• No results found

Agenda. Where are we? Where do we need to go? How does PaaS get us there?

N/A
N/A
Protected

Academic year: 2021

Share "Agenda. Where are we? Where do we need to go? How does PaaS get us there?"

Copied!
49
0
0

Loading.... (view fulltext now)

Full text

(1)

About Me

Red Hat Cloud Evangelist Twitter: @ghaff

Google+: Gordon Haff

Email: ghaff@redhat.com Blog: http://

bitmason.blogspot.com

Formerly: Illuminata (industry analyst), Data General

(minicomputers/Unix/NUMA/ etc.)

(2)

Agenda  

Where  are  we?  

Where  do  we  need  to  go?  

How  does  PaaS  get  us  there?  

(3)

LANGUAGES  

(4)
(5)
(6)
(7)
(8)

Much  of  my  work  has   come  from  being  lazy.  I   didn't  like  wriAng  

programs,  and  so,  when  I   was  working  on  the  IBM   701,  wriAng  programs  for   compuAng  missile  

trajectories,  I  started  

work  on  a  programming   system  to  make  it  easier   to  write  programs.  

(9)
(10)

•  It  should  be  "simple,  object-­‐

oriented  and  familiar"  

•  It  should  be  "robust  and  

secure"  

•  It  should  be  "architecture-­‐

neutral  and  portable"  

•  It  should  execute  with  "high  

performance"  

•  It  should  be  "interpreted,  

(11)

11

(12)

Top  10  Programming  Languages  

•  JavaScript

•  Java

•  PHP

•  Python

•  Ruby

•  C#

•  C++

•  C

•  Objective-C

•  Perl

•  Source:  hOp://redmonk.com/sogrady/2013/02/28/language-­‐rankings-­‐1-­‐13/#ixzz2ecQupc6S  

(13)

SOURCE CONTROL

SYSTEMS (TOOLS)

(14)

Networking:      

None  

OperaAons:            

One  file  at  a  Ame  

Concurrency:  

(15)

Networking:      

Centralized  

OperaAons:              

MulA-­‐file  

Concurrency:    

Merge  before  

commit  

(16)

16

Networking:      

Distributed  

OperaAons:              

Changesets  

Concurrency:    

Commit  before  

merge  

(17)

ARCHITECTURES  

(18)
(19)

19

Source: http://www.slideshare.net/adrianco/gluecon-tutorial-details-2013 Adrian Cockcroft

(20)

http://www.flickr.com/photos/hen_riko/8627290473

(21)

DEVELOPMENT  

METHODOLOGIES  

(22)
(23)

23

Esti Alvarez/flickr under CC

(24)

VELOCITY  OF  RESOURCE  

ACQUISITION  

(25)

25

“The average time for

setting up a new server used

to be 30 days. Now the worst

case is 12 days.”

(26)

VirtualizaAon:  An  iniAal  flexibility  step  

26

Server

Virtualization Virtualization Distributed

Visibility Control

Optimization Automation

l  Consolidation

l  Reduce Capital Expense

l  Flexibility & Speed

l  Reduce Operational Expense

l  Automation

l  Less Downtime

D

rive

rs

(27)

27

(28)

Cost per VM Hour Server: Admin Ratio Resource Utilization App Deployment

New App Development

1x 300: 1 75% Minutes vs. vs. vs. vs. vs. 10x 20: 1 20% Weeks Years

PUBLIC CLOUDS TRADITIONAL ENTERPRISE IT

Weeks/ Months

PERVASIVE NEW EXPECTATIONS

AGILITY. EFFICIENCY. COST SAVINGS.

(29)

Cloud Service Models

STORAGE (RHS) HARDWARE (x86) VIRTUALIZATION (RHEV) OPERATING SYSTEM (RHEL) APPLICATION PLATFORM

(JBOSS, PHP, RUBY, ETC)

APPLICATION

Automated and Managed by the Public or Private Cloud Offering

Managed and Controlled by Customer (IT, Dev, or User)

IaaS PaaS SaaS

Increased Control

(30)

Why  PaaS  now?  

•  Web-­‐friendly,  abstracted  languages  

–  Java,  PHP,  Python,  Ruby,  etc.  

•  Modern  open  source  tools  

–  Distributed  source  control  (git)  

–  Maven,  Jenkins,  Eclipse,  etc.  

•  “Agile”  development  pracAces  

–  More  speed,  more  iteraAon  

•  “Cloud”  applicaAon  design  paOerns  

–  Loosely-­‐couples,  service-­‐oriented  

(31)

Fundamental  accelera@on  

of  business  

(32)

Manufacturing as a Model for IT

Consolidated Aircraft B-24 Liberator

Incredibly sophisticated. ~500k parts, assembled by unskilled labor.

No Mfg process. Parts were cast in rubber molds, so every part was slightly different. Assembled in the heat of San Diego, which warped the metal and required whole

assemblies to be adjusted.

Ford Motor Co. brought a Manufacturing

process … went from 250 planes a year to

650 planes a month.

About re-engineering the complete process.

