March 2003

1.    Executive Summary

In 2001, Communication & Network Services (CNS) began to develop a long-range strategic plan for electronic campus communications. The goal was two-fold: identify a focus for information technology and telecommunications efforts in a resource-constrained environment and re-evaluate the assumptions behind large capital purchases of electronics and infrastructure.

CNS and a consultant, Western Telecommunications Consulting, gathered detailed information on the current communications environment (voice, data and video), surveyed future vendor offerings, and forecasted growth and technology migration through 2012. With this data and analysis, the project team estimated life-cycle operating costs (LCOCs) from 2002 through 2012 and evaluated various alternatives for voice technologies (Centrex, PBX and Voice over IP).

In summary, we found that:

• By 2012, the number of voice and data connections is expected to grow approximately 80% from 68,000 to 122,000.
• LCOCs of $500 million (in 2001 dollars) through 2012 are estimated, regardless of the technology alternative.
• Upgrades to the campus communications infrastructure are critical  to any long-term option.
• Assumptions behind the current infrastructure projects (ICCS, Riser Project, CENIC ONI) are valid.
• To prepare for new technologies, UC Berkeley must control its own communication infrastructure and facilities.
• Although new technologies are emerging, they do not appear to be mature enough to revolutionize CNS’ service offering or financial models.
• There is no compelling reason – financial or technological – to immediately change the voice service platform.

As a result of the study, CNS intends to focus its long-range planning efforts on the following:

• Continue to fund (or even accelerate funding) for Riser Projects, ICCS, and CENIC ONI.
• Use pilot projects to deploy new technologies rather than move the entire campus quickly toward any new service.
• Assume greater control of communications infrastructure and facilities.

2.    Background and Overview

Much of the existing campus communications infrastructure was built over the last few decades on an ad-hoc basis as the technology was conceived and developed at the University of California (UCB). At that time it was unclear what a critical role this technology would play. Today, however, there are forces at work that Communication & Network Services (CNS) to implement a more proactive and systematic approach. These driving forces include:

• Expiration of the campus’ SBC (formerly Pacific Bell) contract in September 2004
• Aging infrastructure and on-going upgrades
• Emerging technologies
• e-Berkeley initiative (increasing demand for on-line information, etc.)
• Increasing demand for information access and transport

Additionally, since 1996, the campus has funded approximately $15 million in improvements to the campus’ communications infrastructure. An additional $25 million has been approved in concept, testifying to the importance of electronic communications to the University’s academic mission and administrative functions. CNS is aware of the significant amount of money and effort these infrastructure projects cost the campus and regularly re-evaluate them. In addition, CNS must ensure that the decisions the campus makes today regarding the communications infrastructure will meet its future needs.

The long-range strategic planning effort started in 1997 as Project VOCAL (Voice Options for CAL), which began as a project to review alternatives for voice service because of the expiration of the contract with Pacific Bell (now called SBC, who currently provides voice services to the campus). Given the above stimuli, Project VOCAL evolved into a more comprehensive project, and in 2001, CNS began creating a long-range strategic plan for electronic campus communications through the year 2012.

CNS identified two major goals for the strategic planning process. First, the campus needs a focus for information technology and telecommunications efforts in a resource-constrained environment. Second, CNS needs to re-evaluate the assumptions behind capital purchases of electronics and infrastructure projects such as the Riser Project, the Inter-building Campus Communications System (ICCS), and CENIC’s Optical Network Infrastructure (ONI) Initiative.

