Data Center is our focus

We help to build, access and manage your datacenter and server rooms

Structure Cabling

We help structure your cabling, Fiber Optic, UTP, STP and Electrical.

Get ready to the #Cloud

Start your Hyper Converged Infrastructure.

Monitor your infrastructures

Monitor your hardware, software, network (ITOM), maintain your ITSM service .

Our Great People

Great team to support happy customers.

Saturday, April 30, 2011

10 things to keep in mind when purchasing a new server

By Brien Posey | April 29, 2011, 3:17 PM PDT

A network server is a big investment and usually represents a long-term commitment, so it's critically important to select one that will meet all your needs. Here are some things you should consider when you go shopping for a new server.

Note: This article is also available as a PDF download.

1: Drivers

One of the first things you need to consider before making any purchasing decisions is whether the server you're interested in offers the necessary driver support. You shouldn't have any trouble getting Windows drivers as long as you're purchasing a server from one of the major manufacturers, such as Dell or HP. However, Linux drivers may not be as readily available. Taking a little time up front to make sure that you can get drivers for your intended operating system can save some heartache down the road.

2: Redundancy

If the server you're purchasing will be used for a mission-critical purpose, you need to make sure it uses redundant parts to avoid having a single point of failure. For example, the server should have at least two power supplies so that if one fails, the other can take over without the server going offline.

Some servers let you install a complete set of redundant memory, while others might include an extra slot you can use to install a spare memory module. The spare memory is used automatically if a memory failure occurs.

3: Hot-swappable components

In the 24/7 world of IT, taking a server down for maintenance practically requires an act of Congress. Try to make sure that the server you choose uses hot-swappable components. Granted, not every component is hot-swappable. For example, you can't swap out a system board or a CPU while the server is running. However, many servers support the use of hot-swappable hard drives, expansion modules, and power supplies.

4: Form factor

It should be fairly obvious that you need to consider the server's form factor when making a purchasing decision, but given the importance of choosing the right one, I wanted to go ahead and mention it anyway.

If you're purchasing a rack mount server, make sure that you have sufficient space left in your rack. Remember that 2U and 4U servers require more rack space than a 1U server does. Likewise, if you are planning to purchase a blade server, ensure that you have adequate space left in the blade server chassis.

5: Fault tolerant memory

Another thing you should look for when purchasing a new server is support for fault tolerant memory technologies, such as Error Correcting Code (ECC). ECC memory can dynamically correct single-bit memory errors. ECC memory can also detect (but not correct) double-bit errors.

6: Storage

Servers vary widely in terms of available internal storage. Although most of the servers on the market allow for the use of two internal hard drives, there are major exceptions. Some servers don't include any internal storage and may boot from a SAN instead. On the other hand, some of the larger form factor servers include support for large internal storage arrays.

Blade servers typically support only two internal drives, but storage can be expanded through a storage module, assuming that there is sufficient room in the chassis.

7: CPU support

You should also consider the server's CPU architecture. I'm not talking about Intel versus AMD (although that can be important, too). I'm talking about leaving room for future growth. Many of the servers on the market today offer multiple sockets that can support various types of processors. Organizations typically make a purchasing decision based on how many processor cores they need right now. However, it's a good idea to purchase a server that will allow you to add CPU cores down the road should the need arise. For example, you might start out with quad-core CPUs and then upgrade to six-core or eight-core CPUs later on. Likewise, you might initially fill one CPU socket but add another CPU when necessary.

8: Connectivity

It's easy to overlook network connectivity when purchasing a server because every server on the market includes built-in networking. However, network connectivity becomes far more important if the server is going to act as a cluster node or as a virtualization host server. Clustering and virtualization typically require more network adapters than a stand-alone server does. In such cases, a server probably won't have a sufficient number of network adapters by default, but you need to make sure it has the capacity to accommodate the additional required network adapters.

9: Memory capacity

If you plan to use your new server as a virtualization host, consider the amount of memory the server can support. Memory is the single most important resource in a virtual server environment, so you must ensure that the server includes enough memory to support all the guest operating systems without skimping on memory.

10: Manageability

If you have more than a handful of servers in your data center, make sure that your server supports manageability (both at the hardware and at the software level). Most of the servers on the market support hardware management through IPMI, which is a standardized management protocol. But software management tools tend to be proprietary, and one vendor's management software usually won't work with another vendor's servers.
build, access and manage your IT infrastructure and web applications

How to interview with a small company vs. a large one

By Toni Bowers | April 26, 2011, 6:34 AM PDT

Though "selling yourself" is still the goal whether you're applying at a small company or a large company, sometimes what you need to emphasize about yourself is different.

Many times in a small company, for example, you may interview with the top dog (the company's founder). This is a person who has a really personal stake in the success of the company, as it represents a substantial investment - in time and money - for him or her.

In a case like that, that person wants to see the same type of enthusiasm and passion from you. Be ware that you may also have to talk to several other people in the company from different departments. In smaller companies separate departments work closely so "fitting in" will be a major consideration here.

In smaller companies, you may not be offered any kind of training. It's important to exhibit in the interview that you are able to work without direct supervision, and that you're flexible and are willing to perform any task.

When it's your turn to ask questions, try to steer things toward company goals. You might be able to get an idea of the company's viability and potential for growth from the responses you get.

Lastly, be prepared to negotiate salary. Since this is not a company that's been around forever and has established pay rates for certain positions, you have some wiggle room.

Large companies

Now, with a large company, your first meeting will more than likely be with someone from HR. This person will be armed with standard, generic questions that are used to hit on the high points of your experience and to see if you drop all the keywords they're looking for (e.g., network administration, project management, Windows 7, etc.) You may even be asked to take a test to measure your technical skills.

I know many of our readers are frustrated by the HR part of the interview but if you're prepared and knowledgeable of what the position requires, then you should do OK.

build, access and manage your IT infrastructure and web applications