Data center administrators should consider the following best practices when physically securing and protecting their network and server hardware in the data center:

1)      Secure Physical Access. Gaining physical access to a server provides direct access to hard drives and sensitive data.  Server access should be limited to a small group of authorized administrators.  This allows a company to track changes to the physical server environment and ensure service availability on the network.

2)      Monitor Access & Activity. Log, correlate and manage security and audit event information. Aggregate and standardize security event information to provide a consolidated view of security.

3)      Surveillance. Surveillance cameras should monitor the perimeter of the building, at all entrances and exits, and at every access point throughout the data center. Video monitoring can also snap pictures of the individuals entering a server room or approaching a server. This is useful in the event of downtime, theft of property, and the event of damage to property.

4)      24×7 Backup Power.  Data center systems should be functioning at all times, complete with uninterruptible power supply (UPS) to ensure continuous operation in case of disaster or unexpected power loss.

5)      Cooling. The introduction of blade servers created a situation where there is more concentrated heat in enclosed area. HVAC are recommended to account for future expansion to make sure systems do not overheat. The data center cooling systems should be designed to maintain cooling of the data center between 70–74 degrees F, with relative humidity between 45%–50%.

6)      Fire detection and suppression. A comprehensive fire detection and suppression system is a major priority to ensure that devices and data are protected. Review local ordinances to determine appropriate options for fire suppression.

7)      Safely Handle Equipment.  Servers are expensive and store mission critical data.  So it is important to use the right tools to mitigate the risk of damage and use. The safest and most effective method of handling servers is to use a specialized lift such as the ServerLIFT® SL500 Server Lifter.

Often times, servers and other data center equipment are handled using manual lifting techniques. Servers often weigh between 50 and 100 pounds and racks can typically hold between 10 – 40 servers depending on the size and type of equipment. Manually moving and lifting the equipment into the racks requires extensive labor and resource, and is not very efficient.

Even with a high level of expertise, IT professionals face potential injury and risk when asked to manually lift or move servers in a data center environment.  Work-related injuries can be very costly, so it is vital to spend adequate time, money and planning to ensure a safe work environment. The U. S. Bureau of Labor Statistics study on Nonfatal Occupational Injuries and Illnesses Requiring Days Away from Work 2009 reported:

-  Sprain, strain, and tear injuries accounted for 40% of total injury and illness cases requiring days away from work in all ownerships; 37% were back-related
-  42% of sprains, strains, and tears were the result of overexertion
-  Musculoskeletal disorders (MSDs), often referred to as ergonomic injuries, accounted for 28% of all workplace injuries and illnesses requiring time away from work in 2009; 60% were of the back and required a median of 7 days to recuperate

There was an instance where a federal contractor sued IBM for $1.4 million after a server fell off a non-specialized forklift and was irreparably damaged while being moved into the customer’s data center.

When selecting a lift to help for your data center, it’s essential to consider specialized lifts specifically designed to handle expensive, heavy, dense, and delicate IT equipment.   A Server Lifter provides a safer and more efficient method for handling servers and other IT assets.

Choosing the correct Server Lifter is a vital decision in making sure the job is done safely and efficiently. In selecting the proper tool for handling your servers and IT assets, there are several important factors to consider:

-          Are you moving heavy equipment? How heavy is it?
-          How many people are involved when you rack equipment?
-          Do you have problems reaching the top of your rack?
-          How narrow are your data centers aisles?

-          What is your risk tolerance?

Industry leading organizations rely on ServerLIFT server handling solutions to reduce risk of injury, improve data center efficiency, reduce operating costs and streamline the server deployment process.

Learn more about our ServerLIFT server handling solutions

According to research firms IDC and Gartner, revenue in the worldwide server market increased between 15.3%-16.4% percent year over year to $15.0 billion in the fourth quarter of 2010 (4Q10). This was the highest quarterly revenue in three years and the fourth consecutive quarter with improving year-over-year revenue growth as the market continued to accelerate throughout the calendar year of 2010.

