Information Systems

Five-Year Plan, FY 1999-2003

Table of Contents Information Systems Information Systems Practice Summaries

Information Systems FY 1999 Schedules

I. Mission

Work in Information Systems is centered on opening new pathways for MIT's core missions of education, research, and service, and is focused on six strategic goals:

Organization

Work in Information Systems is organized around a framework with three explicit elements: IS's customers, its work, and the skills of its staff.

process work flow diagram

The three information technology practices - Academic Computing, Office Computing, and Voice, Data, and Image Networking - focus on customers, advocating on their behalf to Information Systems and on behalf of Information Systems to them. The practices promote and enable technology-based work at MIT. They assist the Institute community in identifying information technology needs and opportunities to use technology in education, research, and administration. The practices also assist in planning for the effective use of these resources.

As a process-centered organization, the information technology work of developing and operating products and services is the responsibility of five I/T processes - Integration, Discovery, Delivery, Service, and Support.

The Integration Process has the responsibility for the information technology infrastructure at the Institute. This infrastructure encompasses the hardware and software components underlying MIT's cross-organizational and Institute-wide academic and administrative applications and related systems. As part of its responsibilities, the Integration Process serves as an advocate for the infrastructure (often referred to as the "commons";) to IS and to the Institute community.

The Discovery Process collaborates with customers from academic, research, and administrative areas to identify ways to leverage information technology for business benefit - increased value, reduced work, and lower costs. Discovery work is done in teams populated by representatives of the business, its customers, and information technology staff from across the Institute. A discovery team's work product is a business case and conceptual design for a new I/T application. Once approved, that design moves to the Delivery Process where a team acquires or builds that application and readies it for use in the Institute community. Both the Discovery and Delivery Processes are influenced by the three I/T practices and the Integration Process.

The Service Process has the primary responsibility for operating and renewing the information technology assets of the Institute. These assets include the servers in the Data Center, the telephone system, the Athena clusters, the communications wire and cable plant, and the computer network. Among Service's responsibilities are the identification of and action on opportunities to improve efficiency and customer service. The Service Process also has responsibility for business continuity in these areas.

The Support Process promotes the widespread effective use of existing I/T products and services by the community. Its focus is on the effective use of standard products and services. To this end, it provides help to the community through training, documentation, and a help desk.

The third focus of the IS organizational framework is staff. Work in this area is the responsibility of the Staff Development and Resource Management Team (formerly known as the Competency Group Team). This team's goals are two-fold:

The IS evolution from a traditional functional organization to the process and team-based network described above has taken longer than anticipated. However, recent experience has begun to show the benefits of this new design. We are now seeing more effective application of MIT's I/T resources through increased staff mobility and concentrated team focus on I/T issues.

II. Accomplishments

Consistent, reliable I/T services

Partnerships with customers for solutions

Partnerships with vendors for products and services

Employees as key resources Policies, procedures, and standards Effective use of resources

III. Strategic Priorities

For the period Fiscal Year 1999 - 2003, Information Systems will focus on three strategic objectives consistent with the goals stated in our mission:
  1. Ensure that MIT processes use modern information technology effectively.
  2. Grow Information System's commitment to customer service and operational excellence.
  3. Continue to develop and implement clear goals, vision, principles, roles and responsibilities, process and subprocess designs, and performance metrics to enable the effective use of the Institute's information technology resources.
To accomplish these objectives, Information Systems will continue to pursue the broad outlines of many program priorities that have been described in earlier plans. This plan includes a number of new program priorities for IS, many of which can be accommodated within IS's current budget parameters.

Other new priorities cannot be so accommodated, however. These include ramping up for the educational computing initiatives envisioned by the Council on Educational Technology; upgrading our network infrastructure on- and off-campus to accommodate both normal growth in network traffic and the Council's initiatives; supporting the Alumni Network Services initiative; and recognizing the on-going cost of implementing and maintaining SAP.

