How Maximum Availability Transforms Our Business Tina Mikula, - - PowerPoint PPT Presentation

1

xMatters (alarmpoint) client training forum

October 5 – 6, 2010, Manhattan

How Maximum Availability Transforms Our Business

Tina Mikula, Global Technology Services Analyst Peggy Steinbrecher, Global Service Delivery Analyst Manpower

CLIENT CASE STUDY - MANPOWER

2

Manpower eliminates 80 hours of downtime

3

Manpower is the one of the world’s largest providers of temporary employees

• Placing over 2 million people in various industries across the

globe for 400,000 clients per year

Manpower has a global network of over 30,000 staff members in 4,000 offices in 82 countries and territories Based in Milwaukee, WI Most critical business process:

• Batch payroll jobs
• Application and asset monitoring

WHO IS MANPOWER

4

Homegrown system:

• Previous system was too manual
• Could take 20 minutes for support team to respond

PREVIOUS SYSTEM

5

Downtime was a huge problem:

• Experienced 293.34 hours of downtime in the year before

xMatters deployment:

• Manpower’s reputation with client was on the line

Required a notification solution that could integrate with multiple systems:

• With four different technologies running in our environment,

we required a solution that could seamlessly communicate with all of them

ISSUES WITH PREVIOUS SYSTEM

6

Evaluate Current

• Our environment was a patchwork of home-grown solutions that did

not work with each other

Determine Need

• We needed an automated solution for critical batch payroll jobs to

streamline the process and keep the possibility of human error to a minimum Implement

• By utilizing xMatters, we have been able to combine four different

notification engines into one simple tool Expand

• We plan to expand the number of notification engines feeding

xMatters so that no matter where the notification originates, all notifications sent will have the same look and feel – making things straightforward for the end user

7

• Ability to integrate
• Currently have four different tools

feeding notifications

• Alert teams prior to service impact
• Alerts needed to be answered before batch jobs were

impacted

• Minimize downtime
• Self service
• Users manage their own schedules and shifts
• Easy customization for unique teams and situations

WHY XMATTERS

8

We used our existing on-call list as a starting point to build groups and teams

Because our current job scheduler used 3-letter codes to identify and associate alerts with the correct on-call group, we simply adopted that architecture for all integrations.

HOW WE DID IT

9

The project team did the initial configuration of groups and users, and then had the teams themselves manage the information from there. We created a self-service help document that would show end users how to do things relevant to them:



On-Call Notifications and Replies ~ 3

• Mercury and OGC ~ 5
• Control-M ~ 6
• Alternate reply method – Logging In ~ 7



Schedules ~ 8

• Adding Users to Your Schedule ~ 11
• Escalation Order ~ 13
• Rotation Schedule ~ 14
• Replacements ~ 15



Adding Devices ~ 18



Who’s on Duty? ~ 23



Subscriptions ~ 25



Sending Messages ~ 28



Adding Users ~ 30



Definitions and Descriptions ~ 32



Configuring Mobile Devices ~ 36

This is continually updated and made available to all end users via knowledge portal

WHAT HAVE WE LEARNED: IT HAS TO BE MONITORED TO BE MANAGED

10

Example from our self written training materials:

HOW WE DO IT

11

Example from our self written training materials:

We held training for all support team managers and members

TRAINING

12

Targeted notifications:

• IT staff only receives alerts that are relevant to them

Automation:

• xMatters will catch issues prior to business impact and start

working to resolve the issue Seamless integration:

• Several applications are able to communicate effectively

Minimized downtime:

• Batch payroll jobs run on-time, ensuring employees are paid

around the world

CURRENT PROCESS

13

Before xMatters After xMatters Time to notify staff of event 15-20 Minutes 3 Minutes Time from alert to accepting ticket 20-30 Minutes 10-20 Minutes Global applications availability 99.41% 99.72% Hours of downtime per year 293.34 Hours 213.35 Hours Our average response time from support has decreased by 85% Our Global application availability has increased by 31% and 80 hours

RESULTS

14

• Home telephone
• Privacy concerns prompted a change in the types of

phones that were listed in a user’s devices

• Generic accounts
• Due to security concerns, we wanted to restrict the

number of generic accounts able to log on to the system, but we still needed the ability to have a generic user to notify

• SMS text
• No SMS solution leads to problems notifying users who

can only get a text message

• We have no standard for mobile devices for offshore
• contractors. As a result, message delivery to them cannot

be guaranteed

PAIN POINTS

15

What’s next?

• Mobile access
• We need to get the information people to need to the

people that need it as quickly as possible. Use the technologies available to us to get that done.

• Use more monitoring tools
• We need to better utilize our already existing tools

and get them to all work together.

• Expansion to more users
• We want to include our Global community so that all

applications and systems get the same level of

• support. Our end users expect it.

FUTURE PLANS

16