Aurel Vlaicu University of Arad [607309]

”Aurel Vlaicu” University of Arad
Faculty of Exact Sciences
Applied Informatics

Distributed Systems
-Cluster Computing –

Made by:
Sorin Abruda n

Contents

1. Summary
2. History
3. Cluster models
4. Conclusion

Summary:

A computer cluster consists of a set of loosely or tightly
connected computers that work together so that, in many respects,
they can be viewed as a single system. Unlike grid computers ,
computer clusters have each node set to perform the same task,
controlled and scheduled by software.
The components of a cluster are usual ly connected to each other
through fast local area networks , with each node (computer used as a
server) running its own instance of an operating system . In most
circumstances, all of the nodes use the same hardware [1]and the
same operating system, although in some setups (i.e. using Open
Source Cluster Application Resources (OSCAR)), different operating
systems can be used on each computer, and/or different hardware. [
They are usually deployed to improve performance and
availability over that of a single computer, while typically being much
more cost -effective than single computers of comparable speed or
availability.
Computer clusters eme rged as a result of convergence of a
number of computing trends including the availability of low -cost
microprocessors, high speed networks, and software for high –
performance distributed co mputing .[citation needed ] They have a wide
range of applicability and deployment, ranging from small business
clusters with a handful of nodes to some of the fastest
supercomputers in the world such as IBM's Sequoia .

History:

Greg Pfister has stated that clusters were not invented by any
specific vendor but by customers who could not fit all their work on
one computer, or needed a backup. Pfister estimates the date as
some time in the 1960s. The formal engineering basis of cluster
computing as a means of doing parallel work of any sort was arguably
invented by Gene Amdahl of IBM, who in 1967 published what has
come to be regarded as the seminal paper on parallel processing:
Amdahl's Law .
The history of early computer clusters is more or less directly tied
into the history of early networks, as one of the primary motivations for
the development of a network was to link computing resources,
creating a de facto computer cluster.
The first production system designed as a cluster was the
Burroughs B5700 in the mid -1960s. This allowed up to four
computers, each with either one or two processors, to be ti ghtly
coupled to a common disk storage subsystem in order to distribute the
workload. Unlike standard multiprocessor systems, each computer
could be restarted without disrupting overall operation.
The first commercial loosely coupled clustering product wa s
Datapoint Corporation's "Attached Resource Computer" (ARC)
system, developed in 1977, and using ARCnet as the cluster interface.
Clustering per s e did not really take off until Digital Equipment
Corporation released their VAXcluster product in 1984 for the
VAX/VMS operating system (now named as OpenVMS). The ARC
and VAXcluster products not only supported parallel computing, but
also shared file systems and peripheral devices. The idea was to
provide the advantages of parallel processing, while maintaining data
reliability and uniqueness. Two other noteworthy early commercial

clusters were the Tandem Himalayan (a circa 1994 high -availability
product) and the IBM S/390 Parallel Sysplex (also circa 1994,
primarily for business use).
Within the same time frame, while computer clusters used
paralle lism outside the computer on a commodity network,
supercomputers began to use them within the same computer.
Following the success of the CDC 6600 in 1964, the Cray 1 was
delivered in 1976, and introduced internal parallelism via vector
processing . While early supercomputers exclu ded clusters and relied
on shared memory , in time some of the fastest supercomputers (e.g.
the K computer ) relied on cluster a rchitectures.

Cluster models:

In order to understand the cluster model from the viewpoint of
relations between firms, researchers have been defined different
models that take into account supplier chains relations, di rect Based
on specific characteristics of urban areas and clusters it is defined a
typology (Malmberg, Solvell, Zander, 1996) that describes four
different agglomerations, in figure 1, which highlights the conceptual
differences between the clusters and th e other three models.

Being a type
of economic
agglomeration,
clusters are formed by firms that conduct activities in the same field
and in which innovation is an important force that fuels the competition
and the firms development, (Porter, 1998 ), (Krugman, 1991). Based
on the role of different cluster members and the interaction between

them, Markusen (Markusen, 1996), has defined four models of
clusters.

Markusen
compare its models of modern clusters with the Marshal one, figure 2,
in which the cluster is rather homogenous, comprising small firms that
collaborate with each other, are in direct competition or in a supplier –
producer relation. In this model, none of the firms has the size and the
force to control directly the cluster and o nly the common market and
the cluster
dynamic define its
shape and
development.

In a hub -and-spoke cluster, figure 2, there are few dominant firms
that represent the core of the cluster and are surrounded by numerous
small firms that are linked d irectly to them. The most part of the cluster
firms represent suppliers of raw materials, of externalized services or

are specialized in a particular phase of the hub production process.
The small firms trade directly with the large ones and depend on thei r
client strategy. The hub firms define the relation inside the cluster and
its dynamics. Clear examples of hub -and-spoke clusters are found in
automotive industry, like Detroit Auto cluster that is concentrated
around the ―Big Three‖ auto manufacturers.

In a satellite
platform cluster, figure 4, a group of branch facilities of externally
based multi -plant firms, (Markusen, 1996) are located in a particular
geographic region in order to benefit from governmental facilities or
low costs with suppl ies and workforce. A particular characteristic of the
satellite platform is that there are no relations between satellite firms
and they are entirely
controlled by the
remotely located
parent firm.

The last category, the state centered (He and Fal lah, 2011) or
state anchored cluster, (Markusen, 1996) is defined around a public,
governmental or non -profit organization that dominates the region and
the economic relation between cluster members. This entity, which in
many US scenarios is a large milit ary base, is surrounded by
numerous small firms that benefit from public -private contracts. The
state centered type can be compared to a hub -andspoke cluster in
which there is one dominant key player that is not controlled by the
private sector.

Conclusions:
Recognizing the benefits of clusters as a form of economic
organization has influenced governments to implement policies
(Sölvell et al, 2003), intended to launch initiatives to support existing
clusters or to form new ones in reg ard with:
• Small and Medium Enterprises (SMEs)
• regional industrial development;
• attracting external funds and foreign investors;
• research and innovation at national or local level.
Economic development based on cluster models represent a
policy adopte d by many economies that can, theoretically, bring
multiple benefits in terms of regional development, competitiveness in
an industry. Also it can generate an economic environment that will
adapt more easily to events such as economic crises or other
econo mic and social transformation. Although existing cluster analysis
highlighted their advantages, the interconnection of factors and their
effect on the cluster, the economic theory has not yet provided a
model that allows both the analysis and the definitio n of a process for
implementing a successful cluster. Due to the complexity of the
economic relations and environment that define a cluster, it is
necessary to continue research from both theoretical and practical
perspectives, so that the theoretical mode l is validated in practice by
overlapping it with any existing cluster case.

Bibliography:

• https://en.wikipedia.org/wiki/Computer_cluster#History
• http://www.sintonia.mx/mm/cmfc.pdf