“The 2010 growth was a combination of pent-up demand and the replacement cycle”, said Jeffrey Hewitt, Research Vice President at Gartner.

Gartner released their worldwide server market research findings on Feb. 24 2011. Days later, IDC came out with their 4Q 2010 worldwide server market revenue report on Feb. 28, 2011. Both reports were very consistent in their findings.

The ‘Big Three’ Server Vendors

IBM held the top position in the worldwide server systems market with 35.5%-37.4%  market share in factory revenue for 4Q10, as revenue increased 21.9%-26.4%  year over year. IBM had strong improvement in demand for its System z mainframe systems and continued demand for x86-based System x servers in the quarter.

HP came in second with 29.9%-30.4% share for the quarter as revenue increased 12.8%-13.2% percent compared to 4Q09. HP was helped by continued demand for its x86-based ProLiant servers.

Dell maintained third place with 12.6%-13.1% revenue market share in 4Q10. Dell experienced 26.4%-26.8% revenue growth compared with 4Q09 due to strength in demand from SMB, enterprise, and public sector customers.

—-

Below is a full breakdown of the state of the worldwide server market for the fourth quarter of 2010 and all of 2010, as compiled by IDC and Gartner

4Q10 Worldwide Server Market Share

Notable highlights:

- Worldwide server revenue grew by more than 15% in 4Q10
- Dell and IBM had the largest server revenue growth by around 26% in 4Q10
- IBM handily claimed the server market share lead in 4Q10
- HP lead the blade server market, with a 53.4% revenue share in 4Q10

Gartner: Worldwide Server Vendor Revenue Estimates, 4Q 2010 (U.S. Dollars)

Source: Gartner (February 2011)

IDC:  Worldwide Server Vendor Revenue Estimates, 4Q 2010 (Revenues are in Millions)

Source: IDC (February 2011)

2010 Worldwide Server Market Share

Notable highlights:

- Worldwide server revenue grew by more than 11% in 2010
- The Big Three server vendors, IBM, HP, and Dell, earned around 75% of worldwide server revenue of $48 billion
- HP and IBM remained in a virtual tie
- Oracle saw its server revenue decline by 14.0%-17.7% in 2010

Gartner: Worldwide Server Vendor Revenue Estimates, 2010 (U.S. Dollars)

Source: Gartner (February 2011)

IDC:  Worldwide Server Vendor Revenue Estimates, 2010 (Revenues are in Millions)

Source: IDC (February 2011)

Here’s a look back at some of the milestones in the evolution of server hardware in the data center:

[slideshow]

As servers evolve, the other technologies in the data center surrounding servers must evolve as well.  The cutting edge designs of ServerLIFT® products focus on safety, efficiency and productivity to help your data center operations meet these evolving needs.

What is a Blade Server?

A blade server is a compact and dense server with its own processor and memory.

Unlike traditional rack servers, blade servers share networking cables, switches, power, and storage with other blade servers in an enclosure (chassis) designed specifically to house and manage them.

Components of Blade Servers

Pictured here are the major components of an IBM HS22V Blade Server:

IBM: Major Components of the IBM HS22V Blade Server

Pictured here are the major components of a typical blade server chassis:

HP: Blade Server Chassis

Benefits of Blade Servers

Blade servers offer significant advantage over traditional rack servers – improving the processing ability while consuming less power per server. The servers, the chassis and all the components of the integrated rack work together as a seamless, efficient unit. This increases efficiency and reduces costs by eliminating many of the overlapping resources often required to run stacks of individual rack servers.

A study by HP identified significant benefits when blade servers are deployed in the data center, including:

- Lower acquisition costs, including networking integration … because ports can be aggregated of and components (like management modules) shared.
- Lower operational costs for deployment, troubleshooting and repair … thanks to reduced complexity, and aggregation of components
- Lower power, cooling, and space requirements … resulting in more efficient power utilization and increase density computing power
- Improved reliability … because of fewer components, thus fewer places for failure
- Manageability improvements … a chassis-level shared view of all assets simplifies resource management and creates efficiencies Blade server chassis’ tend to be considerably heavier than rack servers, and thus more awkward to install or move.