The project team consisted of staff from UCB’s Communication & Network Services and a  consulting firm, Western Telecommunications Consulting, who facilitated the strategic planning process. Project Team members included:

Communications & Network Services:

Margaret Baker – Manager, Planning, Analysis & Outreach
Linda Chu – Project VOCAL Manager
Marv Eckard – Associate Director and Manager, Network Services
Cliff Frost – Director
Nancy Lim – Manager, Financial Services
Michael Sawyer – Manager, Operations, Installations & Repair

Western Telecommunications Consulting

Philip Beidelman – President
Karen Mattis – Practice Director
Dean Oyama – Consultant
Albert Slater, RCDD - Consultant

3.    Scope of Work and Findings

The major components of the project were:

• Determine planning assumptions.
• Gather data describing current environment.
• Release a Request for Technology Update to vendors.
• Forecast growth and technology migration.
• Estimate associated costs and evaluate alternatives.
• Communicate with other campus groups.

3.1.    Planning Assumptions

Planning assumptions were developed to establish guidelines for the project plan. A detailed description of the planning assumptions is available in References 1 and 2. Major assumptions are described in the section below.

The strategic plan

• covers the time period from 2001 through 2012.
• addresses voice, video and data network and needs.
• provides infrastructure to support uniform access to all network functionality at all sites, regardless of location (campus, peripheral or off-campus, see below for definitions).
• includes wired and wireless solutions.
• assumes that, starting in 2005, growth in voice will be Voice over IP (VoIP).

Five alternatives for the voice network (described in Table 1) were considered. Please note that  throughout the planning process, it was assumed that these alternatives were considered “forklift” solutions , i.e., only the one technology (e.g., Centrex, PBX, or VoIP) would be used exclusively, rather than a mixture of one or more (e.g., PBX for campus+Centrex for periphery and off-campus). Additionally, the Centrex alternative should not be considered as a continuation of the existing Centrex service. The alternatives assume proposed upgrades and enhancements to the existing voice and data networks – equipment, infrastructure, new services, etc. – to meet the growing needs of the campus.

Although not explicitly mentioned in the alternatives, the data networks and wireless systems were also included in the planning process. Wireless technology is expected to complement, not replace, wired access through the planning horizon. 

* Line oriented – The use of a Centrex or PBX system to produce dial tone to individual telephones or groups of telephones through a key system arrangement. UCB is currently a line oriented telecommunications environment.

** Port oriented – The use of a PBX system or VoIP technology to produce telecommunications ports for each telephone. The existing key systems would be replaced with individual “port oriented” multi-line telephones.

3.2.    Current Environment

UCB currently supports separate voice, data and video networks. Voice and data are supported by Communication & Network Services. Video services are supported by Educational Technology Services (ETS), formerly Office of Media Services (OMS). The following sections describe the voice, data and video environments as they existed in early 2001, when the strategic planning process started.

Voice:  Wired voice service to CNS-supported buildings is provided primarily through the 1989 SBC/Pacific Bell contract with UCB. Service is supported through a DMS 100 Centrex and local key systems and is distributed via copper. There are approximately 28,000 wired voice connections (or ports). Wireless voice communication is supported via approximately 500 cellular phones, 900 pagers and 700 two-way radios. Detailed information regarding the type (e.g., analog versus digital) and number of Centrex lines was available to the building level.

Data:  The basic campus network service consists of 10 Mb ethernet to the desktop, interconnected by a campus-side FDDI backbone. Additionally, there are several dozen, 100 Mb ethernet cluster and a small amount of Gigabit ethernet. There are approximately 40,000 desktop connections on approximately 400 logical subnets. High-speed data connectivity between off-campus buildings and the UCB network is provided by SBC/Pacific Bell and AT&T Broadband. Detailed information regarding the type (e.g., 10 Mb versus 100 Mb ethernet) and number of data connections, switches, routers and other networking equipment was available to the telecommunications room (or “closet”) level.

