Service Improvement at UBS

1
Service Improvement at UBS

• Presentation to the Computer Sciences Course,
  University of Warwick
• by William Lees
• Executive Director, Information Technology
• UBS Investment Bank

2
Agenda

• UBS Investment Bank who we are
• Current State of IT Service Provision in
  Investment Banking
• ITIL/CMM - A possible framework for improvement
• Case Study Service Management at UBS
• Challenge
• Conclusion and FAQ

3
Who We Are - http//www.ibb.ubs.com
4
The Stamford, Connecticut Trading Floor
5
Investment Bank IT Organisation - 2006
UBS Investment Banks Global IT landscape
London3,240
Regional totals including staff in locations not
shown on map - EMEA exc CH 3320 -
Switzerland 783 - APAC
2506 - Americas 2857
Stamford 1,506
Frankfurt36
Jersey City15
Tokyo168
Hong Kong248
Zurich756
Chicago344
New York898
Pune247
Singapore719
Bangalore545
Sydney368
Source Total IT PAC Dec 2006 ITI IB included
20,000 UNIX/Linux and Wintel servers 3,000 T
Bytes Storage 35,000 workstations 1,300
Applications
78,000 LAN ports 52,000 telephony ports 800
routers 300 firewalls
9,500 technologists 1.1M help desk calls
p.a. 7,100 market data users 220 sites in 130
cities
... a vast network of technology and technologists
6
Current State
7
The Need for Service Improvement

• In 2005, UBS commissioned a private survey of 10
  financial institutions, focussing on IT service
  delivery, governance, IT spend
• Most respondents highlighted governance and
  organisation as key areas of difficulty
• Often the issues had arisen as a result of
  mergers and acquisitions
• Organizations with a centralized as opposed to
  business distributed model appeared to have
  fewer problems
• 8 out of 10 respondents were implementing a
  business driven one firm initiative

8
Governance Key Concerns

• Many governing bodies, yet little clarity around
  where decisions are made
• Decision outcomes unrelated as differing
  governing bodies do not do a good job of
  communicating with each other
• Governance not strong enough to mandate and
  influence acceptance of agreed outcomes such as
  architectural standards and technology decisions
• Roles like chairperson, office of technology
  strategy, office of standards exist, but little
  clarity on their responsibilities.
• Mergers have played a significant role in
  governance dysfunction. Too many legacy
  committees exist in the new entities they cannot
  rationalize themselves.
• For those organizations that have made changes,
  governing bodies were created simply as a
  rationalization exercise without a holistic look
  at the organization
• Governing bodies are formed to counter other
  governance bodies or processes within the
  organization. For example, multiple technology
  councils, each within different business units,
  were found to be mandating their own technology
  standards without taking into account corporate
  strategy

9
The Impact on IT Service Delivery

• Fragmented IT organizations provide a fragmented
  service users have to contend with multiple
  contact numbers and teams
• Example I need to use AOL Instant Messenger.
  This could require
• Business justification and security approval
• Installation of the software on the users PC
• Training and support
• Modification of firewall rules
• In a fragmented organization, the user might
  have to discover these steps and initiate them
  separately
• Fragmentation makes it difficult to measure
  performance or control cost
• It makes it difficult to respond to changing
  requirements
• It can be self perpetuating departments have to
  do it themselves because of the poor service
  they receive from others

10
ITIL and CMM As A Framework For Improvement
11
The ITIL model

• Information Technology Infrastructure Library
  (ITIL) is a framework describing best practice
  approaches to the delivery of IT services
• It was originally developed by the UK Government
  and is now widely used in commercial
  organisations, particularly in Europe
• Most importantly, it provides a common language
  for discussing IT service provision, and a common
  framework
• The Capability Maturity Model (CMM) provides a
  mechanism for categorizing capability and
  planning improvement. It started as a software
  development initiative (Watts Humphrey, Carnegie
  Mellon University) and has recently been extended
  to IT service delivery.

12
The ITIL Model
13
Important Concepts Service Support
Configuration Management

• Hardware
• Software
• Related documentation
• Relationships between CIs

Configuration Item
Service Desk
Incident Management
Problem Management
Change Management
Release Management
Change
Release
Incident
Problem

• Root cause of related incidents
• When understood, becomes a Known Error

• Significant changes to the IT infrastructure
• Does not include standard or pre-approved
  changes

• Errors
• Outages
• Service Requests

• 1 or more changes built and tested together

14
Important Concepts Service Delivery
Availability Management
Availability
Capacity Management
Capacity

• Reliability
• Maintainability
• Serviceability

Service Level Management

• Business capacity
• Service capacity
• Resource capacity

Service

• Service catalog
• Service level
• Service level agreement

IT Customer
IT Service Continuity Management
Continuity
Financial Management for IT Services
Cost

• Business continuity
• Threats
• Vulnerabilities
• Risks

• Budgeting
• Accounting
• Charging
• Cost types

Service Provider
15
CMM - An Approach to Service Process Improvement
Categorize service maturity in each discipline
Initial (1)
Repeatable (2)
Defined (3)
Managed (4)
Optimizing (5)
People
Process Technology Information
Capabilities are vulnerable to change
Repeatable processes but outcomes are not
consistent
Capable of anticipating and balancing supply
demand
Capable of anticipating and predicting outcomes
Capable of driving strategic outcomes
16
Service Improvement Planning Desired State
Build up a matrix of existing or desired state
see http//itservicecmm.org
17
Case Study Service Management at UBS
18
Why do we need Service Management?