Blade server chassis’ tend to be considerable heavier than rack servers, and thus more awkward to install or move. ServerLIFT® products are designed to smoothly install, uninstall and relocate servers of all kinds, providing a speedy, safe and economical solution to this issue.

Whatever your materials handling need may be, there’s probably a specialized tool to fill it. For instance, each of the lifts shown here has its own set of features and benefits, depending on the type of move or materials involved.

Forklift Types (industrialforklifttruck.org)

Forklift Hazards

Regardless of the type of lift being used, forklift safety should be a major concern. With OSHA estimates of 1 million forklifts in workplaces across the country, forklifts are not only among the most common type of motorized equipment, they also cause many of the most serious workplace accidents – more than 20,000 forklift-related injuries each year.

Employees are often injured or killed when forklifts:
• Tip over
• Drop loads
• Strike employees working nearby
• Battery charging and fueling accidents

Most incidents involve not only injury, but property damage as well: overhead sprinklers, racking, pipes, walls, materials, and machinery can all be damaged or affected.

According to OSHA, most employee injuries and property damage are caused by:
• Lack of safe operating procedures
• Inconsistent or nonexistent safety rule enforcement
• Insufficient or inadequate training

OSHA Forklift Safety Tips and Best Practices

Setting up these OSHA forklift safety tips with your employees will help promote workplace safety:
1. Forklift operators must be 18 or older.
2. Only people properly trained on forklifts should drive them.
3. The forklift should be stacked correctly to insure nothing falls off during lifting.
4. Take turns slowly to prevent tipovers.
5. All forklifts should have a seat belt, that operators should wear it at all times.
6. The operator’s view should be clear of any obstructions caused by a load before driving.
7. NEVER overload the lift above its maximum capacity.
8. Never leave the forklift running unattended; the operator should switch it off with every
dismount.
9. Always use the horn to warn others while backing up. Also sound the horn when entering doorways or any place
where vision may be obstructed.
10. NEVER lift another living being: human or pet. Forklifts are made for freight, and lifting people
can result in serious injury or death.

Specialized Material Handling Lifts: Moving Servers

In selecting the proper tool for your material handling needs, there are several important factors to consider:
How heavy and what size are your typical loads?
How high do you need to lift the load?
How much room do you have to maneuver?
How wide are your narrowest aisles?

A federal contractor sued IBM for $1.4 million after a server fell off a non specialized forklift while being moved into the customer’s data center, and was irreparably damaged.

Moral: Addressing these issues, it’s essential to consider specialized lifts, purpose- designed to handle specific materials safely and efficiently. Especially when moving or racking expensive, delicate servers.

One such is the ServerLIFT® SL500 Server Lifter – uniquely designed to lift and move servers quickly, effectively and safely; and to accelerate the server deployment process in any data center environment.

While mechanical lifts have been common on shop and factory floors for decades, only the ServerLIFT® SL500 server lifter has been designed exclusively and specifically for the tasks of moving and lifting rack-mounted data center equipment.

The ServerLIFT® SL500 offers the following exclusive features:
• Compact Design. Its slim 24″ base allows for easy navigation down narrow data center aisles.
• Side-Loading Usability. A clever side shifting platform adds 6 inches of travel to the right or left
side for effortless server installation by a single employee.
• Safe for data center floors. The ServerLIFT® SL500 is hydraulics-free and has a leak-proof
battery

These specialized design features not only provide increased safety, efficiency and productivity, but actually reduce costs measurably, to benefit the user’s bottom line.

Side-loading tray makes installation a snap.