Video:  Video service on campus is primarily baseband audio and video over specialized circuits and is separate from the campus voice and data networks. A few buildings (without copper AV facilities) are served via wireless from the campus ITFS transmitter on San Bruno Mountain. The central AV control point on campus is Dwinelle Hall, with the audio and video cables radiating from Dwinelle Hall to other campus buildings. One or two video circuits are provisioned to each connected building, so there is a limit of one or two simultaneous programs to a building. The bandwidth of the video circuits is at least 6 MHz; audio bandwidth is at least 15 kHz. Signals are NTSC composite video for display in classrooms, with image sizes ranging from a 23-inch direct view CRT to a 25-foot high quality projection system. There are several “return” signal circuits from large lecture halls or specialized classrooms to bring programming back to the Dwinelle Hall control room for recording or further distribution. Presently 44 rooms in 15 buildings are connected, and there are 8 video return circuits.

Structures served by CNS
The plan encompassed over 500 buildings and structures served by CNS, including student residence halls, family student housing, space housed by UC affiliates (e.g., UC Extension). For the purposes of this study, a building is defined as any structure with a voice or data connection fed to it. Buildings were categorized into the categories in Table 2.
 

The list of buildings included in the planning process is listed in the Infrastructure Assessment Worksheets (Attachment III of Reference 1) or Table VIII of Reference 2: Buildings and Structures Included in the Project.

Inter-building infrastructure
Although virtually all campus buildings are connected to the campus data network, most of the buildings are connected through ad hoc pathways, including old telecommunications conduits, steam tunnels, or other conduits previously used for other functions. These conduits are at capacity in many places. Others are damaged and no longer usable. Still others pose a hazard to people working in them (e.g., steam tunnels). This limits future connectivity and service upgrades, which makes some projects impossible and lengthens the implementation times of others. To meet the vision of a common campus “communications backbone” in support of multiple communications services, The Inter-building Campus Communications System (ICCS) project is designing and constructing a system of underground conduits and vaults specifically for communications cabling. Three of the seven major portions of the ICCS have been completed. CNS estimates that the remainder of this project will cost approximately $11 million, with a total cost of approximately $19 million. (The cost of the ICCS project is funded through the Central Campus and is not included as part of this project.) Figure 1 shows the late 2002 status of the ICCS.

Voice service is provided via outside plant (OSP) copper as depicted in Figure 2 of Reference 1: Outside Copper Plant Based on PacBell Drawings.

Riser systems: Approximately two-thirds of campus buildings use coaxial cabling in their vertical (or riser) cabling systems, which is not capable of supporting more than shared 10 Megabit per second ethernet and does not allow flexibility in how services are provided and supported. CNS has created a project, the Riser Project, that will upgrade these systems to the industry standard of multi-mode and single-mode fiber optic cables. The Riser Project will also redesign and remodel the telecommunications rooms and cabling pathways, if necessary. CNS has proposed the Riser Project but has been unable to procure a dedicated funding source for the upgrades. CNS estimates that the remaining riser upgrades will cost approximately $14 million, with a total project cost of $21 million. (The cost of the Riser Project is included in the infrastructure costs of this project.) Figure 1 shows the late 2002 status of the Riser Project.

Telecommunications Rooms (“Closets”)
There are over 900 telecommunications rooms or “closets” serving the voice and data networks. A preliminary list was created from CNS’ database (Mystic) and was reviewed by a representative group of engineers, technicians and managers. This technical review group was to correct errors in the list (room numbers, locations, etc.) and determine the viability of the closets through the term of the planning horizon (i.e., would the closet be able to continue to provide service through 2012). The condition of the closets were compared to those recommended in the BICSI Telecommunications Distribution Methods Manual – Ninth Edition and were categorized as described in Table 3. Detailed information regarding the condition of the telecommunications rooms is located in Infrastructure Assessment Worksheets (Attachment III of Reference 1).

Based on the technical review, although the majority of the closets (Category Y) can provide service through 2012, significant improvements still are required before non-Centrex voice services can be deployed.

3.3.    Vendor Offerings

To ascertain what new technologies might emerge during the planning horizon, vendors were to provide information on their research and development projects in the areas of telephony, wireless voice, wireless data, campus backbone, campus metropolitan area network/wide area network, and video. The following vendors were asked in June 2001 to respond to UCB’s “Request for Technology Update” (Reference 3).