In part, the predominant driver for this set of new priorities is the continued growth in the capabilities of desktop computers at a constant or decreasing price. This phenomenon is primarily the result of Moore's Law - the number of devices on a microchip will double in number approximately every eighteen months as the chip halves in price - and a similar recognition by Jerry Saltzer that the price of disk memory decreases by a factor of two every 12 months. Take for example the changes in the Macintosh computer since its introduction in 1983. In 1983, its processor was the Motorola 68000 running at 8 Mhz; it had 128 Kbytes memory, no hard drive, and a 9-inch monochrome screen. The machine cost about $2000. Today, for a similar price in constant dollars, you can buy a Macintosh with a Motorola G3 processor running at 266 Mhz, with 64 Mbytes of memory, a 4 Gbyte hard drive, and a 17-inch color monitor display. Today's machine represents an increase of over two orders of magnitude in performance over the 15-year period. Similar increases in performance at a constant price are anticipated to continue for at least one more decade and likely for two.

Academic Computing

The Council on Educational Technology's proposed initiatives are based largely on the possibilities of today's technology and the promise of tomorrow's. During an undergraduate's four-year tenure at MIT, we can anticipate improvements in hardware performance of a factor of about five. Such performance improvements pave the way for waves of new educational applications, as well as new ways of learning, with successive generations of technology. To prepare for the full implementation of the Council's recommendations (which will require major funding), we propose three start-up initiatives:

  1. Initiate a technology discovery project to examine how we might move from our present infrastructure to one that will support the initiatives proposed by the Council. This project would take a first look at the network infrastructure, the required system services, the application development environment, and a prototypical toolkit; it would also produce a draft technical plan and a process for changing the plan as more is learned. The project would be staffed with a small group of technical experts who would be expected to complete their work within a three-month period of time.

    This work will be done in the second half of FY 1998. Funding may be required if it is necessary to backfill the operational responsibilities of staff assigned to this task.

  2. Provide a core staff of three to begin to develop and provide a set of common tools and services for use in the next generation of educational initiatives, and to provide support for the initial explorations. Strengthening support for educational development is consistent with a direction first recommended by the Committee for Academic Computing in the 1990s and Beyond, and more recently by the Council. Development of initial tools and services must precede the development of actual educational explorations.

    This proposal requires additional funding for salary, employee benefits, and additional software for these staff.

  3. Provide two additional staff to begin development and operation of the additional services - higher bandwidth on- and off-campus, network drops for serial reuse across the campus, wireless services, etc. - required for the initial phases of the Council's proposed initiatives.

    Funding will be required to support this work.

Campus-wide Communications Infrastructure

MIT installed its present 5ESS telephone system in 1988. This system consisted of the switch and a completely new internal and external wiring plant, including a network of multimode fiber cables connecting each building to one of the three 5ESS switch sites and interconnecting the switch sites. It is now time to begin to replace the switch modules. The first of these replacements occurred this year, and one or two modules will be replaced each year over the next several years. In the coming year we will also upgrade the switch software (as we do each year) and replace the switch management software. Costs of this hardware and software will be funded via the rate structure as was the initial installation.

The demand for on-campus network bandwidth will soon surpass the capacity of our current multimode fiber plant. Replacing the fiber plant with single-mode fiber that will permit data rates of 2.4 Gb/s with current technology will cost about $2,250,000. Most of the present multimode fiber was installed as a gift of AT&T when we purchased the switch from them. Gift possibilities for the replacement fiber need to be explored at this time. Also, we need to explore the possibility of funding all or part of the replacement cost from the telephone and network rate bases.

Increasing the bandwidth to the desktop requires not only a high-bandwidth backbone provided by the fiber plant and the central network switches but also a high-bandwidth building infrastructure of network switches, hubs, and wiring to the desktop. Current estimates for the building infrastructure costs are not complete. Hardware available today is early in its production cycle and therefore expensive. The original intrabuilding wiring predated the specifications now used to guarantee the support of higher speed networks. Since wiring installed after 1993 meets these specifications, the need for new wiring will have to be determined on a case-by-case basis.

