Operational Excellence in a Cloud World

Operational Excellence in a Cloudy World

www.rackspace.com

WAYNE WALLS Director, Cloud Strategy Rackspace

!2

RYAN RICHARD Team Lead - DevOps Automation Service Rackspace

Presenters

http://www.rackspace.com/

www.rackspace.com

RACKSPACE VISION

To be recognized as one of the world’s great service companies.

!3


www.rackspace.com

Managed Cloud

• A New Category

!4

Managed cloud is the offering for customers who want to tap the power of cloud computing without the pain and expense of managing it themselves. !“We’re in it with you every step of the way.”


www.rackspace.com

Gartner Validates Managed Cloud Market

Cloud-Enabled Managed Hosting

Magic Quadrant July 2014

!5

ABILI

TY T

O EX

ECUT

E

COMPLETENESS OF VISION As of July 2014

CHALLENGERS LEADERS

NICHE PLAYERS VISIONNARIES

Rackspace

Datapipe

Verizon Terremark

AT&TNaviSite

Layered Tech

Virtustream

Peak 10Latisys

Hosting CenturyLinkFireHost

Dimensior DataCSC

IBM

NTT Communications

Sungard Availability Services

2014

This graphic was published by Gartner, Inc., as part of a larger research document and should be evaluated in the context of the entire document. The Gartner document is available upon request from Rackspace. Gartner does not endorse any vendor, product, or service depicted in its research publications, and does not advise technology users to select only those vendors with the highest ratings. Gartner research publications consist of the opinions of Gartner's research organization and should not be construed as statements of fact. Gartner disclaims all warranties, expressed or implied, with respect to this research, including any warranties of merchantability or fitness for a particular purpose.


www.rackspace.com

All Backed by Fanatical Support®

!6

It’s way more than raw infrastructure. It’s the one-on-one expert guidance you need—from launch, to ongoing security guidance—whenever you need it, 24x7x365. We call it Fanatical Support®. And we’ve been doing it for over 15 years. It means we’re in it with you every step of the way, to deliver a best-fit environment for your app or workload.


www.rackspace.com

Lets talk apps…

!7


www.rackspace.com

POLL:

How would you rate your production application operations today?

!8


www.rackspace.com

•Responsible for stability, reliability and scalability to our Cloud products!•Identifying and documenting all products, systems and their interactions!•Identify systems prone to incidents!•Communicate major incidents and help to ensure resolutions are integrated into QE!

•Define long term best practices for all products and systems

Product Operations Reliability Team

!9


Level 1: Ad Hoc Level 2: Standardizing Level 3: Industry Standard Level 5: Industry Leader Level 8: World Class

Llama! Logging, Alerting, Monitoring

Standards

Architectural Operability

Recoverability

Automation

Documentation

• Lorem ipsum dolor sit amet, consectetur adipiscing elit.!• Partim cursu et peragratione laetantur, congregatione aliae

coetum quodam modo civitatis imitantur; !• Traditur, inquit, ab Epicuro ratio neglegendi doloris. !• Qua tu etiam inprudens utebare non numquam. !• A mene tu? Quamquam ab iis philosophiam et omnes

ingenuas disciplinas habemus; !• Sine ea igitur iucunde negat posse se vivere? !• Duo Reges: constructio interrete. !• Quod vestri non item. Dulce amarum, leve asperum, prope

longe, stare movere, quadratum rotundum.





















































































































Ops Scorecard

www.rackspace.com

Ops Scorecard Progression

!11

Diff

icul

ty

Progression1 2 3 5 8

Most Companies


Level 1: Ad Hoc Level 2: Standardizing Level 3: Industry Standard Level 5: Industry Leader Level 8: World Class


Standards


Recoverability

Automation

Documentation


























































































































AR

CH

ITEC

TUR

AL

OPE

RA

BIL

ITY • Stack running on a single server under a desk, or maybe a single

cloud server!• Passwords stored in plaintext!• Manual user management!• Local user management, no directory services!

• Often primarily developer-driven

Level 1: Ad Hoc

AR

CH

ITEC

TUR

AL

OPE

RA

BIL

ITY • Application runs on external platform services!

• Operations team involved in planning!• Backend development is prioritized!

• Shared accounts and shared access!• Beginning the separation of stateful and stateless services

Level 2: Standardizing

AR

CH

ITEC

TUR

AL

OPE

RA

BIL

ITY • Capable of running on any Infrastructure-as-a-Service provider!

• Operations team consensus on backend system deployment!• Continuous Integration required prior to deployment!• Bastions or “jump-servers” used to access production environment!• Aided by centralized access!• SSH key management!

• Dynamic Networking (Software-defined networking)!• Regional redundancy

Level 3: Industry Standard

AR

CH

ITEC

TUR

AL

OPE

RA

BIL

ITY • Data dictates the technology used, not the opposite!

• Fine grained access controls!• Clear separation of roles and responsibilities!

• Consolidated centralized user access management!• Full audit trails with logging!• Early adoption of emerging technology researched and woven into

solutions where applicable

Level 5: Industry Leader

AR

CH

ITEC

TUR

AL

OPE

RA

BIL

ITY • Systems are self-healing !

• Zero customer impact!• Geo-redundant systems!• Manual (i.e. human) node/system interaction is rare!• “Unauthorized” changes trigger automated response!• Removing nodes from load balancers!• Replacing worker nodes automatically!

• Service discovery!• Zookeeper!• etcd

Level 8: World Class

Level 1 Ad Hoc Level 2: Standardizing Level 3: Industry Standard Level 5 Industry Leader Level 8: World Class


Standards


Recoverability

Automation

Documentation

























































































































AU

TOM

ATIO

N • Manual deployment!• Zero configuration management!• Commands run directly on command line

Level 1: Ad Hoc

AU

TOM

ATIO

N • Follow runbooks / wiki build instructions!• Use bash or other scripts copied manually to nodes for job

execution!• Manual tool run via SSH!• “Fire & Forget”!

Level 2: Standardizing

• Tool is auto-run when something breaks!• Auto deployment post continuous integration!• Continuous integration tool has deployment capability

AU

TOM

ATIO

NLevel 3: Industry Standard

• One-click deploy!• Adding/removing nodes is fully automated with configuration

management!• Monitoring is baked into config management systems!• Local access to nodes is restricted

AU

TOM

ATIO

NLevel 5: Industry Leader

• Deploys new version of application in a continuous rolling upgrade manner!

• Full root-cause analysis (RCA) at the application level; operations does not band-aid solutions!

• Service discovery paired with configuration management for node deployment and application release management

AU

TOM

ATIO

NLevel 8: World Class

www.rackspace.com

POLL:

How would you rate your production application operation now?

!25


RACKSPACE® | 1 FANATICAL PLACE, CITY OF WINDCREST | SAN ANTONIO, TX 78218 US SALES: 1-800-961-2888 | US SUPPORT: 1-800-961-4454 | WWW.RACKSPACE.COM

© RACKSPACE LTD. | RACKSPACE® AND FANATICAL SUPPORT® ARE SERVICE MARKS OF RACKSPACE US, INC. REGISTERED IN THE UNITED STATES AND OTHER COUNTRIES. | WWW.RACKSPACE.COM

THANK YOU

Operational Excellence in a Cloud World

Technology

tu etiam

mene tu

consectetur adipiscing

cloud products

leve asperum

dulce amarum

duo reges

constructio interrete