• Alcatel
• AT&T Broadband
• Avaya
• Caspian Networks
• Cisco
• Ericsson
• 1st Virtual Communications
• Juniper Networks
• Lucent Technologies
• Malibu Networks
• Nortel Networks
• SBC/Pacific Bell

Five vendors responded: Avaya, Caspian Networks, Cisco, Nortel, and SBC/Pacific Bell. These responses confirmed the project team’s understanding of the market for Centrex, PBX , VoIP and wireless technologies. However, it did not uncover any new technologies that would significantly change the directions developed in the Planning Assumptions (see Section 3.1) or that would dramatically enhance or reduce pricing of the current or planned service offerings through the planning horizon. Two of the more promising technologies that CNS is continuing to explore are:

Non-blocking internal fabric core offered by Caspian Networks – This core would allow for 4 microsecond delay through the UCB core at a speed in excess of 60 to 100 Gigabits per second, allowing UCB to forego a 10 Gigabit core and migrate directly to a higher speed core with greater control of the network.

IP Centrex – The roll-out for IP Centrex will continue for the next three years (through 2004). IP Centrex currently does not provide the same level of reliability as traditional Centrex and therefore should not be used for emergency telephones or voice circuits. If VoIP were deployed, special analog Centrex circuits need to be deployed to ensure E911 capabilities.

3.4.    Growth and Migration

The project team also estimated the amount of growth in the network for the time periods listed in Tables 4, 5 and 6: 2004, 2006 and 2012. First, the team categorized the space serviced by telecommunication closets into various usage categories and estimated growth in voice and data connections separately, as described in Tables 4 and 5. Growth in video was assumed to be in IP and is included in the data growth estimates.

Then each telecommunications closet was categorized by the technical review group (see Telecommunications Rooms section above) as serving areas with a particular usage. Growth in the number of voice ports was calculated at a building level; data available to CNS did not allow the team to associate numbers of voice ports to specific closets. Growth in data ports was calculated at the closet level. For example, a closet that serves several floors of office space is estimated to increase the number of voice and data connections by 10% and 20%, respectively, between 2001 and 2006. When a telecommunications closet (or building, in the case of voice connections) functions under multiple usage functions, the most aggressive growth classification was used.
 

 * Working at Cut is defined as the number of ports currently in use at the time of the study.
 **Please see Table 6 for a complete definition of this term.

                        
Table 6 describes the design requirements of the networks at the various time periods used to estimate costs.

Growth due to planned capital projects of new buildings were also considered. A list of these projects is included as Appendix A.

In addition to growth in the number of data ports, the type of network technology also was expected to change, as older equipment is replaced. Table 7 summarizes the timetable for the planned technology conversions through the planning horizon.

3.5.    Estimated Costs and Evaluation of Alternatives

Using costing data from WTC experience and previous UCB projects, the UCB project team estimated the cost to convert to the proposed alternative and evaluated the relative merits or each.

Life Cycle Operating Costs (LCOC)
Estimates of the Life Cycle Operating Costs (LCOC) were developed for each of the five alternatives, as shown in Figure 3. The costs described below assume a complete cutover of all CNS-supported structures from the current Centrex voice service to the proposed alternative. Pricing was based on the university and WTC’s experience. The cost estimates include:

• Incremental increases in staffing and training, where applicable.
• Capital costs for voice and data equipment, etc.
• Capital costs for infrastructure, excluding the cost of the Inter-building Campus Communications System (ICCS), which is to be funded separately.
• Maintenance costs.
• Growth in features and station equipment.
• Conversion to VoIP telephones after 2005
• Centrex contract costs through 2004
• Refreshing the data network every 4 years
• Refreshing the PBX hardware every 6 years (for PBX option only)
• Capital expenditures for data equipment, voice equipment, and infrastructure financed over 4, 6 and 10 years, respectively.
• Financing rate of 6%

Please see the Capital and 10-year Life Cycle Operating Cost Model Summary Charts and Detail Spreadsheets (Attachment VI of Reference 1) for more details, such as annual life cycle costs and capital costs models.

