Present

1
Present Future Directions

• Sam Carothers
• Strategic Directions Manager
• JCPenney

2
The Role of Strategic Directions

• Maintain an awareness of present and future IT
  related business needs
• Stay current on existing and emerging
  technologies
• Develop practical IT based solutions to our
  business problems
• Where possible, assure that solutions leverage
  existing investments
• Assure that solutions are congruous, and
  complementary to our overall direction

3
Projects Currently Underway

• Enterprise Management
• Storage Area Network / Storage Consolidation
• Unix and NT Server Consolidations

4
Enterprise Management
What is it?

• When we refer to Enterprise Management, we are
  talking about the ability to do end-to-end
  management of the entire IT infrastructure from a
  central location.
• The ideal implementation of Enterprise Management
  would allow us to perform any Management
  Discipline on any IT resource, whether that
  resource is a Network, a Mainframe, a Midrange, a
  Server, an Application, a Database, Middleware,
  the Internet, or any other type of resource.

5
Enterprise Management
Examples of Management Disciplines...

• Output Management
• Performance Management
• Problem Management
• Remote Control
• Recovery / Rerun / Restart
• Security Management
• Service Level Reporting
• Single sign on
• Software Distribution
• Software Inventory Management
• Storage Management
• User Account Management
• Workload Scheduling

• Backup Recovery
• Capacity Planning
• Change Management
• Configuration Management
• Data File Transmission
• Database Management
• Directory Management
• Disaster Recovery
• Event Management
• Hardware Inventory Management
• Installation Management
• Internet / Extranet Management
• License Management

6
Enterprise Management
Tactical Directions and Actions

• Our tactical direction is to expand the
  management of the Unix and NT platforms by
  implementing the same Operational Model used to
  manage the Mainframe and Network platforms today

7
Enterprise Management
Tactical Directions and Actions - Phase One

• We have identified five Management Disciplines as
  critical building blocks for implementing the
  Operational Model for Managing Unix and NT.
• Event Management
• Remote Control
• Change Management
• Workload Scheduler
• Backup and Recovery

8
Enterprise Management
Tactical Directions and Actions - Status

• A review of packaged Framework type products is
  complete
• Findings Framework Solutions are expensive,
  difficult to deploy, and have high deployment
  failure rates among adopters
• We have developed requirements documents for the
  Management Disciplines identified for Phase One
  and have presented those requirements to our
  Systems Architectures department
• The Systems Architectures department will make a
  build or buy decision for each Management
  Discipline

9
Enterprise Management
Event Management

• Distributed monitoring that detects exceptions
  and either addresses the problems locally or
  escalates the problem to an enterprise console
  where further automation can be performed.
  Automation, in this case, could either include
  resolving the problem or opening a problem ticket
  and assigning it to the appropriate support
  group.
• Status
• Internal development of an Enterprise Console /
  Event Management Console was underway but is now
  on hold while people resources are deployed on
  other projects
• BMCs CommandPost product has undergone
  evaluation and is now involved in a Proof of
  concept

10
Enterprise Management
Remote Control

• Used to enable support associates access the
  servers for problem determination and problem
  resolution.
• Status
• 5000 Timbuktu licenses have been purchased to
  enable remote access to servers and desktops
• The product has been deployed to the Corporate
  Servers
• It is being deployed to desktops as needed.
  (Currently on the Standard Merchandising desktops)

11
Enterprise Management
Change Management

• As problem determination responsibilities move to
  the Command Center it will be important to have a
  Change Management process in place that will
  enable Problem Solvers to view what changed in
  the environment. Fully mature Change Management
  also implies integration with Software
  Distribution and Software Inventory type
  capabilities.
• Status
• Work to integrate CRISS and SDS is underway and
  is expected to be complete before the end of
  1999.
• This will enable us to put the Change Management
  of Mainframe, Unix and NT under one umbrella.
• CRISS - An internally written Change Management
  System
• SDS - An Internally written Software
  Distribution System

