Unit 4 - Data Center Facilities
Unit 4 - Data Center Facilities
Unit 4
Data Center Facilities
BY Zahir Al Rashdi
Data Center Facilities
Data center facilities are custom-designed locations that are outfitted with specialized computing,
storage, and network equipment. These facilities have several functional layout areas, as well as
various power supplies, cabling, and environmental control stations that regulate heating,
ventilation, air conditioning, fire protection, and other related subsystems.
Computing Hardware
Much of the heavy processing in data centers is often executed by standardized commodity
servers that have substantial computing power and storage capacity. Several computing
hardware technologies are integrated into these modular servers, such as:
Computing Hardware
● Rackmount form factor server design composed of standardized racks with
interconnects for power, network, and internal cooling.
● Support for different hardware processing architectures, such as x86-32bits, x86-64,
and Reduced Instruction Set Computer (RISC).
● A power-efficient multi-core CPU architecture that houses hundreds of processing
cores in a space as small as a single unit of standardized racks.
● Redundant and hot-swappable components, such as hard disks, power supplies,
network interfaces, and storage controller cards.
Computing architectures such as blade server technologies use rack-embedded physical
interconnections (blade enclosures), fabrics (switches), and shared power supply units and
cooling fans. The interconnections enhance inter- component networking and management while
optimizing physical space and power. These systems typically support individual server
hot-swapping, scaling, replacement, and maintenance, which benefits the deployment of
fault-tolerant systems that are based on computer clusters.
These arrays inherently divide and replicate data among multiple physical drives, and increase performance and redundancy
by including spare disks. This technology is often implemented using redundant arrays of independent disks (RAID)
schemes, which are typically realized through hardware disk array controllers.
I/O Caching
This is generally performed through hard disk array controllers, which enhance disk access times and performance by data
caching.
Hot-Swappable Hard Disks
These can be safely removed from arrays without requiring prior powering down.
Storage systems, cont
Storage Virtualization
This is realized through the use of virtualized hard disks and storage sharing.
Storage systems encompass tertiary redundancies, such as robotized tape libraries, which are used as backup and recovery
systems that typically rely on removable media. This type of system can exist as a networked IT resource or direct-attached
storage (DAS), in which a storage system is directly connected to the computing IT resource using a host bus adapter
(HBA). In the former case, the storage system is connected to one or more IT resources through a network.
Networked storage devices usually fall into one of the following categories:
Physical data storage media are connected through a dedicated network and provide block-level data storage access using
industry standard protocols, such as the Small Computer System Interface (SCSI).
Hard drive arrays are contained and managed by this dedicated device, which connects through a network and facilitates
access to data using file-centric data access protocols like the Network File System (NFS) or Server Message Block (SMB).
NAS, SAN, and other more advanced storage system options provide fault tolerance in many components through controller
redundancy, cooling redundancy, and hard disk arrays that use RAID storage technology.
Network Hardware
Data centers require extensive network hardware in order to enable multiple levels of connectivity. For a simplified version
of networking infrastructure, the data center is broken down into five network subsystems, followed by a summary of the
most common elements used for their implementation.
A subsystem related to the internetworking infrastructure, this interconnection is usually comprised of backbone routers that
provide routing between external WAN connections and the data center’s LAN, as well as perimeter network security devices
such as firewalls and VPN gateways.
This subsystem comprises Web acceleration devices, such as XML pre-processors, encryption/decryption appliances, and
layer 7 switching devices that perform content-aware routing.
LAN Fabric
The LAN fabric constitutes the internal LAN and provides high-performance and redundant connectivity for all of the data
center’s network-enabled IT resources. It is often implemented with multiple network switches that facilitate network
communications and operate at speeds of up to ten gigabits per second. These advanced network switches can also perform
several virtualization functions, such as LAN segregation into VLANs, link aggregation, controlled routing between
networks, load balancing, and failover.
SAN Fabric
Related to the implementation of storage area networks (SANs) that provide connectivity between servers and storage
systems, the SAN fabric is usually implemented with Fibre Channel (FC), Fibre Channel over Ethernet (FCoE), and
InfiniBand network switches.
NAS Gateways
This subsystem supplies attachment points for NAS-based storage devices and implements protocol conversion hardware
that facilitates data transmission between SAN and NAS devices.
Data center network technologies have operational requirements for scalability and high availability that are fulfilled by
employing redundant and/or fault-tolerant configurations. These five network subsystems improve data center redundancy
and reliability to ensure that they have enough IT resources to maintain a certain level of service even in the face of multiple
failures.
Ultra high-speed network optical links can be used to aggregate individual gigabit-per-second channels into single optical
fibers using multiplexing technologies like dense wavelength-division multiplexing (DWDM). Spread over multiple locations
and used to interconnect server farms, storage systems, and replicated data centers, optical links improve transfer speeds and
resiliency.
Other Considerations
IT hardware is subject to rapid technological obsolescence, with life cycles that typically last between five to seven years.
The on-going need to replace equipment frequently results in a mix of hardware whose heterogeneity can complicate the
entire data center’s operations and management (although this can be partially mitigated through virtualization).
Security is another major issue when considering the role of the data center and the vast quantities of data contained within
its doors. Even with extensive security precautions in place, housing data exclusively at one data center facility means much
more can be compromised by a successful security incursion than if data was distributed across individual unlinked
components.
PRIMARY ROOMS
Electrical Room
Reserved for electrical equipment and installations, such as power distribution and bypasses, this space is divided into
individual rooms dedicated to housing power generators for temporary emergency use, UPS, battery banks, and other
electrical subsystems.
Mechanical Room
This space houses mechanical equipment, such as air conditioning and cooling engines.
This space is dedicated to safely storing both new and used consumables, such as removable media used for
backups.
Primary Rooms, cont
A building space that is usually isolated from the computer room for the placement of personnel involved in data
center operations.
Telecommunications Entrance
Typically located outside of the computer room, this space functions as a demarcated area that houses
telecommunications equipment and the ends of the external cabling that enter the data center boundaries.
Primary Rooms, cont
Computer Room
A highly critical zone with strict environmental control and access that is limited to authorized personnel, this room
usually has raised floors and safety vaults that are designed to protect the data center equipment from physical
hazards. The computer room is subdivided into the following specialized areas:
Main Distribution Area (MDA) – Encloses backbone-level telecom and network equipment, such as core
switches, firewalls, PBX, and multiplexers.
Horizontal Distribution Area (HDM) – Encloses network, storage, and keyboard, video, and mouse (KVM)
switches.
Equipment Distribution Area (EDM) – This is where computing and storage equipment is installed on
standardized rack cabinets. Cabling subsystems, usually divided into backbone cabling (main interconnects) and
horizontal cabling (individual equipment connects), interconnect all of the data center equipment, as illustrated in
Figure D.1.
Figure D.1. A data center’s internetworking regions, divided into backbone and horizontal cablings (adapted from TIA-942).
ENVIRONMENTAL CONTROLS
Utility power infrastructure interconnects with external power utility providers, and is usually supplied by
high-voltage power lines in larger data centers. Voltage conversion requires the use of on-site utility power
stations, while redundant interconnections are required for electrical grid configurations.
Power Distribution
An electrical subsystem that traditionally uses low-voltage alternating current (AC) for its operations, the power
distribution system comprises power distribution units that provide electrical power to all data center equipment.
The power supplies that are embedded in the computer equipment can require an AC/DC conversion, since
some electronic circuits in IT equipment run on DC power. The common inefficiencies of both voltage and AC/DC
conversions are notorious for causing power outages.
Uninterruptible Power Source (UPS)
Many IT resources, most notably physical servers, undergo data loss and other types of malfunctioning if they are
unexpectedly shut off. This subsystem is comprised of equipment dedicated to powering the data center during
temporary primary power source failures. Multiple UPS devices can operate alongside other power sources to
quickly fulfill the data center’s power requirements. The UPS subsystems are also responsible for removing
voltage flow fluctuations in order to normalize the incoming current and prevent undue strain on the IT
infrastructure. UPS equipment usually relies on DC battery banks, which provide only a few hours’ worth of
backup power.
Power Engine-Generator
Gas generators are standard combustion engines that are used by larger data centers to sustain operations
during natural disasters and power-grid failures. Energy efficiency is commonly measured by the power usage
effectiveness (PUE) metric, which is expressed as the ratio of the total facility power entering a data center to
the power used by its IT equipment, as follows:
The PUE is determined by the power required by the IT equipment’s supporting subsystems and should ideally
be a ratio of 1.0. An average data center would have a PUE over 2.0, while the PUE of a more complex and
efficient data center would be closer to 1.2.
INFRASTRUCTURE REDUNDANCY SUMMARY
The TIA-942 classification specifies the minimum requirements for infrastructure redundancy in four tiers, an
approach that is useful for comparing and evaluating data center facilities (briefly described in Table D.1).
Technologies used to increase storage capacity include DAS, SAN, and NAS.
Computing hardware technologies include rack mounted server arrays and multi-core CPU
architectures, while specialized high-capacity network hardware and technology, such as
content-aware routing, LAN and SAN fabrics, and NAS gateways, are used to improve
network connectivity.