A server farm or server cluster is a collection of
computer servers, usually maintained by an
organization
An organization or organisation (Commonwealth English; see spelling differences), is an entity—such as a company, an institution, or an association—comprising one or more people and having a particular purpose.
The word is derived f ...
to supply server functionality far beyond the capability of a single machine. They often consist of thousands of computers which require a large amount of power to run and to keep cool. At the optimum performance level, a server farm has enormous financial and environmental costs.
They often include backup servers that can take over the functions of primary servers that may fail. Server farms are typically collocated with the
network switch
A network switch (also called switching hub, bridging hub, and, by the IEEE, MAC bridge) is networking hardware that connects devices on a computer network by using packet switching to receive and forward data to the destination device.
A ...
es and/or
routers that enable communication between different parts of the cluster and the cluster's users. Server "farmers" typically mount computers, routers, power supplies and related electronics on
19-inch rack
A 19-inch rack is a standardized frame or enclosure for mounting multiple electronic equipment modules. Each module has a front panel that is wide. The 19 inch dimension includes the edges or "ears" that protrude from each side of the equ ...
s in a
server room
A server room is a room, usually air-conditioned, devoted to the continuous operation of computer servers. An entire building or station devoted to this purpose is a data center.
The computers in server rooms are usually headless systems that ca ...
or
data center
A data center (American English) or data centre (British English)See spelling differences. is a building, a dedicated space within a building, or a group of buildings used to house computer systems and associated components, such as telecommun ...
.
Applications
Server farms are commonly used for
cluster computing. Many modern
supercomputer
A supercomputer is a computer with a high level of performance as compared to a general-purpose computer. The performance of a supercomputer is commonly measured in floating-point operations per second ( FLOPS) instead of million instructio ...
s comprise giant server farms of high-speed
processors
A central processing unit (CPU), also called a central processor, main processor or just processor, is the electronic circuitry that executes instructions comprising a computer program. The CPU performs basic arithmetic, logic, controlling, a ...
connected by either
Gigabit Ethernet
In computer networking, Gigabit Ethernet (GbE or 1 GigE) is the term applied to transmitting Ethernet frames at a rate of a gigabit per second. The most popular variant, 1000BASE-T, is defined by the IEEE 802.3ab standard. It came into use ...
or custom interconnects such as
Infiniband
InfiniBand (IB) is a computer networking communications standard used in high-performance computing that features very high throughput and very low latency. It is used for data interconnect both among and within computers. InfiniBand is also use ...
or
Myrinet
Myrinet, ANSI/VITA 26-1998, is a high-speed local area networking system designed by the company Myricom to be used as an interconnect between multiple machines to form computer clusters.
Description
Myrinet was promoted as having lower protocol ...
.
Web hosting
A web hosting service is a type of Internet hosting service that hosts websites for clients, i.e. it offers the facilities required for them to create and maintain a site and makes it accessible on the World Wide Web. Companies providing we ...
is a common use of a server farm; such a system is sometimes collectively referred to as a ''web farm''. Other uses of server farms include scientific simulations (such as
computational fluid dynamics
Computational fluid dynamics (CFD) is a branch of fluid mechanics that uses numerical analysis and data structures to analyze and solve problems that involve fluid flows. Computers are used to perform the calculations required to simulate ...
) and the rendering of
3D computer generated imagery (see
render farm
A render farm is a high-performance computer system, e.g. a computer cluster, built to render computer-generated imagery (CGI), typically for film and television visual effects.
Origin of the term
The term ''render farm'' was born during the ...
).
Server farms are increasingly being used instead of or in addition to
mainframe computer
A mainframe computer, informally called a mainframe or big iron, is a computer used primarily by large organizations for critical applications like bulk data processing for tasks such as censuses, industry and consumer statistics, enterprise ...
s by large enterprises. In large server farms, the failure of an individual machine is a commonplace event: large server farms provide
redundancy, automatic
failover
Failover is switching to a redundant or standby computer server, system, hardware component or network upon the failure or abnormal termination of the previously active application, server, system, hardware component, or network in a computer net ...
, and rapid reconfiguration of the server cluster.
Performance
The performance of the largest
server
Server may refer to:
Computing
*Server (computing), a computer program or a device that provides functionality for other programs or devices, called clients
Role
* Waiting staff, those who work at a restaurant or a bar attending customers and su ...
farms (thousands of processors and up) is typically limited by the performance of the
data center
A data center (American English) or data centre (British English)See spelling differences. is a building, a dedicated space within a building, or a group of buildings used to house computer systems and associated components, such as telecommun ...
's
cooling systems and the total electricity cost rather than by the processors' performance. Computers in server farms run 24/7 and consume large amounts of electricity. For this reason, the critical design parameter for both large and continuous systems tends to be
performance per watt
In computing, performance per watt is a measure of the energy efficiency of a particular computer architecture or computer hardware. Literally, it measures the rate of computation that can be delivered by a computer for every watt of power con ...
rather than cost of peak performance or (peak performance / (unit * initial cost)). Also, for
high availability
High availability (HA) is a characteristic of a system which aims to ensure an agreed level of operational performance, usually uptime, for a higher than normal period.
Modernization has resulted in an increased reliance on these systems. F ...
systems that must run 24/7 (unlike supercomputers that can be power-cycled to demand, and also tend to run at much higher utilizations), there is more attention to power-saving features such as variable
clock-speed and the ability to turn off both computer parts, processor parts, and entire computers (
WoL and
virtualization
In computing, virtualization or virtualisation (sometimes abbreviated v12n, a numeronym) is the act of creating a virtual (rather than actual) version of something at the same abstraction level, including virtual computer hardware platforms, stor ...
) according to demand without bringing down services. The network connecting the servers in a server farm is also an essential factor in overall performance, especially when running applications that process massive volumes of data.
Performance per watt
The
EEMBC EnergyBench,
SPECpower {{Short description, Type of computer benchmarking tool
SPECpower_ssj2008 is the first industry-standard benchmark that evaluates the power and performance characteristics of volume server class computers. It is available from the Standard Performan ...
, and the
Transaction Processing Performance Council TPC-Energy are benchmarks designed to predict performance per watt in a server farm. The power used by each rack of equipment can be measured at the
power distribution unit.
Some servers include power tracking hardware so the people running the server farm can measure the power used by each server. The power used by the entire server farm may be reported in terms of
power usage effectiveness or
data center infrastructure efficiency.
According to some estimates, for every 100 watts spent on running the servers, roughly another 50 watts is needed to cool them.
For this reason, the siting of a server farm can be as important as processor selection in achieving power efficiency.
Iceland
Iceland ( is, Ísland; ) is a Nordic island country in the North Atlantic Ocean and in the Arctic Ocean. Iceland is the most sparsely populated country in Europe. Iceland's capital and largest city is Reykjavík, which (along with its ...
, which has a cold climate all year as well as cheap and
carbon-neutral
Carbon neutrality is a state of net-zero carbon dioxide emissions. This can be achieved by balancing emissions of carbon dioxide with its removal (often through carbon offsetting) or by eliminating emissions from society (the transition to the " ...
geothermal electricity supply, is building its first major server farm hosting site.
Fibre optic cables are being laid from Iceland to
North America
North America is a continent in the Northern Hemisphere and almost entirely within the Western Hemisphere. It is bordered to the north by the Arctic Ocean, to the east by the Atlantic Ocean, to the southeast by South America and th ...
and
Europe
Europe is a large peninsula conventionally considered a continent in its own right because of its great physical size and the weight of its history and traditions. Europe is also considered a Continent#Subcontinents, subcontinent of Eurasia ...
to enable companies there to locate their servers in Iceland. Other countries with favorable conditions, such as Canada,
Finland, Sweden and Switzerland,
are trying to attract cloud computing data centers. In these countries, heat from the servers can be cheaply vented or used to help heat buildings, thus reducing the energy consumption of conventional heaters.
See also
References
{{DEFAULTSORT:Server Farm
Cluster computing
Server hardware
Servers (computing)
Fault-tolerant computer systems