*The Occupational Safety & Health Administration

To ensure a safe working environment, it is important never to overlook the potential risks the data center and its personnel may be exposed to. These include fire, natural disasters, faulty equipment and outages, electrocution, physical injuries, and the list goes on. These risks can be mitigated and avoided with proper inspections, procedures and/or training.

Failure to prepare for such risks can not only cause costly outages in your data center, but also result in an OSHA noncompliance issue or injury to an employee. Safe workplaces don’t just happen – they’re created when data center safety is a primary concern and when active steps are taken to plan a safe work environment.

Want to make your data center workplace as safe as it can be?

Follow these Data Center Safety Tips:

1. Design an Organized and Safe Data Center. Data Center Safety should be planned into your design from the start. By analyzing the physical steps that operators take to load and unload servers … to relocate racks … to monitor consoles … and to perform other routine physical tasks, you can develop a well-designed floor plan that minimizes time and motion and maximizes safety.

2. Eliminate Data Center Hazards. One of the simplest actions you can take to improve data center safety is to remove hazards like network cables (as in the spaghetti dinner below), or holes in raised flooring that employees might trip on. While that sounds simple and straightforward, it’s all too often neglected when facilitating server upgrades or network expansions. Such hazards can be dangerous, with the real possibility of physical injury.

Look Familiar… We Hope Not

3. Inspect and Maintain Water and Fire Detection Systems. Most data center maintenance focuses on testing the reliability of backup electrical systems (UPS’s, batteries, generators, power distribution units), but there typically isn’t much focus on fire detection and suppression systems. It is absolutely critical that you regularly schedule inspections and maintenance of these systems, as well as keep your evacuation plans fully up to date.

4. Provide Clear Work Instructions. Make sure your employees know precisely what you expect of them by providing thorough training and clear, written instructions or unmistakable diagrams such as those below. When you document your work processes, make sure you include basic data center safety instructions for lifting heavy objects, or operating a server lift.

BE SURE TO PROVIDE CLEAR Data Center SAFETY INSTRUCTIONS LIKE THIS DIAGRAM

5. Watch and Learn How Each Employee Performs Their Job. Closely watch how your employees do their jobs to make sure they’re following procedures … or if they’re taking shortcuts that could compromise data center safety and expose them to injury or other risks.

6. Ensure Your Staff is Prepared for Natural Disasters. Make sure your staff is trained on preparedness for natural disasters like floods, tornadoes, hurricanes, and earthquakes, particularly in geographic areas most prone to these types of disasters. Verify common practices, such as anchoring equipment, latching cabinets, and properly storing materials. Procedures involving evacuation routes, holding areas and communication should all be tested and well understood.

These data center safety tips are simple and effective to use. Combined with the elimination of common risk exposures, they can significantly enhance the safety of your data center.

Efficiency and effectiveness are top priorities in data centers.  Attention is given to performance and hardware, including servers, storage, backups, recovery, power, and cooling.  The most critical pieces which create a successful data center also create a more dangerous work environment.

Maintaining a safe data operation for all personnel is commonly under prioritized (or overlooked).  Work-related injuries can be very costly, so it is vital to spend adequate time, money and planning to ensure a comfortable and safe work environment.

The Risks of Heavy Lifting in the Data Center

A common task for data center workers is moving and lifting servers, networking equipment and power supplies as they reorganize, expand or migrate to a new location. These tasks are unsafe and sometimes cause injury and data damage when performed without the proper tools.

