Important factors within a data center
A recent article from ZDNet Asia author Erik Eckel relayed that not every data center is currently up to task.
By Donna DonowitzFebruary 25, 2011
A recent article from ZDNet Asia author Erik Eckel relayed that while cloud computing is advantageous for many companies because of its benefits pertaining to data storage, virtualization and remote access, not every data center is currently up to task. For these centers not “ready for prime time,” Eckel discussed several factors companies need to focus on to improve their quality.
First, data capacity is a major factor at data centers, so companies need to make sure their systems can handle significant amounts of data transmission and can scale appropriately. Next, centers need to have secondary electrical sources, or else they run the risk of catastrophe if the initial power is knocked out.
Like the power issue, data centers need to have backup internet access as well - secondary and tertiary, just in case. “Buried cables get cut. Overhead cables fall when trucks strike poles. Vendors experience network-wide outages,” the article states. Thus, companies need to make sure multiple internet backups are in place.
While having backup power and internet is important, even essential, configuring these systems as hot online spares for automatic hardware failover is just as important. “It's critical that data centers employ redundant online switches, routers, UPSes, and HVAC equipment that automatically fail over when trouble arises,” according to Eckel.
Data centers need access control, too. These centers need to be protected from any potential detractors to ensure the security of the sensitive data. Physical access controls, such as keys and scanner cards, are important, as they can block access to certain sections if someone were to gain general access. Also, centers need to have security and support at all times, “24x7x365,” as Eckel terms it. “Systems can't fail. Constant monitoring and maintenance is a must,” the article states. These centers need to employ in-house break or fix service should a server, system, piece of software or anything else go down.
Centers should also have independent power, which will eliminate heavy reliance on other companies should a problem occur. Written service level agreements, with specific words are also necessary in order for the data center to guarantee certain factors, such as uptime, service response, bandwidth and physical access protections. SLAs should also state what the process is should a data center fail to meet these terms.
Focusing on these areas is important, especially considering how companies are being held to strict compliance in regards to data storage. Various regulations are now in place, which companies must comply with, based upon the physical location of these centers. This is yet another area companies need to pay strict attention to.
First, data capacity is a major factor at data centers, so companies need to make sure their systems can handle significant amounts of data transmission and can scale appropriately. Next, centers need to have secondary electrical sources, or else they run the risk of catastrophe if the initial power is knocked out.
Like the power issue, data centers need to have backup internet access as well - secondary and tertiary, just in case. “Buried cables get cut. Overhead cables fall when trucks strike poles. Vendors experience network-wide outages,” the article states. Thus, companies need to make sure multiple internet backups are in place.
While having backup power and internet is important, even essential, configuring these systems as hot online spares for automatic hardware failover is just as important. “It's critical that data centers employ redundant online switches, routers, UPSes, and HVAC equipment that automatically fail over when trouble arises,” according to Eckel.
Data centers need access control, too. These centers need to be protected from any potential detractors to ensure the security of the sensitive data. Physical access controls, such as keys and scanner cards, are important, as they can block access to certain sections if someone were to gain general access. Also, centers need to have security and support at all times, “24x7x365,” as Eckel terms it. “Systems can't fail. Constant monitoring and maintenance is a must,” the article states. These centers need to employ in-house break or fix service should a server, system, piece of software or anything else go down.
Centers should also have independent power, which will eliminate heavy reliance on other companies should a problem occur. Written service level agreements, with specific words are also necessary in order for the data center to guarantee certain factors, such as uptime, service response, bandwidth and physical access protections. SLAs should also state what the process is should a data center fail to meet these terms.
Focusing on these areas is important, especially considering how companies are being held to strict compliance in regards to data storage. Various regulations are now in place, which companies must comply with, based upon the physical location of these centers. This is yet another area companies need to pay strict attention to.
