operations Manager if the VM is there, so let’s talk about what happens when a VM disappears. First thing to understand is that by “disappear” I mean that the BOSH agent is not functional; the VM could be there, but Ops Manager no longer knows what it is up to so for all intents and purposes it’s “gone”. How does Ops Manager know? One of the things that a BOSH agent is responsible for is sending out heartbeat messages and by default it does so every 60 seconds. The OMHM is constantly listening for those heartbeats and when it finds that one is missing it will itself produce and alert and pass that through the list of responders. Just as described above, this could result in emails, pages and operations dashboard alerts, but in this case there is one more responder that kicks in – the “resurector”. The resurector will communicate with the IaaS over which PCF is running and will ask that the failed VM be replaced. Of course it will be replaced with a VM running the appropriate part of the elastic runtime – i.e. a health manager or DEA, etc. That’s right, Operations Manager will restart failed cluster components.
Of course, the BOSH agent on a VM can only communicate back to the Operations Manager if the VM is there, so let’s talk about what happens when a VM disappears. First thing to understand is that by “disappear” I mean that the BOSH agent is not functional; the VM could be there, but Ops Manager no longer knows what it is up to so for all intents and purposes it’s “gone”. How does Ops Manager know? One of the things that a BOSH agent is responsible for is sending out heartbeat messages and by default it does so every 60 seconds. The OMHM is constantly listening for those heartbeats and when it finds that one is missing it will itself produce and alert and pass that through the list of responders. Just as described above, this could result in emails, pages and operations dashboard alerts, but in this case there is one more responder that kicks in – the “resurector”. The resurector will communicate with the IaaS over which PCF is running and will ask that the failed VM be replaced. Of course it will be replaced with a VM running the appropriate part of the elastic runtime – i.e. a health manager or DEA, etc. That’s right, Operations Manager will restart failed cluster components.