Managing IT resources with an agent<\/p>\n
By Zatni Arbi<\/p>\n
JAKARTA (JP): Imagine yourself working for a bank with
\nhundreds of branches scattered all over Indonesia as the manager
\nresponsible for its huge, distributed network. How can you make
\nsure that all your company's IT resources are well-maintained and
\nthat their capacity is always adequate?<\/p>\n
With today's computer hardware and software technologies, you
\nwon't have to physically travel around the country and visit each
\nof the branches to check each individual PC, server, mini and
\neven mainframe computer that it may have. Nor will you have to
\nhire highly skilled -- and highly paid -- technical support personnel
\nto man the help desks. A type of software suite called
\n\"enterprise management framework\" will do it for you. Currently
\ntwo strong contending products are fiercely battling it out in
\nthis field: Tivoli Enterprise from Tivoli (an IBM company) and
\nUnicenter TNG (TNG stands for 'The Next Generation') from
\nComputer Associates.<\/p>\n
These suites of end-to-end management tools promise to help
\nlarge corporations manage their widely dispersed IT resources
\nthrough a centralized control system by automating a lot of
\nfunctions, including maintenance and asset management. As you can
\nimagine, in such a huge environment there may be UNIX servers, NT
\nservers, OS\/2 servers, a Novell network operating system and many
\nother platforms. Because of such diversity, these \"umbrella\"
\ntools are supposed to be able to also work on top of all these
\ndifferent platforms and facilitate management so that the
\nutilization of the resources can be maximized and disruption can
\nbe minimized.<\/p>\n
Two typical problems that IT personnel in such huge computing
\nenvironments usually face are downed servers and an unexpected
\njump in the number of users that means unanticipated increased
\ndemands for resources. Unlike our home PC, which we can replace
\nat any time if it crashes, the servers in these organizations
\ncannot stop functioning just because maximum capacity has been
\noverstepped or one component has failed. Just imagine how
\nfrustrated we will become each time we are told by the teller
\nthat the bank's server is down and we cannot make any
\ntransaction.<\/p>\n
Both Tivoli Enterprise and Unicenter TNG use the so-called
\n'agents' that are deployed to the individual resource in the
\nenterprise network, including the PCs and the notebooks. These
\nsoftware agents regularly collect data from their assigned
\nlocation and report any extraordinary event in their target
\nenvironment to the central management software. Based on these
\nreports, the IT personnel, who may be notified through pagers or
\nE-mail in case the light has turned from green to yellow, will
\nthen be able to manage the network better by reallocating
\nresources as needed, take preventive action or pinpoint the
\nsource of a problem quickly when it occurs.<\/p>\n
The increasing importance of complete management of IT
\nresources has led companies such as Computer Associates (CA) to
\nfind a new way of simplifying the immensely complex task. One
\nreasonable option is to use a neural network to predict when
\ncapacities will be used up to their maximum so that actions can
\nbe taken to prevent disasters.<\/p>\n
Neural Agent<\/p>\n
You may still recall the technology called neural networks.
\nSimply put, a neural network is an attempt to imitate the
\nsophistication and intelligence of the human brain. It is called
\na network because it consists of numerous \"units\" of simple
\nprocessors--each with its own memory--that are connected to each
\nother to form a system that tries to mimic the brain.<\/p>\n
The most important aspect of the resemblance is the ability to
\nsort out patterns and learn from trial and error. With it comes
\nthe ability to discern and extract the relationships underlying
\nthe data that we feed it with. With these two abilities, the
\nartificial intelligence system can even make decisions based on a
\nset of dynamic rules. In addition, the more data it has the more
\nintelligent a neural network will become.<\/p>\n
As an application of the neural network, the new Neugents from
\nCA also learns as it goes. The name Neugents actually comes from
\n'Neural' and 'Agent'. This application is actually a part of CA's
\nstrategy in moving to its next version of Unicenter TNG, i.e.,
\nUnicenter TND (D for 'Dimension').<\/p>\n
What, specifically, does Neugents do to help simplify the
\nmanagement of an enterprise IT infrastructure? Its agents, which
\nare deployed to mission critical servers, will continuously
\nmonitor a number of parameters in the systems. Based on these
\nparameters and with the help of the historical data that has been
\ncollected at the Unicenter TNG, Neugents will then be able to
\nmake a predictive statement such as \"There is a 70 percent
\nprobability that Server X will crash in an hour.\"<\/p>\n
The server crash may be caused by overloading, for example, as
\nthe number of its users keeps rising while its capacity remains
\nconstant. When the IT administrator sees such a message in his
\nUnicenter console, he will then be able to make a proper decision
\nto prevent the crash, for example by temporarily shutting down
\nother services and allocating the freed resources to accommodate
\nthe increasing number of users at server X.<\/p>\n
Can a human being do what Neugents does? As Kieandy Susanto
\nfrom CA reminded me, human beings can only process information in
\na limited number of dimensions, perhaps three of them at the
\nmost. A computer server, on the other hand, has tens of
\nparameters that determine its overall performance, including
\nmemory, hard disk, I\/O and processing capacities. It will be very
\nhard for a human brain to process all these data and make a quick
\ndecision based on them. A computer, on the contrary, has no
\nproblem handling multidimensional data. Herein lies the benefit
\nof Neugents. It can process data in dozens of different
\ndimensions and make a prediction based on them.<\/p>\n
Interestingly, although we may not realize it, the neural
\nnetwork-based systems and parent technologies (artificial
\nintelligence and expert systems) are already widely in use all
\nover the world although they have long ceased to be buzzwords.
\nThe latest generation of optical character recognition and voice
\nrecognition tools already incorporates the neural network
\ntechnology, and so do the systems that are used by banks to
\nestimate investment risks, for example. However, CA has been the
\nfirst to bring the technology to the enterprise management tools,
\nand it may be some time before other vendors will follow suit.<\/p>",
"url": "https:\/\/jawawa.id\/newsitem\/managing-it-resources-with-an-agent-1447893297",
"image": ""
},
"sponsor": "Okusi Associates",
"sponsor_url": "https:\/\/okusiassociates.com"
}