A benchmark LCOC was also developed. The benchmark LCOC assumed that there would be no change in the services CNS offers through the planning horizon. It is a straight-line projection of CNS’ 2000 budget, assuming 3% growth. This benchmark cost was developed for comparison purposes only. It does not meet the desired design requirements for growth and redundancy of the voice and data networks.

Evaluation of Alternatives
The project team noted the advantages and disadvantages of each alternative (See Appendix B). WTC also provided its own analysis of the relative merits of the various alternatives for voice, as described in Table X of Reference 1. Although each technology is viable, it is unlikely that any one solution is ideal for all the structures served by CNS. This is in part due to the campus’ diverse geography with structures throughout the cities of Berkeley, Albany, Oakland, Richmond and the rest of the San Francisco Bay Area. Also, VoIP currently cannot adequately address the E911 requirements of a university environment. If VoIP were deployed, special analog Centrex circuits need to be deployed to ensure E911 capabilities.

The existing OSP copper can be used to deploy a PBX system on the campus proper, using a star configuration with 25 node systems (See Figure 4 of Reference 1: PBX Node locations and the resulting areas of coverage). The node systems would be connected via meshed fiber backbone using the campus backbone fiber (the proposed ICCS). This design will require the splicing of OSP copper in many maintenance manholes and remodel of closets to accommodate PBX switches (see Attachment IV of the Reference 1).

However, much of the existing communications infrastructure will not be able to support enhanced voice and data services. Approximately $30 million in infrastructure improvements will be required to implement any of the alternatives. Furthermore, as currently planned, the conversion from hubs to switches by 2006 would not be in time for the expiration of the Centrex contract in September 2004. Either this conversion will need to be accelerated to accommodate VoIP solutions or conversion will need to happen in stages as the hubs are replaced.

Wireless access needs in outdoor areas of the campus were also identified, as shown in Figure 5 of Reference 1: Wireless Areas Needing Coverage. Assuming a 150-radius of coverage, potential locations of wireless access points were determined. Please see Figure 6 of Reference 1: Wireless Outside Design.

3.6.    Communication with Other Campus Groups

As part of the planning process, Communication & Network Services solicited input from and communicated progress with other campus service providers (Capital Projects, Physical Plant–Campus Services (PPCS), Educational Technology Services), faculty, and other campus IT professionals (through the Network Advisory Committee). Presentations were also made to the Vice Chancellors Advisory Committee (VCAC) and various academic departments (as requested). Updates were also published in the Berkeley Computing and Communications and on the project’s web page http://cns-pao.berkeley.edu/Vocal.

4.    Conclusions and recommendations

As described previously, the project team evaluated Centrex, two versions of Private Branch Exchange (PBX) and two versions of Voice over IP (VoIP) and their associated annual and life cycle operating (LCOC) costs. Many of the findings and conclusions were much as expected:

• Upgrades to the campus communications infrastructure (e.g., ICCS, riser systems) are critical to any long-term option. – Upgrades to the Inter-building Campus Communications System (ICCS) and riser systems are needed to maintain current offerings. They become even more critical when considering emerging technologies, such as VoIP, with the additional power, equipment, and space requirements.

• Assumptions behind the ICCS, Riser Project and CENIC ONI continue to be valid. – Estimated costs for the ICCS and Riser Projects matched with those developed independently by CNS.  The strategic plan assumed a more aggressive scheduling and funding than currently exists to provide services like PBX and VoIP.

• To prepare for new technologies, UC Berkeley must control its own communication infrastructure and facilities. – To maintain and fully leverage the advantages of new technologies, CNS will need to better control access to and flexibility of the voice and data networks. This becomes of special concern if voice traffic is routed through the traditional data network, as issues surrounding security and quality of service arise. UCB can work with vendors to create more partnership and a more cooperative relationship then it currently experiences, or UCB can bring the voice services in-house.