(33)

Streamlining App Dev with PaaS

With PaaS

How to Build an App:

1.  Have Idea 2.  Get Budget

3.  Code

4.  Test

5.  Launch

6.  Automatically Scale

How to Build an App:

1.  Have Idea 2.  Get Budget

3.  Submit VM Request request

4.  Wait

5.  Deploy framework/appserver

6.  Deploy testing tools

7.  Test testing tools

8.  Code

9.  Configure Prod VMs

10.  Push to Prod

11.  Launch

12.  Request More Prod VMs to

meet demand 13.  Wait

14.  Deploy app to new VMs

15.  Etc.

Virtualized

How to Build an App:

1.  Have Idea

2.  Get Budget

3.  Submit hardware acquisition request

4.  Wait

5.  Get Hardware

6.  Rack and Stack Hardware

7.  Install Operating System

8.  Install Operating System Patches/

Fix-Packs

9.  Create user Accounts

10.  Deploy framework/appserver

11.  Deploy testing tools 12.  Test testing tools

13.  Code

14.  Configure Prod servers (and buy them if needed)

15.  Push to Prod 16.  Launch

17.  Order more servers to meet demand

18.  Wait…

19.  Deploy new servers

20.  Etc.

Physical

“The use of Platform-as-a-Service technologies will enable

IT organizations to become more agile and more responsive to the business needs.” –Gartner*

(34)

Good  fences  make  

good  neighbors  

(35)

Walls  tradiAonally  hindered  

(36)

DevOps  about  tearing  down  wall  

Dev2ops.org

(37)
(38)

How Can I Consume OpenShift?

STORAGE (RHS) HARDWARE (x86) VIRTUALIZATION (RHEV) OPERATING SYSTEM (RHEL) APPLICATION PLATFORM

(JBOSS, PHP, RUBY, ETC)

APPLICATION Public Cloud Service On-Premise or Private Cloud Software Developer Controls Developer Controls Operated by Red Hat at Scale for 18

Months IT Ops

Provides OpenShift Automates, IT Ops Controls

(39)

Developers  

•  Abstracts  away  

irrelevant  plaform   details  

•  Provides  self-­‐

service  

•  Reduces  Ame  to  

code  

•  Increases  

producAvity  

•  Supports  familiar  

tools  

Nelson Pavlosky/flickr under CC

(40)

Developers Choose Languages, Frameworks and Middleware

(41)

OperaAons  

•  Delegates  and  

automates   provisioning  

•  Automates  tasks  

such  as  scaling  

•  Improves  IT  

service  levels    

•  Provides  control  

over  

development   environments  

Leonardo Rizzi/flickr under CC

(42)

Efficient  &  Secure  MulA-­‐Tenancy  

Linux Linux

Super-efficient, secure, multi-tenant containerization model using:

SELinux, Control Groups, Namespaces

Broker Node Node Node

Linux

(43)

Automates Application Scaling

Broker Node Node Node

RHEL

HA-Proxy Java MySQL

Cod e

Java Cod e

Java

Code

(44)

Architects  

•  Enables  

standardized   development   workflows  

•  Reduces  Ame  to  

develop  

applicaAons  

•  Increases  developer  

producAvity  

401(K) 2013//flickr under CC

(45)

Helps IT Architects Streamline & Automate Developer Workflows

RHEL RHEL

A Developer creates a

new application à

OpenShift creates a GEAR

Web Console Eclipse IDE Cmd Line

Broker Node Node Node

RHEL

OpenShift Gear

(46)

Buyers  

•  Decouples  plaform  

from  applicaAon   development  

•  Enables  plaform  

changes  to  be  made   independently  based   on  economic  and  

other  factors  

Will Scullin/flickr under CC

(47)

Code Deploy Enjoy

PaaS leverages automation technologies and a cloud architecture…

…to drive Velocity, Efficiency, and Scalability in IT

(48)

Get Started Today for Free!

http://openshift.redhat.com

•  Deploy Apps to the OpenShift OnLine Developer Preview

•  Request an Evaluation of OpenShift Enterprise

(49)

THANK YOU!

QUESTIONS?

Gordon Haff

@ghaff

References

Related documents

To take advantage of this resource, we would develop a tutor training program for community members or current college students who wish to volunteer in basic skills classes at

After an SAH , blood vessels near the burst aneurysm can go into spasm and prevent blood from getting to the brain. Why this happens isn’t clear, but a lack of blood can lead to

• the designer’s knowledge is limited, and/or • the sheer complexity and variability precludes an..

● Estimated Parameters ● SAS/PROC GENMOD ● Example 2: Lung Cancer ● Plot of the Rate by Age ● Plot of the log(Rate) by Age ● Model 1: Age and City both Nominal. ●

Using those behavioural and spatial cognitive tests, we found that Eurasian harvest mice behaved constantly and that personality traits formed a behavioural syndrome in the

In order to solve theses problems, first, new location reference system for the National Highways was introduced by installing kilometer posts which were designed to have

Our findings showed that the HLS made a large number of contextually inappropriate substitutions by using direct evidentials in places where an indirect evidential would

The wisdom that lies behind this quote started a project for finding best integrated Busi- ness Intelligence (BI) solutions in the market. Progress of this project is documented in