1.What is Grid computing?

Interconnected computer systems where the machines utilize the same resources collectively.
Grid computing usually consists oI one main computer that distributes inIormation and tasks to a
group oI networked computers to accomplish a common goal. Grid computing is oIten used to
complete complicated or tedious mathematical or scientiIic calculations.
Grid computing, or simply grid, is the generic term given to techniques and technologies
designed to make pools oI distributed computer resources available on-demand. Grid computing
was originally conceived by research scientists as a way oI combining computers across a
network to Iorm a distributed supercomputer to tackle complex computations. In the commercial
world, grid aims to maximize the utilization oI an organization's computing resources by making
them shareable across applications (sometimes called virtualization) and, potentially, provide
computing on demand to third parties as a utility service. When used with speciIications such as
WSRF and WS-NotiIication, grid resources can appear as web services within a service-oriented
architecture.
Grid computing (or the use oI a computational grid) is applying the resources oI many
computers in a network to a single problem at the same time - usually to a scientiIic or technical
problem that requires a great number oI computer processing cycles or access to large amounts
oI data. A well-known example oI grid computing in the public domain is the ongoing SETI
(Search Ior Extraterrestrial Intelligence) Home project in which thousands oI people are
sharing the unused processor cycles oI their PCs in the vast search Ior signs oI "rational" signals
Irom outer space. According to John Patrick, IBM's vice-president Ior Internet strategies, "the
next big thing will be grid computing."
Grid computing requires the use oI soItware that can divide and Iarm out pieces oI a program to
as many as several thousand computers. Grid computing can be thought oI as distributed and
large-scale cluster computing
and as a Iorm oI network-distributed parallel processing. It can be conIined to the network oI
computer workstations within a corporation or it can be a public collaboration (in which case it is
also sometimes known as a Iorm oI peer-to-peer computing).
A number oI corporations, proIessional groups, university consortiums, and other groups have
developed or are developing Irameworks and soItware Ior managing grid computing projects.
The European Community (EU) is sponsoring a project Ior a grid Ior high-energy physics, earth
observation, and biology applications. In the United States, the National Technology Grid is
prototyping a computational grid Ior inIrastructure and an access grid Ior people. Sun
Microsystems oIIers Grid Engine soItware. Described as a distributed resource management
(DRM) tool, Grid Engine allows engineers at companies like Sony and Synopsys to pool the
computer cycles on up to 80 workstations at a time. (At this scale, grid computing can be seen as
a more extreme case oI load balancing.)
Grid computing appears to be a promising trend Ior three reasons: (1) its ability to make more
cost-eIIective use oI a given amount oI computer resources, (2) as a way to solve problems that
can't be approached without an enormous amount oI computing power, and (3) because it
suggests that the resources oI many computers can be cooperatively and perhaps synergistically
harnessed and managed as a collaboration toward a common objective. In some grid computing
systems, the computers may collaborate rather than being directed by one managing computer.
One likely area Ior the use oI grid computing will be pervasive computing applications - those in
which computers pervade our environment without our necessary awareness.
Wbat is virtualization Wby Sbould You Care?
'irtualization is the creation oI a virtual (rather than actual) version oI something, such as an
operating system, a server, a storage device or network resources.
You probably know a little about virtualization iI you have ever divided your hard drive into
diIIerent partitions. A partition is the logical division oI a hard disk drive to create, in eIIect, two
separate hard drives.
Operating system virtualization is the use oI soItware to allow a piece oI hardware to run
multiple operating system images at the same time. The technology got its start on mainIrames
decades ago, allowing administrators to avoid wasting expensive processing power.
In 2005, virtualization soItware was adopted Iaster than anyone imagined, including the experts.
There are three areas oI IT where virtualization is making headroads, network virtualization,
storage virtualization and server virtualization:
0twork virtualization is a method oI combining the available resources in a network by
splitting up the available bandwidth into channels, each oI which is independent Irom the others,
and each oI which can be assigned (or reassigned) to a particular server or device in real time.
The idea is that virtualization disguises the true complexity oI the network by separating it into
manageable parts, much like your partitioned hard drive makes it easier to manage your Iiles.
Storag0 virtualization is the pooling oI physical storage Irom multiple network storage devices
into what appears to be a single storage device that is managed Irom a central console. Storage
virtualization is commonly used in storage area networks (SANs).
S0rv0r virtualization is the masking oI server resources (including the number and identity oI
individual physical servers, processors, and operating systems) Irom server users. The intention
is to spare the user Irom having to understand and manage complicated details oI server
resources while increasing resource sharing and utilization and maintaining the capacity to
expand later.
'irtualization can be viewed as part oI an overall trend in enterprise IT that includes autonomic
computing, a scenario in which the IT environment will be able to manage itselI based on
perceived activity, and utility computing, in which computer processing power is seen as a utility
that clients can pay Ior only as needed. The usual goal oI virtualization is to centralize
administrative tasks while improving scalability and work loads.
Wby Sbould You Care?
'irtualization can help you shiIt your IT Iocus Irom managing boxes to improving the services
you provide to the organization. II you are managing multiple servers and desktops,
virtualization can help you to:
O Save money Companles ofLen run [usL one appllcaLlon per server because Lhey don'L wanL Lo
rlsk Lhe posslblllLy LhaL one appllcaLlon wlll crash and brlng down anoLher on Lhe same machlne
LsLlmaLes lndlcaLe LhaL mosL x86 servers are runnlng aL an average of only 10 Lo 13 percenL of
LoLal capaclLy WlLh vlrLuallzaLlon you can Lurn a slngle purpose server lnLo a mulLlLasklng one
and Lurn mulLlple servers lnLo a compuLlng pool LhaL can adapL more flexlbly Lo changlng
workloads
O Save energy 8uslnesses spend a loL of money powerlng unused server capaclLy vlrLuallzaLlon
reduces Lhe number of physlcal servers reduclng Lhe energy requlred Lo power and cool Lhem
O Save t|me WlLh fewer servers you can spend less Llme on Lhe manual Lasks requlred for server
malnLenance Cn Lhe fllp slde poollng many sLorage devlces lnLo a slngle vlrLual sLorage devlce
you can perform Lasks such as backup archlvlng and recovery more easlly and more qulckly lL's
also much fasLer Lo deploy a vlrLual machlne Lhan lL ls Lo deploy a new physlcal server
O educe desktop management headaches Managlng securlng and upgradlng deskLops and
noLebooks can be a hassle ueskLop vlrLuallzaLlon soluLlons leL you manage user deskLops
cenLrally maklng lL easler Lo keep deskLops updaLed and secure
Wbat to Consider
Since virtualization makes it easy to set up new virtual servers, you may end up with a lot oI
servers to manage. Each server needs to be managed just as iI it was a physical server. Keeping
track oI where everything and how your virtual resources are using physical resources is
vital, so shop Ior solutions that have easy-to-use tools that help you monitor and measure use.
'irtualization isn`t a magic bullet Ior everything. While many solutions are great candidates Ior
running virtually, applications that need a lot oI memory, processing power or input/output may
be best leIt on a dedicated server.
For all oI the upside virtualization isn`t magic, and it can introduce some new challenges. But in
most cases the many cost and eIIiciency advantages will outweigh any issues, and virtualization
will continue to grow gain popularity.





'irtualization
Technology insiders tend to throw around technical terms and business jargon, assuming people
outside the industry understand what it all means. By its nature, technology vocabulary is oIten
conIusing and complicated, and insiders oIten add to the conIusion by over-complicating things.
To help add a sense oI clarity to the conIusion, each month, Laurie McCabe, a partner at Hurwitz
& Associates (a business consulting Iirm), will pick a technology term, explain what it means in
plain English, and then discuss why it may be important to you. This month Laurie looks at
'irtualization.
When people talk about virtualization, they`re usually reIerring to server virtualization, which
means partitioning one physical server into several virtual servers, or machines. Each virtual
machine can interact independently with other devices, applications, data and users as though it
were a separate physical resource.
DiIIerent virtual machines can run diIIerent operating systems and multiple applications while
sharing the resources oI a single physical computer. And, because each virtual machine is
isolated Irom other virtualized machines, iI one crashes, it doesn`t aIIect the others.
Hypervisor soItware is the secret sauce that makes virtualization possible. This soItware, also
known as a virtualization manager, sits between the hardware and the operating system, and
decouples the operating system and applications Irom the hardware. The hypervisor assigns the
amount oI access that the operating systems and applications have with the processor and other
hardware resources, such as memory and disk input/output.
In addition to using virtualization technology to partition one machine into several virtual
machines, you can also use virtualization solutions to combine multiple physical resources into a
single virtual resource. A good example oI this is storage virtualization, where multiple network
storage resources are pooled into what appears as a single storage device Ior easier and more
eIIicient management oI these resources. Other types oI virtualization you may hear about
include:
O -eLwork vlrLuallzaLlon spllLs avallable bandwldLh ln a neLwork lnLo lndependenL channels LhaL
can be asslgned Lo speclflc servers or devlces
O ppllcaLlon vlrLuallzaLlon separaLes appllcaLlons from Lhe hardware and Lhe operaLlng sysLem
puLLlng Lhem ln a conLalner LhaL can be relocaLed wlLhouL dlsrupLlng oLher sysLems
O ueskLop vlrLuallzaLlon enables a cenLrallzed server Lo dellver and manage lndlvlduallzed
deskLops remoLely 1hls glves users a full cllenL experlence buL leLs l1 sLaff provlslon manage
upgrade and paLch Lhem vlrLually lnsLead of physlcally
'irtualization was Iirst introduced in the 1960s by IBM to boost utilization oI large, expensive
mainIrame systems by partitioning them into logical, separate virtual machines that could run
multiple applications and processes at the same time. In the 1980s and 1990s, this centrally
shared mainIrame model gave way to a distributed, client-server computing model, in which
many low-cost x86 servers and desktops independently run speciIic applications.
While virtualization Iaded Irom the limelight Ior a while, it is now one oI the hottest trends in the
industry again, as organizations aim to increase the utilization, Ilexibility and cost-eIIectiveness
in a distributed computing environment. 'MWare, Citrix, MicrosoIt, IBM, RedHat and many
other vendors oIIer virtualization solutions.
Grid computing
Pooled computer resources. Grid computing, or simply grid, is the generic term given to
techniques and technologies designed to make pools oI distributed computer resources available
on-demand. Grid computing was originally conceived by research scientists as a way oI
combining computers across a network to Iorm a distributed supercomputer to tackle complex
computations. In the commercial world, grid aims to maximize the utilization oI an
organization's computing resources by making them shareable across applications (sometimes
called virtualization) and, potentially, provide computing on demand to third parties as a utility
service. When used with speciIications such as WSRF and WS-NotiIication, grid resources can
appear as web services within a service-oriented architecture.
A Iorm oI networking. Unlike conventional networks that Iocus on communication among
devices, grid computing harnesses unused processing cycles oI all computers in a network Ior
solving problems too intensive Ior any stand-alone machine.
A well-known grid computing project is the SETI (Search Ior Extraterrestrial Intelligence)
Home project, in which PC users worldwide donate unused processor cycles to help the search
Ior signs oI extraterrestrial liIe by analyzing signals coming Irom outer space. The project relies
on individual users to volunteer to allow the project to harness the unused processing power oI
the user's computer. This method saves the project both money and resources.
Grid computing does require special soItware that is unique to the computing project Ior which
the grid is being used.