Servers

Data Center Upgrade: Not All Or Nothing

Data Center Upgrade.png

Everyone wants one screen to rule them all: one console display to monitor and manage all systems in a data center. According to some vendors, the only way to achieve this management “nirvana” is to bring in the forklifts and replace every system you own in one swoop. If you have an unlimited budget and no executive committee to report to, that's a suitable and realistic plan.

For the rest of us, the good news is that there is a better alternative: updating servers as they hit the end of their life cycle according to the refresh cycle. A mixed fleet of servers can be highly manageable if you consider a handful of key issues while you're building and evolving the servers within.

Picking The Server Management Framework

Several decisions follow the initial decision of which server management framework is selected. The basic split is between a framework from a hardware vendor and a framework independent of any hardware tie. While they are alike in many ways, there are key differences that will have major implications for your hardware choices.

Hardware-Tied Or Vendor-Neutral?

First, it seems obvious that a hardware-tied management framework should be at the top of your candidate list if all your servers are from a single vendor. While each new generation of servers has features that work more closely in concert with management applications, the vendor's software will typically work with at least three previous generations of hardware.

Vendor-neutral frameworks may lack the ability to take advantage of some specific server features, but they tend to offer consistent management across all servers of a particular generation and across two or three previous generations. They also can be cheaper depending on a multitude of factors. The real advantage of these frameworks involves existing analytics packages that you want to continue using. Integration with a wide range of third-party software is a strength of several vendor-neutral management systems.

Preparing For The Future

With all of these management frameworks, one of the most important considerations is how well the package prepares you for the future, since changing the software that manages a fleet of servers is not something to be taken lightly. Whether the management framework comes from a hardware vendor or not, it will be the tool that allows you to manage new servers and server blades as they are brought into service through the normal hardware refresh cycle.

As servers become part of a growing ecosystem of platforms that support virtual or software-defined functions, a management framework that supports all components of an integrated environment, from the server to storage to the network, becomes more important.

A single pane of glass that allows you to monitor and manage absolutely everything in the infrastructure is not yet available, but you can have a data center management system that will provide direct management of the servers in a diverse fleet while allowing integration with platforms that manage networking, storage, and other functions.

Another valid option for network monitoring and management is hiring a managed service provider (MSP). The advantages of hiring an experienced MSP like Current Technologies are numerous and include spending less on IT personnel, the MSPs are experts, you can rest easy knowing that someone else is protecting your network, and it is cheaper than and easier than doing it yourself.  It isn’t surprising that 70 percent of CIOs partnered with outside experts to plan manageable growth in 2018, find out what we can do for you today.

Experience The Current Technologies Advantage

Name *
Name

We Value Your Feedback

Was this information helpful?
Was this an interesting read?
CurrentTech_Horizontal.jpg

Servers Designed For The Real World

Server.png

Extreme conditions call for extreme servers


When most people think about servers and the rooms in which they live, they think of serious air-conditioning, dust-less raised floors, and row after row of pleasantly blinking lights to let everyone know that all is well. Sometimes in the real world of IT, the world where the data and applications actually live, there are no clean floors or carefully controlled temperatures.

A new generation of servers designed for the real world is emerging. With some tweaks, you can make the most adverse conditions tolerable for your hardware. So how do you deploy servers in those situations and have the confidence that they’ll be reliable for months or years on end?

1. Room To Breathe

The first parameter to consider is temperature. It is remarkable how many servers manage to function in small, unventilated closets with no real airflow and an in-closet temperature that would be reasonable in a sauna. 

Modern blade servers tend to be designed around convection cooling as well as fan-enhanced forced air. If you know that a server will be deployed in extreme conditions, don’t stuff it full of processors or storage boards. Allow the air to flow unimpeded between the components. Modern servers are capable of keeping themselves cool as long as internally air is allowed to flow.

2. Rack Space

Give some thought to how components are stacked in the rack, as well. In many cases, storage is placed at the bottom of the rack because it’s heavy and the stability is good. In extreme conditions, it can be worth looking for secondary sources of stability — bolting the rack to the floor wall — while the heat-generating spinning disks are situated above the processor units.

With careful rack construction, physics can work in your favor with convection currents adding to the airflow and heat dissipation.

3. Stop The Spinning

Another consideration is whether it might be possible to eliminate spinning disk storage altogether and replace it with solid state drives. This is one of the tradeoffs that will involve thinking about:

  1. The data that will be generated and used on site

  2. The budget for the system

  3. Whether network connectivity is available to make cloud or central storage a realistic possibility

4. And The Rest

Other considerations will include networking, backup, and connectivity for any on-site instrumentation that will be part of the deployment. With everything that is connected to the system, look for jacketed connectors and ask your vendor about rack fan units that can keep air moving in the warmest situations.

One more thing: monitor the environment. There are a number of possibilities for environmental monitoring and reporting, possibilities that range from those that are standard in blade server frames to separate temp/humidity/vibration reporting units. Current Technologies has been designing hardware networking and storage racks for 20+ years. Our experienced and knowledgeable team can deliver you a set up that will survive any conditions long into the future

After designing the server and rack properly, then keep an eye on the conditions inside the rack. There’s no reason why your server can’t survive in the most demanding circumstances.

Functionality In Every Climate

Name *
Name

We Value Your Feedback

Was this information helpful?
Was this an interesting read?
CurrentTech_Horizontal.jpg