12
Enterprise Management
Workload Scheduler

• This is not needed everywhere but is needed for
  specific servers that have or will have complex
  scheduling needs. The product selected will
  replace the traditional Chron type scheduling
  on Unix and NT
• Status
• We have purchased 60 copies of the Tivoli Maestro
  product (50 Unix and 10 NT)
• Nine copies have been rolled out. Maestro is
  installed on eight Openview servers and on one
  Peoplesoft server
• Workload Management responsibilities for servers
  is now in the ITCC

13
Enterprise Management
Backup and Recovery

• Replace existing point solutions with a centrally
  managed and fully automated backup system that
  will keep copies of data onsite for quick
  recovery and offsite for business continuity.
• Status
• 300 ADSM licenses have been purchased in order to
  perform backups on the Unix and NT platforms.
• Over 200 Dallas servers are being backed up and
  vaulted. Dallas is complete.
• Plans are under development to roll out this
  solution to Columbus, Lenexa and Reno in 2000.
• Server Backup responsibilities are now in the
  ITCC (for those servers backed up by ADSM).

14
Storage Area Networks
What and Why?

• A SAN is a high performance Network built using
  Storage protocolsWe are reviewing Fibre Channel
  based solutions
• Why
• To enable storage consolidations
• To enable better storage management
• To enable faster backups and business continuity
• To offload backup traffic from the network
• To increase availability of the storage
• To address distance limitations
• To enable greater flexibility in disk allocations

15
Storage Area Networks
Status

• We have done a considerable review of technology
  offerings
• Findings This is a rapidly evolving area. Many
  early adopters (early 1999) experienced
  interoperability issues between HBAs, Switches,
  and Storage. Many of those problems are now
  resolved.
• We are currently bringing in two different
  solutions to evaluate their value, effectiveness
  and ease of use.
• Combination of HDS disk and Ancor Switches
• Combination of EMC disk and McData Director

16
Unix and NT Consolidations
Background

• As the popularity of Unix and NT systems
  increased, many end users established systems in
  their own departments that were outside the
  control of IT
• Over time, most end user departments that
  installed their own servers have requested that
  IT take on the management of the servers
• The process of relocating the Unix and NT servers
  has been underway for several years
• The vast majority of application servers that
  were in end user locations are now in the glass
  house

17
Unix and NT Consolidations
Status

• Physically consolidating the Unix and NT servers
  has enabled IT to begin applying more standard
  policies and management techniques to these
  servers
• As Server Management tools improve it has further
  enabled us to apply our Operational Model to the
  Servers
• As the technologies improve, we will be able to
  begin consolidating workloads on large enterprise
  class Unix and NT systems
• There are very few success stories for combining
  unlike workloads on Unix and NT

18
Areas Under Investigation

• The following areas are under investigation
  because they are emerging technologies that
  address existing or anticipated business needs.
• These technologies are at different stages of
  development, but none are fully mature yet.
• Unix and NT Clustering
• Remote Backups / Remote Mirroring
• Virtual Tape

19
Unix and NT Clustering
What and Why?

• Clustering refers to setting up a group of
  independent systems so that they can function as
  a single system
• Why
• To reduce unplanned downtime
• To reduce planned downtime
• To increase manageability
• To increase scalability
• To reduce the total number of platforms under
  management
• Status
• Technologies are under investigation
• No Production implementations yet

20
Remote Backup / Remote Mirror
What and Why?

• Remote Mirroring and Remote Backups refer to the
  ability to mirror or copy data from one
  geographic area to another
• Why
• To reduce backup times
• To reduce risk / enable Business Continuity
• As an enabler to clustering
• To enable planned workload migrations between
  locations
• Status
• Solutions from EMC, IBM, HDS, Compaq, Veritas and
  others are under review
• No Production implementations yet

21
Virtual Tape
What and Why?

• Virtualization of tape will us to break the old
  paradigms of One tape mount to one tape
  transport and one tape cartridge equating to
  one tape volume
• Why
• To reduce the numbers of robotics libraries
  needed
• To reduce physical floor space requirements
• To reduce Operations labor
• To reduce tape job runtimes
• To enable Business Continuity by using less tapes
  for vaulting
• Status
• We are investigating solutions from STK, IBM,
  Sutmyn Sterling
• No Production implementations yet

22
end

• Sam Carothers
• Strategic Directions