The U. S. Bureau of Labor Statistics study on Nonfatal Occupational Injuries and Illnesses Requiring Days Away from Work 2009 reported:

-  Sprain, strain, and tear injuries accounted for 40% of total injury and illness cases requiring days away from work in all ownerships; 37% were back-related
-  42% of sprains, strains, and tears were the result of overexertion
-  Musculoskeletal disorders (MSDs), often referred to as ergonomic injuries, accounted for 28% of all workplace injuries and illnesses requiring time away from work in 2009; 60% were of the back and required a median of 7 days to recuperate

The OSHA Technical Manual on Back Disorders and Injuries reports that Back disorders:

-  One of the leading causes of disability for people in their working years
-  Afflict over 600,000 employees each year
-  Cost of about $50 billion annually
-  Up to one-third of compensable back injuries could be prevented through better job design

Techniques to Help Prevent Lifting-Related Injuries in the Data Center

The Occupational Safety & Health Administration is an agency dedicated to help employers and employees reduce workplace injuries, and has developed a number of resources, guidelines and regulations for lifting in the workplace.  Data center training and job performance should be in compliance with OSHA regulations.

OSHA has looked at both major categories of methods for preventing lifting injuries:  administrative controls and engineering controls. Administrative controls include carefully selecting and/or training workers so they can safely perform lifting tasks.

Engineering controls attempt to redesign a job so lifting becomes less hazardous.  Primary among these are installation of mechanical aids such as pneumatic lifts, conveyors, and/or automated materials handling equipment, and also include adjustment of the physical environment (height of shelves, reduction in size or weight).  Since most racks and equipment sizes are standardized, the optimal solution is a pneumatic lift machine such as a ServerLIFT.

Newer rack systems are desinged to be mobile with equipment installed.  Unless your server racks are specifically designed to be moved with equipment installed, you should NEVER move racks that are fully loaded.

Consider the following when moving server racks:
-Remove all components from server racks before a move and reinstall when the rack is in place
-Move empty server rack by pushing as low on the rack as possible to avoid tipping
-This is a good time to reorganize of cable management if sloppy
-If you are shutting down the servers, this is also a good time to install any firmware updates

 
Reasons not to move server racks when fully loaded:
-The server rack could collapse or fall and injure personnel
- The rails could get bent rendering them useless (bent rails are difficult, if not impossible, to bend back)
- The server rack could collapse or fall and damage all of your equipment
- Most server racks, wheels, bolts, mounting hardware, etc… are not rated to handle being moved while fully loaded
- Internal components (i.e. hard drives, CPUs, fans, etc…) are at risk of shock damage when being moved while mounted
- The move could cause a disk crash to your server and loss of data

Some quotes to consider (from Dell servers – Moving a rack):

“Definitely remove everything! I pushed a rack down a cable hole once and that was scary enough and it was only tilted a few degrees over (I wasn’t even trying to move it). Sounds like a good way to waste a weeks work anyway “

“Safest is certainly to strip it down first. Personally though when I had to move a rack I just cranked it back up on it’s wheels and carefully pushed it to where I wanted it – the biggest hassle was keeping the cable spaghetti coming out of the back of the rack from getting in the way.”

(Could have saved some time and heavy lifting by utilizing the SL 500 Server Lifter)

In creating an efficient data center, pay special attention to the space and design as well as implementing the right technologies.   The success of a data center design is realized through the long-term scalability, flexibility and availability of the facility.

To help optimize network performance, create operational efficiencies, achieve optimal availability and produce cost savings we suggest:

-  The design should be flexible, adaptable and agile. The facility and technologies should be ready to change and align with future business objectives and requirements.

-  Design should emphasize appropriately right-sizing IT equipment to support cabling, power, and cooling infrastructures.

-  Data centers should be designed to anticipate next-generation equipment, which is likely to have even-higher power, cooling, and cabling requirements. This will help avoid the cost and disruption of frequent technology refreshes and additional infrastructure.

-  It is important to provide adequate, scalable power for the load. Comprehensive load analysis can produce a reasonable estimate of the facility’s power requirements.

-  “Going green” is not only sustainable, but makes good business sense, especially when it involves simple techniques such as using outside air for cooling.  That can save significant amounts of money over time.

-  Data centers should be equipped with intelligent monitoring. When evaluating monitoring systems, look for solutions that have intuitive alarming, reporting and escalation features.

That wraps up the Data Center Efficiency Series. Please feel free to leave any comments or questions that you may have!