• Can we manage RTB and CTB change efficiently and
  securely? 4,000 changes per month are
  implemented into this environment, and it is
  growing, at a rate of 600 servers pm.
• How do we prevent Business operational outages?
  We need to provide consistent end-to-end
  application availability and cost information, as
  well as performing effective capacity planning,
  in order to future proof the infrastructure.
• Growth Business areas such as Prime Services are
  an aggregator of service offerings from other
  streams, and as such need to be able to deliver a
  consistent and professional service offering to
  their external customers.
• How do we provide a consistent approach to the
  resolution of problems ensuring the minimum of
  business downtime and the elimination of repeat
  incidents through root cause analysis? Move
  from heroics to a professional approach.
• How do we know which components of this vast
  infrastructure landscape are the most critical,
  which have the most dependencies? What routers,
  App servers, racks, DB servers does your
  application depend on?
• In the event of a disaster (eg DC power outage)
  which components do we bring back on line first,
  in order to ensure business critical apps are
  resumed in priority order?
• How do we ensure we operate a Continuous
  Improvement approach, avoiding repeat incidents
  and incorporating technology change?

19
Why do we need Service Management?
1. Number of Incidents (IRAT) by month
2. Outages, hours (IRAT) by month
Legend for Charts 1, 2
3. Outages in July, hours by CO (IRAT)
We need to move from self crafted methods to a
professional and consistent approach to operate
effectively at this scale
20
How will Service Management help?

• Process
• Through the use of defined processes, we are able
  to ensure that all support teams are applying a
  proven methodology for the identification and
  resolution of infrastructure incidents
• Tools (automation)
• Enabling common reporting and metrics, allowing
  the business to feel confidence in the approach
  and ensuring that best practices are replicated
• Provide a scalable approach to infrastructure
  support, which otherwise will risk becoming
  overwhelmed by the growth in business volumes and
  infrastructure (vols up 60, server estate
  growing 50 pa, storage increasing by 70 pa)
• People and Culture
• Allows an end-to-end approach to incident
  resolution we can follow a problem management
  workflow from the triggering incident, through
  trending and analysis, resolution option
  identification, to root cause resolution from the
  application through to all components of the
  infrastructure
• Consistent processes and tools enable staff
  mobility across teams

A Service Management culture needs to become a
way of life
21
What does the Service Mgt roadmap look like?
SLM Release Mgt Change Mgt Problem Mgt Incident
Mgt Ticketing/Service Desk
Configuration Mgt
4
roadmap to maturity
Advanced/World class

• Advanced automation supporting releases into
  production and linkage to monitoring tools for
  proactive problem resolution methods

3

• 2008 onwards

Intermediate stage

• Common Incident, Problem, Change and Release
  Management processes built around an integrated
  CMDB

2

• 2006 - 2007

Early stage

• Use of common ticketing system and incident type
  definitions, introduction of Problem Management
  processes

1

• 2004 - 2006

No Common strategy

• IB IT teams operating ad-hoc incident and problem
  mgt processes

• Pre-2004

A multi-year journey that does not have
short-cuts
22
Benefits

• Reduction in business outages through elimination
  of repeat incidents, faster problem resolution
  and effective root cause analysis
• Faster business resumption in the event of a
  major infrastructure disaster (DC power outage,
  major network failure, virus attack/DoS)
• Improved resource utilisation, providing more
  effective resource management and reduction in
  resource scaling required to support application
  and infrastructure growth
• Enables the offshoring or automation of routine
  GSD and infrastructure workload, through
  repeatable processes and improved reporting and
  metrics
• Proactive lessons learnt and best practice
  sharing, culture of continuous improvement
• Hard Benefits 44 FTEs, 4.9M per annum savings
  (2008 on)
• Intangible Benefits Improved Operational Risk
  management, greater Cost Transparency, reduced
  Change Management issues, improved consistency
  between divisions

23
Challenge / Conclusion
24
Your Challenge

• Explain to a business user, who doesn't really
  understand IT but ultimately has to pay for it
• what ITIL is all about
• how it differs from a more conventional approach
  to service management
• why service improvement is important
• what tangible benefits its adoption will provide
  to the business as a whole

25
Conclusion

• UBS IB is a large graduate / postgraduate
  employer. We have a popular summer internship
  programme for students about to enter their final
  year or postgraduate year
• All information and application forms are on our
  website http//www.ibb.ubs.com