“Control” also includes maintaining an accurate accounting of its infrastructure and facilities. This will also facilitate current development, operations and maintenance efforts. Eventually, CNS should eliminate, as much as possible, field visits (except for major installations, e.g., new equipment) and allow more remote control. The existing database was based on a commercial-off-the-shelf billing system, which was modified, is not likely to meet this need and should be replaced.

The project team also found that:

• By 2012, the number of voice and data connections (nodes) is forecasted to grow approximately 80 percent from 68,000 to 122,000. Figure 1 illustrates the growth over the planning horizon. Growth in video services is assumed to be over IP and is included as “data” nodes.

• Total Life Cycle Operating Costs (LCOC) of $500 million (in 2001 dollars) through 2012 can be expected, regardless of the technology alternative implemented. (See Figure 2.) This corresponds to annual operating costs of $50 million per year or a $35 to $45 per month cost per voice, video, or data node (in 2001 dollars). Please note that the options were “forklift” solutions. They assumed that only the one technology (e.g., Centrex, PBX, or VoIP) would be used exclusively, rather than a mixture of one or more (e.g., PBX+Centrex, PBX+VoIP).

• Although new technologies are on the horizon, they do not appear to be mature enough to revolutionize our service offerings or financial models. Although five vendors responded to our Request for Technology Update, most of the technologies offered were nearly production-ready (within the next 1 to 3 years) and did not reflect future offerings during the mid- or latter stages of the planning horizon. Of the few research and development efforts discussed, none were compelling enough to make significant changes to our existing operations or planning.

• There is no compelling reason, financial or technological, to immediately change our voice service platform. Consequently, CNS should pursue extending or renewing the Centrex voice services contract while focusing planning efforts as described in the following sections.

When compared to comparable studies conducted by other universities and available industry data, CNS found that they had made similar findings with respect to LCOC and the cost of major platform changes.

As a result of this study, CNS intends to focus its long-range communications planning efforts on the following:

• Continue to fund (or even accelerate funding) for Riser Projects, ICCS, and CENIC ONI.

• Use pilot projects to deploy new technologies rather than move the entire campus quickly toward any new service.

• Assume greater control of communications infrastructure and facilities.

Specific implementation tasks are noted in the following section.

5.    Implementation Plan

The implementation plan separates tasks based on the time frame in which they should be completed – either before the current Centrex contract expires (short-term) or within the 2012 horizon (long-term).

5.1.    Short-term projects (by Fall 2004)

These projects should be completed before the end of the current Centrex contract expires in September 2004.

Extend Centrex Services

• Review current contract: What services to keep? Any additional services? Lower costs? Any terms not fulfilled during the last contract that need revisiting? Any other points of negotiation? 
• Review vision with SBC.
• Re-negotiate contract with SBC.

Begin a pilot project for voice services at Richmond Field Station (RFS)

• The existing ROLM switch (PBX) must be replaced in early 2003. CNS should view RFS as a pilot project for future voice services on campus. Can a PBX or VoIP solution be implemented at this site that also may be suitable (at a larger scale) on the campus?
• Review potential solutions with vendors and release a Request for Proposals.

Coordinate infrastructure projects

Do design changes need to be made to ICCS and riser systems (CEVs, conduit counts, closets)? Do locations of entry points onto campus need to be changed or additional ones added? How should upgrades be prioritized to facilitate the deployment of pilot projects or new technologies (e.g., upgrade risers in buildings where the ICCS is completed)?

Meet at least once a quarter to discuss high-level voice planning – copper capacity on-campus and peripheral buildings (e.g., Strawberry Canyon).

Coordinate with SBC (Tim Randel and Greg Wong) for information on cable counts, potential solutions to capacity issues (e.g., dial tone over fiber, transferring pairs from one building to another).

Are there opportunities to take better control of the infrastructure/facilities?

Continue to learn about emerging technologies

• Technical briefings with vendors (similar to the Request for Technology Update for this project) should be scheduled every year.
• Pilot projects for the following:

• VoIP – An IPCentrex technical pilot was completed in late 2002. CNS concluded that the technology was still limited, but that it could be deployed at remote sites if a suitable cost structure was offered by SBC. Other pilots of VoIP technology should be implemented (e.g., limited tests with equipment in the controlled environments vaults, using the ICCS infrastructure and facilities).
• Unified Messaging – A technical pilot for unified communications offerings began in spring 2002. At the time of this report, none of the solutions seemed to meet all the identified needs and requirements either technically or financially. However, the pilot has identified two potential vendors and should be offering some new services by mid-2003.
• Wireless – Beginning in Fall 2001, a pilot wireless LAN service became available in a few selected locations around campus. The new service, known as AirBears, enables one to connect to the campus network with a laptop equipped with a wireless Ethernet card. The service is based on the IEEE standard 802.11b wireless LAN technology, which supports speeds of up to 11 Mbps. Current efforts are introducing 802.11a technology. Similar pilot projects should be considered for wireless voice technology.

Migrate to an infrastructure-based database

• Identify the needs and requirements of the database.
• Determine whether the new database will be commercial-off-the-shelf (COTS) or designed in-house.
• Buy COTS or begin design and creation of new database.

5.2.    Long-term issues (2012 horizon)

These issues need be addressed during the 2012 planning horizon.

• Academic Strategic Plan – Do CNS’ plans fit the Academic Strategic Plan?
• New Century Plan – Do CNS’ plans fit the New Century Plan? Where are the growth areas?
• Long-Range Development Plan - Do CNS’ plans fit in the Long-range Development Plan? Where are the growth areas?
• Be prepared in 2012 to take voice services for campus proper back in-house. What are the regulatory issues with providing in-house voice service? Staffing? Training? Organizational Structure? Billing and financial systems and processes?
• Infrastructure improvements – Will the Riser and ICCS be completed?
• What demands will video services put on the network(s)?
• What role will wireless and VoIP technologies play?
• Outside plant and facilities maps and records – Are we keeping these updated? Do we have what we need? Can we get the entire infrastructure onto a single CAD file for planning purposes? Will we have the information that we need when we take control of the voice facility?
• General staffing issues.
• Space and power issues.

6.    References

1. Western Telecommunications Consulting, Development of the Telecommunications Infrastructure Business Plan – University of California, Berkeley – Final Report – Volumes 1 through 5, December 15, 2001.

Volume 1: Final Report
Volume 2: Planning Assumptions
Volumes 3 – 5: Attachments
	Campus Outside Plant Drawings
	Physical Node Room Closet Survey
	Technology Update – List of Vendor Response Documents
	Capital and 10-year Life Cycle Operating Cost Model Summary Charts and Detail Spreadsheets

2. Western Telecommunications Consulting, Planning Assumptions, Final Draft, July 6, 2001.

3. Western Telecommunications Consulting, Technology Update, June 28, 2001.

4. Western Telecommunications Consulting, Station Port Model – Volumes 1 through 3, July 9, 2001.

General Notes/Assumptions

• In Centrex and PBX, we will be moving network growth to VoIP in 2005.
• We are moving to MPOE the entire campus.
• We handle the relationship with long distance vendors for FRS only.
• Voice and data will be on the same network for VoIP.
• Regardless of the alternative chosen, there will be an increase in staff and training costs (will need more staff, staff with different skills). How much varies with each alternative.
• For PBX and VoIP, we will need 24X7 staffing coverage.
•  Whatever we do, it won’t be one of these (more likely some mixture).
• If we don’t offer a service/technology, our customers will do it on their own. We need to keep up with demand, but we do not need to be early adopters. We should learn from others’ experiences.