While the costs for equipment and wiring for all of MIT's buildings is significant, these costs likely will not be the limiting factor in increasing network bandwidth to the desktop. When the 5ESS wiring infrastructure was installed, many non-standard spaces were used for wiring closets. For example, many janitor's closets were turned into shared space with wiring blocks simply mounted on the wall. In other locations (e.g., second floor of Building 3) small indentations in the corridor were walled off to serve as phone closets. These spaces which constitute about half our telephone closet inventory will not accommodate the equipment required in order to deliver 100 Mb/s network service to the desktop. Replacing these closets will require difficult negotiation with the current occupants of the necessary space and construction. However, higher bandwidth cannot be provided in the areas of the Institute served by these non-standard wiring closets until they are replaced. (New closets require 150 square feet of space to accommodate the network equipment racks for 150 network drops.)

The total costs involved in these activities to upgrade the network infrastructure will be significant although they are not fully known at this time. Options for funding the costs are being developed and will be presented later in the year.

Office Computing

The principle strategic priority in the Office Computing Practice is to place even greater emphasis on completing the SAP rollout to the departments, laboratories, and centers (DLCs) by the end of June 1999. This will include the work remaining in Rollout98, the conversion of the Physical Plant system to the SAP infrastructure, the configuration and rollout of SAP HR-Payroll (if a decision is made that this software meets MIT's requirements), and introduction of SAP 4.0c release. By any definition, this is an ambitious undertaking; however, it must be accomplished within this time schedule so that we can reach some measure of stability in our SAP environment.

This work will likely require funding beyond that already budgeted for one-time reengineering expenditures.

Before the end of FY 1999 we must structure the on-going operational partnership for SAP and see it fully staffed. In all likelihood this will be a cross-functional team with membership drawn from the organizations that support or have custodial responsibility for SAP. These would include Audit, Budget, CAO, Information Systems, Personnel, Plant, Purchasing, and the Management Reporting Reengineering Team.

Also of strategic importance for Office Computing is the development of processes to systematically renew the administrative computing desktop hardware environment; to better manage the integration, testing, packaging, and release of software for this environment; and to provide on-going training and support for the Institute's staff who use the new business applications to do their work.

MIT Alumni Network Services

Based upon work begun in FY 1996 jointly by the MIT Alumni Association and Information Systems, the Alumni Association launched its first Alumni Network Service, Email Forwarding for Life. Today some 12,000 alumni have registered with the service. Enthusiasm is high and additional services - electronic alumni directory; on-line discussion forums for students, faculty, and alumni; secure financial transactions between alumni and MIT permitting on-line gifts and registration for MIT events; and career and professional support services for alumni - are already planned. This work has been performed, and continues to be done, on the margin, on a time-available basis, by staff in both Alumni and Information Systems.

Because of its strategic importance to the Institute, the Alumni Association and Information Systems are now jointly requesting an increase in base budget funding beginning in FY 1999 to continue the development and support of MIT's Alumni Network Services (ANS). This budget increase would fund four new staff positions - two each in Alumni and IS - plus the resources required to support the ANS infrastructure and the working environment for these four individuals. In the Alumni Association the two individuals would focus on customer service for the ANS products and on managing the content provided through ANS services; in IS, one individual would have responsibility for the continued development of the ANS infrastructure, and its support and integration into the MIT network, and the other would focus on building scaleable applications to deliver ANS services.

III. Current Program Priorities

IS's mission is to support MIT's commitment to education, research, and service. To accomplish its mission, IS focuses its work in six broad areas:

Work in these areas crosses boundaries between the Academic, Office, and Voice, Data, and Image Networking Practices. Key initiatives for the remainder of FY 1998 and for FY 1999 are described in this section. The listing is not exhaustive, however. Given the dynamic nature of information technology, it is difficult to predict all the projects that will be done within this period or which projects may be affected by unexpected work of higher priority.

Consistent, reliable I/T services

Partnerships with customers for solutions

Partnerships with vendors for products and services

Employees as key resources

Policies, Procedures, and Standards

Effective use of resources

As IS reaches towards operational excellence, it will rely on continuous improvement and performance measurement efforts. IS must ensure that MIT's resource allocations are guided by the overarching goal of supporting the continued improvement of education and administration at MIT. Initiatives should be consistent, as much as possible, with these central principles:

V. Issues

As we look ahead to FY 1999 and beyond, IS faces these issues: