Assess data communication and networking options for a computer system

IT332_M4 Assess data communication and networking options for a computer system.

Purdue University Global

When looking into computer clustering and everything that goes along with it the

benefits, challenges, and potential applications it can be very confusing. Breaking it down to

one type and focusing on the one that makes the most sense for our company is where we

should place our focus. I want to place my focus on load balancing because it is one type of

clustering that will make the biggest overall impact to our company and how we operate .

Load balancing is a way for IT departments in companies or organizations to evenly

distribute workloads across many different servers or to allow computer resources to better

optimize capacity, network efficiencies and reliability. In order to use load balancing there

needs to either be a virtual or physical appliance to work through in order to determine in a

real time state which server in a server farm can best meet the load being sent out .

Depending on your load balancer set up during periods where the load is heavier the load

balancer can add servers to mitigate spikes in traffic and in the reverse of that drop servers

when demand is low. Load balancers can be a software application, an appliance, or a

combination of both. Network load balancing is a great feature that allows setup to use a

single IP address to balance the loads between two or more machines. You can setup IIS or

Internet Information Services on bother servers as well as Windows Operating System, as

well as a static IP address .

When looking at the server itself they come in multiple configurations such as blade, rack,

tower and all have various performance levels as well as space, budget and scalability

depending on their configuration. I want to focus on blade servers because they offer excellent

processing power in heavy dense environments and are small circuit boards that

really act as servers .

Our company could choose to also use a COTS server which is a commercial of the shelf

server. They offer versatility, tons of cost savings, ease of integration, & easy maintenance,

plus, their cost of ownership is lower, but when looking at the two blade servers are more

powerful and easier to service so there is a cost savings on the service end which can be

where you see a lot of cost over time .

Whatever server configuration you ultimately choose will in some way impact the way in

which your load balancing functions. Having a network of multiple devices that are dependent

upon consistent, fast, and efficient server quality and in order for the failover from server to

server to happen you will want a configuration that supports the software of appliance of the

load balancer.

Load balancing are usually agnostic applications that normally sit outside the server

application. Your load balancer is really in charge of the decisions on how and when your

network of devices functions. Load balancing controls what servers the devices travel through

and how to direct the traffic if one of the servers goes offline for some reason or has a failure

of any kind.

When looking at high availability which is the constant availability of a service which is why

load balancing redundancies is so lucrative. Although it is sought after it alone is not a stand

a lone strategy to ensure high availability in your system. High availability actually works to

address failure on many levels that include storage, network, and even the data center. Using

products across all your platforms in your network that utilize redundancies to ensure there is

not one single component failure .

Working in the IT world is fast moving and is always changing staying in tune with how

things operate and how to make the various components of your overall system function

together and reduce failure is key. Using load balancing on your servers reduces failure on a

server level. Also having monitoring on your system to watch functions in various levels of

the system from the database to the data center also helps track issues and see trends. Load

balancing offers the ability to help distribute your system load across many different servers

in the system. Having overall high availability also helps mitigate failure inside your system

on many levels .

When building your IT infrastructure its best to look at your company and what needs they

have look at what fail points they may currently have and what level of risk they are willing to

accept. Next you need to know what ramifications come from a system outage and what the

cost of being down would be for that company. If you work for a company that is dependent

upon network like a gaming company having an outage is not only inconvenient but is also

impacting paying customers. If you take the flip side of a small candy shop it may not be

convenient if their system is partially down or down completely because they probably have a

plan B for that situation. Building your system to deal with redundancy and high availability

really comes down to the type of company you’re dealing with.

All companies would love to say they are 24/7 and are never down, but in reality, there are

outside forces you have to deal with as well as internal. Since ensuring that everything will

run 24/7 it is best to focus on the various ways to help your company deal with the risk of

being down and having redundancy where it is available and having a plan of action where it

is not. If cost is a factor and you can’t make it all redundant that what major components can

you afford redundancy in.

Talking to your CTO and business leaders will help address where they want to focus their

money and at what levels of risk, they are willing to accept. Having redundancy and knowing

your weak points will help you have an overall picture of your system and how to best

maintenance it will also help as you scale the system up where to put focus on areas that may

be lacking redundancy of any kind.

Part B