Network-Critical

Physical Infrastructure
for Medical Imaging
and Diagnostic
Equipment

By Viswas Purani

White Paper #86

Executive Summary
Medical imaging and diagnostic equipment (MIDE) is increasingly being networked to

Picture Archiving and Communications Systems (PACS), Radiology Information Systems

(RIS), Hospital Information Systems (HIS), and getting connected to the hospital intranet as

well as the Internet. Failing to implement the necessary Network-Critical Physical

Infrastructure (NCPI) can result in unexpected downtime, and safety and compliance issues,

which translates into lost revenue and exposure to expensive litigations, negatively

affecting the bottom line. This paper explains how to plan for NCPI when deploying medical

imaging and diagnostic equipment, with emphasis on power and cooling.

2005 American Power Conversion. All rights reserved. No part of this publication may be used, reproduced, photocopied, transmitted, or 2
stored in any retrieval system of any nature, without the written permission of the copyright owner. www.apc.com Rev 2005-0
Introduction
The proliferation of information technology (IT) and other high technologies into medical imaging and
diagnostic equipment (MIDE) in the last decade has resulted in the evolution of powerful new devices in the
field of diagnostics and interventional radiology. The information generated and carried by these images is
crucial to the treatment in cardiology, neurology, oncology, gynecology & obstetrics, orthopedics, surgery
and pulmonary medicine. These new developments have helped in the early detection and treatment of
diseases and significantly improved patient care. A typical MIDE Network is illustrated in Figure 1. MIDE
can be broken down into five broad sub-categories:

1. Modalities that capture or generate the images

2. Picture Archiving and Communications Systems (PACS) that store the generated images and make
them available to the physicians and nurses for diagnosis and treatment
3. Radiology Information Systems (RIS) and Hospital Information Systems (HIS) that monitor and
manage the work flow of radiology departments and entire hospitals, all the way from the patient
check-in, to scheduling and billing, to generating electronic medical records and management
reporting
4. Computed Radiography (CR) that helps convert films to digital images on cassettes or Digital
Radiography (DR) that provides cassette-less digital images
5. Laser printers that print films when required and other peripherals

Figure 1 – Typical MIDE network

Consulting Cardiologist Modality

CT, MRI, etc.

Consulting Physician

RIS
Internet VPN

Storage - PACS HIS

Viewing
Workstations

2005 American Power Conversion. All rights reserved. No part of this publication may be used, reproduced, photocopied, transmitted, or 3
stored in any retrieval system of any nature, without the written permission of the copyright owner. www.apc.com Rev 2005-0
Modalities are high tech medical imaging systems including Computed Tomography (CT), Magnetic
Resonance Imaging (MRI), Positron Emission Tomography (PET), Ultrasound (US), and Electro-Cardiogram
(ECG). They get connected to a PACS and to RIS/HIS through Local Area Networks (LANs) or Wide Area
Networks (WANs). Newer, modern hospitals have even started to deploy Wireless Local Area Networks
(WLANs). PACS may have there own Storage Area Networks (SANs) or Network Attached Storage (NAS)
and the RIS and HIS may be made of several clusters of servers and a number of workstations distributed in
different departments of the hospitals. Because of their numerous benefits combined with the enormous
pressure on the hospitals to improve quality of care, reduce errors, comply with federal regulations like
HIPAA, and simultaneously cut cost, adoption of all these technologies is inevitable, converting the
traditional hospital into a “digital enterprise”.

The backbone of this new digital hospital is a network made up of different modalities, PACS and RIS/HIS,
CR/DR, printers and peripherals. This highly complex network and its components has to comply with
relevant standards like Digital Imaging and Communications in Medicine (DICOM), Health Language Seven
(HL7), Underwriters Laboratory (UL), Federal Communications Commission (FCC), National Electrical Code
(NEC), and other applicable local and national codes such as BS7671:2001 (U.K.), NFC15-100 (France),
and VDE (Germany), as well as international such as CEI IEC 60364. This imposes a huge challenge to the
IT and the facilities manager to provide the right Network-Critical Physical Infrastructure (NCPI). NCPI is the
foundation upon which the critical equipment, systems and networks reside, but is often ignored. It has to be
reliable, scalable, highly available, and manageable. It consists of:

1. Power systems such as UPS, power distribution units (PDUs), isolation transformers, and
generators to provide uninterrupted, conditioned, clean power to the critical loads
2. Precision cooling systems that provide optimal environment by regulating temperature and humidity
3. Racks that house the critical network equipment like servers, switches, routers, and gateways
4. Physical Security and fire protection systems
5. Cabling to interconnect equipment
6. Management systems to monitor and manage these systems, locally as well as remotely to ensure
their satisfactory operation 7x24x365
7. Services to design, deliver, install, commission, operate and maintain these systems

Special attention should be given to the hospital wiring closets which allow networking of the modalities to
PACS and RIS/HIS as well as other workstations and peripherals within the hospital premises. It is these
backbone closets that support this complex hospital network carrying critical data, voice and video, keeping
the network up and running.

The hospital power system is a large complex electrical system consisting of high voltage transformers,
automatic transfer switches (ATS), generators, isolation transformers, power distribution units (PDUs), etc.
This power system feeds a variety of electrical loads including lighting, heating, ventilation air-conditioning
(HVAC) systems, elevators, escalators, large pumps, fans, motors and more. The random nature of these
loads (turning on and off randomly) creates an unstable power environment (i.e. sags and surges) that more

2005 American Power Conversion. All rights reserved. No part of this publication may be used, reproduced, photocopied, transmitted, or 4
stored in any retrieval system of any nature, without the written permission of the copyright owner. www.apc.com Rev 2005-0
sensitive MIDE loads and other IT networks that support them, must endure. Healthcare companies and
hospitals faced with these challenges should engage partners like American Power Conversion with
engineering expertise to perform complete NCPI assessments that identify weaknesses and suggest
corrective actions. This paper discusses and reviews the challenges imposed on NCPI while deploying
medical imaging and diagnostic equipment (MIDE), with a focus on power and cooling.

Note: This paper does NOT address infrastructure related to patient safety, life support equipment,
operation rooms (ORs), Intensive Care Units (ICUs) and any other similar environments.

Modalities
Depending on the patient’s ailment, a physician can use different modalities for diagnosis and treatment (i.e.
X-Ray or CT for orthopedics, ECG or MRI for cardiology or Ultrasound for obstetrics). These modalities can
be broadly classified into two categories: portable and stationary. Portables can be further classified into
hand held (i.e. blood glucometer) and trolley or cart mounted (i.e. ultrasound) while stationary devices can be
further classified into desk mounted (i.e. blood, urine analysis equipment) or floor mounted (i.e. CT, MRI).
Figure 2 illustrates a cart mounted ultrasound and Figure 3 illustrates a floor mounted MRI machine. The
floor mounted, desk mounted and cart mounted devices need the most NCPI planning.

Figure 2 – Cart mounted ultrasound Figure 3 – Floor mounted MRI

Environment
Modalities are generally used in an indoor office environment. The cart mounted and desk mounted
modalities generally use 120VAC, 208VAC or 230VAC single phase power less than 5 kVA. The floor
mounted devices typically require 208VAC, 400VAC or 480VAC three phase power, ranging from 20 kVA to
300kVA or more. They require a lot of space and often times have their own separate room within the main
hospital building or adjacent to it. They may be cooled with the building’s comfort air-conditioning system or
they may have a precision computer room air-conditioning (CRAC) system, which more tightly controls
temperature and humidity in the environment. Figure 4 illustrates a typical MRI facility with floor mounted
equipment.

2005 American Power Conversion. All rights reserved. No part of this publication may be used, reproduced, photocopied, transmitted, or 5
stored in any retrieval system of any nature, without the written permission of the copyright owner. www.apc.com Rev 2005-0
 Figure 4 – Typical CT or MRI facility

Challenges
Modalities need to be protected from power anomalies that cause hardware failures such as blown power
supplies or printed circuit boards (PCBs) as well as from system software crashes. Physical space is a
major constraint for large modalities like CTs and MRIs, more so in big hospitals in urban areas, as they
have no room to expand. These modalities consume a lot of power, so heat dissipation is a major challenge
to the building cooling system. Often times, comfort cooling is not sufficient and precision cooling is
required. One of the most critical requirements is to provide electrical isolation from the electrical utility input
to protect the patient and the technicians from any shock hazards. Compliance to National Electrical Codes
(NEC) and other local, state, national, and international codes is of paramount importance.

Best practices
• Since the hospital power grid is electrically “noisy and dirty” with a lot of electrical surges and sags,
it is a good practice to provide UPS protection to all sensitive, expensive electronics systems, LCD
displays, workstations, printers and peripherals. The UPS system protects the hardware, avoids
unwarranted system crashes while tests are in progress, prevents loss of patient data files, and
provides safe, reliable radiology examinations.
• UPS systems used in the hospital should meet the following stringent standards:
o UL1778 – American Standard for UPS
o CSA22.2 No. 107.1 – Canadian Standard of UPS
o FCC Part 15 Class A – American Standard for Electromagnetic Radiation
o ANSI C62.41- American Standard for surge withstand capability
o IEC60950 – International Standard for UPS Systems
o EN50091-1- European Standard for Electromagnetic compatibility
Depending on their usage, cart mounted and desk mounted devices may require the UPS to
comply with UL544 (instead of UL1778) which is the Underwriters Laboratory (UL) standard for
medical & dental equipment. The newer UL 2601-1 which is similar to the international standard
IEC60601-1 is getting widely adopted for patient vicinity applications and will replace the UL544.

2005 American Power Conversion. All rights reserved. No part of this publication may be used, reproduced, photocopied, transmitted, or 6
stored in any retrieval system of any nature, without the written permission of the copyright owner. www.apc.com Rev 2005-0
 • For large floor mounted modalities, a large UPS (50-300kVA typical) should be installed to protect
the entire room. Isolation transformers and the appropriate circuit breakers should be used to limit
the leakage currents and electrical shock hazards for such modalities. These isolation transformers
are sometimes built into the UPS or may be optional outside of the UPS box. All equipment should
be installed in compliance with relevant codes like NEC, NFPA 70, NFPA99 (for U.S), and all other
applicable local and national codes such as BS7671:2001 (U.K.), NFC15-100 (France), and VDE
(Germany), as well as international such as CEI IEC 60364. Understanding and interpreting the
codes can, in some unique cases, become very controversial and the final say should be from the
authorities having jurisdiction. In instances where budget constraints prohibit investment in a large
facility level UPS, a smaller UPS (5-10kVA) dedicated to the sensitive electronics and computer
system of the CT, MRI, PET should be considered. In addition, UPS protection should be provided
to all their viewing stations & workstations.
• Sizing of the UPS for many devices like CTs and MRIs can be challenging since they draw very
high amounts of inrush current. Ample precautions should be taken while sizing their power
systems (including UPS, generators, transformers, and switchgear). Their normal power
consumption as well as the inrush current ratings are available from their manufacturers. Allow
enough margins for miscellaneous loads and future growth. Many companies like American Power
Conversion have dedicated Systems Engineers, power protection specialists, etc. who can help
evaluate the right solution for every unique customer situation.
• Adequate cooling and air flow should be provided for all modalities that have sensitive electronics
dissipating heat. For most of the cart and desk mounted modalities, building HVAC should be
sufficient. However for large floor mounted modalities like CTs, MRIs or PETs supplemental
cooling may be required. Precision cooling should be preferred as it can provide temperature and
humidity control in the CT/MRI room.
• All of the networked modalities and their physical infrastructure should be monitored and managed
(i.e. environmental conditions of the radiology room, UPS battery life, runtime and capacity, and
generator fuel) so that anomalies can be quickly detected and a corrective action be taken
proactively to avoid any downtime. Refer to APC white paper #100 “Management Strategies for
Network-Critical Physical Infrastructure” for more details.

Picture Archiving & Communication Systems (PACS)

PACS make it possible to electronically store, manage, distribute and view images. Fundamentally, these
systems are a network of all image acquisition devices, display workstations and storage systems. They are
made up of a broad range of technologies that enable digital radiology and digital hospitals that will
eventually be capable of tele-radiology, tele-medicine and tele-surgery. In the last decade, PACS have
become more complex, encompassing systems that digitally acquire, convert, interpret, transmit and store
medical images. It is growing at a significant pace and is expected to completely transform the legacy
hospitals into digital enterprises. Diagnostic images will be available anytime, anywhere with little or no

2005 American Power Conversion. All rights reserved. No part of this publication may be used, reproduced, photocopied, transmitted, or 7
stored in any retrieval system of any nature, without the written permission of the copyright owner. www.apc.com Rev 2005-0
human intervention, making their distribution faster, easier and more reliable. Because of its numerous
benefits, adoption of this technology is inevitable. Figure 5 illustrates the components of a typical PACS.

Figure 5 – Typical PACS

RIS / HIS

Workstations
PACS
Core

Scanning &
Modalities
Capture
CT, MRI, PETs etc

Printers &
Peripherals

Environment
The core of the PACS is made up of high availability RAID storage and server clusters running Windows,
Unix, Linux or a propriety operating system. These RAID storage and server clusters are housed in racks in
a computer / data room or data center environment. Typically they draw less then 10 kVA, single phase AC
power at 120V, 208V or 230VAC. Very large systems may draw three phase power.

Challenges
PACS need to be available on demand to the nurse, physician, clinician or specialist surgeon, providing
latest imaging data of the patient under treatment. It needs to be highly available, 7x24x365 and there is
little tolerance to downtime. Since the RAID drives and server clusters are confined to rack enclosures,
handling their heat dissipation within the racks often becomes a bigger challenge.

Best practices
• PACS should be protected with an N+1 redundant UPS system. This N+1 UPS system not only
protects the hardware but also protects the software from malfunctioning and gracefully shuts down
and reboots the operating system if needed, thereby preventing a hard crash. The N+1 redundancy
of the UPS system mirrors the redundancy of the RAID storage drives and server clusters which
are at the core of the PACS, providing high availability. For smaller, simpler systems, a basic UPS
can be provided.

2005 American Power Conversion. All rights reserved. No part of this publication may be used, reproduced, photocopied, transmitted, or 8
stored in any retrieval system of any nature, without the written permission of the copyright owner. www.apc.com Rev 2005-0
 • Often times, additional receptacles are needed for plugging in all required devices. Rack based
power distribution units (PDUs) should be used to provide additional outlets. PDUs that can
measure and display current, which can help prevent accidental overloading and shut down of the
PACS are recommended. PDUs that allow remote outlet control via the web are desirable as they
can help reboot a hung server or a storage drive efficiently.
• At a minimum, surge suppression should be provided for LCD/CRT based passive view-stations.
For PC based active workstations running software applications, UPS protection with graceful shut
down and reboot capabilities is highly recommended.
• The PACS storage and servers should be housed in secured, lockable, rack enclosures. These
racks should be in a temperature controlled environment. The racks housing PACS storage and
servers are generally very dense physically and in terms of power consumption. The rack doors
should be perforated, allowing for maximum airflow. When power draw in the rack exceeds 2kW,
rack-based cool air distribution units should be provided so that the RAID drives and the servers in
the top portion of the rack do not get overheated and fail. For rack power densities approaching 6
to 8kW, rack-based hot air removal units should be considered. For power densities in excess of
8kW, self-contained high-density cooling systems should be used (such as APC’s High-Density
Cooling Enclosure shown in Figure 6).

Figure 6 – Example of high density cooling solution

• A good management strategy involves the management of PACS servers, storage and their entire
NCPI including UPS, PDUs, batteries and their critical environment (temperature and humidity).

2005 American Power Conversion. All rights reserved. No part of this publication may be used, reproduced, photocopied, transmitted, or 9
stored in any retrieval system of any nature, without the written permission of the copyright owner. www.apc.com Rev 2005-0
 This will give early warning of any anomaly or impending disaster so that corrective actions can be
taken and prevent all avoidable shutdowns.

Radiology Information Systems (RIS) and Hospital

Information Systems (HIS)
RIS and HIS are server based systems running special software that make it possible to store, monitor,
manage and distribute patient medical information. They help patients in scheduling appointments,
registration, and billing, and help hospitals in generating, maintaining and managing patient’s electronic
medical records as well as generate workflow, work-list, management reporting and variety of other tasks.
These RIS and HIS are really becoming one large HIS and are integrated / networked with PACS as well as
various other modalities within the hospitals providing complete automation. By converting them into “digital
hospitals”, they can significantly improve patient-care, minimizing human errors, saving lives and reducing
costs. Figure 7 shows a typical RIS / HIS and its subcomponents.

Figure 7 – Typical RIS/HIS System

Hospital Wards Management

Reporting

Medical
Transcriptionists Technical
Department

Reception Clinical
Departments

Environment
These systems are generally housed in a data center environment drawing 10kW single phase 208VAC or
230VAC power on the lower side, to hundreds of kilowatts of three phase 400VAC or 480VAC power on the
higher side. The majority of data centers in hospitals have a UPS with battery back-up, precision air
conditioning units, and a back-up generator.

2005 American Power Conversion. All rights reserved. No part of this publication may be used, reproduced, photocopied, transmitted, or 10
stored in any retrieval system of any nature, without the written permission of the copyright owner. www.apc.com Rev 2005-0
Challenges
RIS / HIS are the most important systems within the data center requiring longer runtime and higher
redundancy and availability then most other equipment. Since these systems are merging to form one big
HIS on which the entire hospital depends for normal functioning, their availability requirements are generally
99.999% (five nines) or higher which translates to average unplanned downtime of 5 minutes per year or
less. Additionally these systems may be located in high rise buildings and attention should be paid to the
floor load (weight) handling capacity, elevator hauling capacity, door heights and widths to ensure that the
NCPI elements like UPS, batteries, and air-conditioning can be rolled in to their planned positions.

Best practices
• The physical infrastructure supporting the RIS / HIS should provide highest levels of redundancy
while minimizing the total cost of ownership. An N+1 Redundant UPS with automatic and manual
bypass is very common and often times it is extended to the generator as well as the precision air-
conditioning systems to ensure the highest levels of availability. The entire infrastructure should be
scalable to allow for future expansion, be manageable like the other IT equipment, and be
serviceable to reduce mean time to recover. An example of such a system is the APC
InfraStruXure and is shown in Figures 8. All of these characteristics contribute to the overall
availability of the system.

Figure 8 – APC InfraStruXure

• Servers & systems requiring the highest levels of availability should be identified and grouped so
that they can be provided with longer runtime and higher levels of redundancy in a separate area,
and in separate racks within the data center. This concept of “targeted availability” helps increase
availability of business critical systems without having to incur a large capital expense for the entire
data center. Higher levels of redundancy like dual feeds with dual generators and dual N+1 UPS

2005 American Power Conversion. All rights reserved. No part of this publication may be used, reproduced, photocopied, transmitted, or 11
stored in any retrieval system of any nature, without the written permission of the copyright owner. www.apc.com Rev 2005-0
 with dual power paths all the way to the rack should be considered for highly-critical data centers
and networks.
• PDUs should be able to measure and display current, which can help prevent accidental
overloading and shutdown of the RIS / HIS. PDUs that allow remote outlet control via the web are
desirable for rapidly rebooting a hung server or a storage drive. Isolation transformers should be
used wherever required and mandated by local laws.
• Precision air conditioning equipment should have the capability to allow for expansion. Redundant
air conditioning units should be considered for higher availability. For high power density racks (>2
kW / rack) additional air distribution and air removal units should be used to avoid hot spots. For
more information on cooling best practices refer to APC White Paper #49, "Avoidable Mistakes that
Compromise Cooling Performance in Data Centers and Network Rooms".

Wiring Closets or Intermediate Distribution Frame (IDF)

Medical imaging and diagnostic equipment are getting more and more networked. Modalities like CTs and
MRIs get connected to PACS which are connected to RIS and HIS which in turn are connected to the
hospital intranets and extranets. The wiring closets or IDFs, as shown in Figure 9, play a very vital role in
ensuring the connectivity of this equipment and the availability of the network, 7x24x365. Wiring closets
comprise of layer 2 and layer 3 access and distribution switches, hubs, routers, patch panels, UPS systems
with a battery back-up as well as any other miscellaneous telecommunications equipment mounted generally
in a two post rack. Newer IDFs or wiring closets also supply power over the Ethernet (PoE) to networked
devices likes IP phones, web/security camera and any other devices drawing power up to 15W. This
imposes a lot more challenges on the power and cooling requirements in the closets.

Figure 9 – IDF (wiring closet)

Patch Panel

Midspan Power
Supply

Network Telephony
System

Network Switches

Uninterrutible
Power
Supply

2005 American Power Conversion. All rights reserved. No part of this publication may be used, reproduced, photocopied, transmitted, or 12
stored in any retrieval system of any nature, without the written permission of the copyright owner. www.apc.com Rev 2005-0
Environment
These IDFs or wiring closets are typically hidden in some remote location of the building with little or no
ventilation and illumination. Legacy telecommunication networks typically used wiring closets mainly for
punch-down blocks, patch panels, and a few small stackable hubs or switches. However, new networking
equipment supply power over the Ethernet uses and dissipates considerably more power. These new
switches support data, voice and video, generally 19” rack mount type, and have varying air flow patterns
depending on the manufacturers (i.e. side to side vs. front to back). A typical IDF will house 1-3 racks worth
of equipment and draw 500 W to 4000 W of single phase AC power or more. Two post racks are very
popular in the wiring closet, however, four post racks are getting popular as the new equipment is getting
heavier and deeper.

Challenges
While deploying PACS, RIS, HIS or new modalities that are being networked, these wiring closets or IDFs
need the most attention in terms of power and cooling. Ensuring the right type of receptacles (i.e. L5-20, L5-
30, L6-20, IEC 320 C19, IEC 320 C13) and the right amount of power with the right circuit breaker protection
to all the networking equipment, UPS and PDU in the wiring closet is a challenge. Cooling and airflow are
often a bigger but ignored problem to address in these wiring closets.

Best practices
All equipment in the IDF should be protected by a UPS system. The selection of UPS should be based on:

• The total power required in Watts

• The run time required in minutes
• The level of redundancy or fault tolerance desired
• The voltages and receptacles required

The UPS system is sized by taking the sum of the Watt ratings of the loads. A common rack-mount UPS
like the APC Smart-UPS will provide approximately four nines (99.99%) of power availability, while an N+1
redundant, UPS with built in bypass, like the APC Symmetra RM, with one hour runtime will provide
approximately five nines (99.999%), which may be sufficient for most applications. See Appendix of APC
White Paper #69, “Power and Cooling for VoIP and IP Telephony Applications” for details on this availability
analysis. UPS products are available with battery packs to provide different durations of run time.

Identify the plugs and receptacles required for all the equipment including the UPS in the wiring closet.
Ideally all of the equipment should be plugged directly into the back of the UPS or the transformer, and the
use of additional outlet strips or rack PDUs should be avoided. However, if there are many devices, it may
not be practical and a Rack PDU should be used. In that case a high-grade rack PDU specifically designed
for the purpose should be used. The PDU should have enough receptacles to plug all the current equipment
with some spares for future needs. PDUs with a meter displaying the current power consumption are

2005 American Power Conversion. All rights reserved. No part of this publication may be used, reproduced, photocopied, transmitted, or 13
stored in any retrieval system of any nature, without the written permission of the copyright owner. www.apc.com Rev 2005-0
preferred as they reduce human error like accidental overloading and resultant load drops. For the correct
selection of the appropriate UPS model meeting the required power level, redundancy, voltage, and run time,
the process is simplified by using a UPS selector such as the APC UPS selector at
http://www.apcc.com/template/size/apc/. This system has power data for all popular switches, servers and
storage devices, which avoids the need to collect this data. In systems like this, the choice of configuring a
UPS will provide various receptacle options.

To ensure continuous operations of the equipment in the wiring closet, 7x24x365, cooling and airflow issues
must be identified and addressed. The problem of heat dissipation and need for supplemental air-
conditioning is most pronounced in the wiring closets which have no vents. Power dissipation in the wiring
closet should be calculated to decide on a cost effective way to solve the problem (see Table 1 & Table 2 in
APC White Paper #69, “Power and Cooling for VoIP and IP Telephony Applications” for details).
Finally, environmental monitoring (i.e. temperature and humidity) within these wiring closets is highly
recommended as it will help flag any abnormal conditions, allow for enough time to take proactive measures
and avoid downtime.

Conclusions
To ensure high availability and reliability to medical imaging and diagnostic equipment, including PACS, RIS,
HIS, modalities and their network, special attention must be paid to their Network-Critical Physical
infrastructure (NCPI). The biggest challenges are in terms of power, cooling, physical space, management
and services. Providing UPS protection to all such devices protects the hardware, prevents the software
from unwarranted crashes and increases their availability significantly. Cooling is a special problem for
bigger floor mounted modalities, high density storage and servers for PACS as well as RIS / HIS and
hospital wiring closets. In some cases, a building’s HVAC system along with proper ducting, ventilation and
airflow may be sufficient. However in many situations, additional cooling in the form of precision air
conditioning is required. Companies like American Power Conversion have dedicated team of systems
engineers, power protection specialists and availability consultants who can help in doing assessments and
audits of Network-Critical Physical Infrastructure and provide detailed actionable reports on improving overall
system reliability and availability while minimizing the total cost of ownership.

2005 American Power Conversion. All rights reserved. No part of this publication may be used, reproduced, photocopied, transmitted, or 14
stored in any retrieval system of any nature, without the written permission of the copyright owner. www.apc.com Rev 2005-0
Bibliography
1. APC White Paper #1: "The Different Types of UPS Systems"
2. APC White Paper #5: "Essential Cooling System Requirements for Next Generation Data Centers"
3. APC White Paper #37: "Avoiding Costs From Oversizing Data Center and Network Room
Infrastructure"
4. APC White Paper #43: "Dynamic Power Variations in Data Centers and Network Rooms"
5. APC White Paper #49: "Avoidable Mistakes that Compromise Cooling Performance in Data
Centers and Network Rooms"
6. APC White Paper # 69 "Power and Cooling for VoIP & IP Telephony Applications"
7. APC White Paper # 100 “Management Strategies Network-Critical Physical Infrastructure”

References
1. American Power Conversion Corporation, www.apc.com
2. Agfa Healthcare, www.agfa.com/healthcare
3. GE Healthcare, www.gehealthcare.com/worldwide.html
4. Kodak Medical Systems, www.kodak.com/global/en/health
5. Philips Medical Systems, www.medical.philips.com/us
6. Siemens Medical Systems, www.medical.siemens.com
7. Toshiba Medical Systems, medical.toshiba.com
8. IEEE White book “IEEE Recommended Practice for Electric Systems in Health Care Facilities”
IEEE std. 602-1996

About the Author:

Viswas Purani is a Director of Emerging Technologies and Applications with APC based in RI, USA having
extensive global experience. He has a Bachelors degree with a major in power electronics engineering from
India in 1988 and has been involved with technology transfers of UPS Systems and AC/DC drives from
leading American and European companies. He has successfully started a data center support company in
the Middle-East as well as Motorola semiconductor distribution in western India. He has a Masters degree in
business administration with a major in international business from USA in 1999. He joined APC in 1997
and has been product and program manager for Symmetra and InfraStruXure product lines, intimately
involved with their design, development, launch and support worldwide.

2005 American Power Conversion. All rights reserved. No part of this publication may be used, reproduced, photocopied, transmitted, or 15
stored in any retrieval system of any nature, without the written permission of the copyright owner. www.apc.com Rev 2005-0