Issue 32
Publisher: Jack Burlin Editor: Patti Hammonds
February 3, 2008
IN THIS ISSUE
Your success story
Monthly featured product
Kevinisms
Trivia
Articles of Interest: Temperature's Rising . . . Or Not? The "Follow Me" Truck
| Jack, Here is the photo I promised you. It is a Tactical Data Network (TDN) Gateway mock-up. It has 1/16 the capability of the real system at 3% of the cost. I used many different vendors to include ISC. I went with ISC because I could get a lot of the items I needed at one stop and the price was very competitive. The quality of the products is superb, and I would use ISC again for further applications. Thanks again. GySgt Ross "H-Bomb" Hrynewych
Publisher's Note: The photo to the right shows the completed system. Components supplied by ISC include the PTRK cabinet, rackmount powerstrip, blank panels, UPS, utility drawers, 10-32 rack screws, and cable management (in the future).
|
 |
|
|
Temperature's Rising
. . . Or Not?
January 11, 2008 • Vol.30 Issue 2
Spurred by rising energy costs, many enterprises have
considered raising the baseline temperature, or set point, inside their
data centers. While there’s no denying that thermostat settings can
impact equipment performance, deciding just what a set point should be
is hardly a straightforward issue.
“The temperature setting is important to IT equipment performance, which
is why manufacturers have temperature ranges specified for their
equipment,” says consultant Don Beaty, president of DLB Associates and
former chair of the technical committee on mission-critical facilities,
technology spaces, and electronic equipment of ASHRAE. “The temperature
setting also impacts the performance of the cooling plant. The optimum
ranges do not quite mesh, but over a period of several years, IT
manufacturers, facilities engineers, and operations personnel worked
together to develop a set of thermal guidelines that would allow for
both acceptable IT equipment performance and reasonably low energy
consumption of the cooling plant.”
The culmination of this effort, he notes, is an ASHRAE publication
titled "Thermal Guidelines for Data Processing Environments,” published
in 2004.
ASHRAE Guidelines
Beaty states that based on these guidelines, data centers should be able
to operate “in complete safety” at a range of 68 to 77 degrees
Fahrenheit. He believes most data center set points do not reach the
latter number in part because it is difficult to achieve uniform
temperature distribution throughout the facility. Measuring a range of
10 degrees or more between the inlet of one server and another—for
example, 68 F at the bottom of a rack and 78 F at the top of a rack—is
not unusual, Beaty notes. “If facilities could solve the uniformity
problem, they would likely be able to work at the higher distribution
temperatures and achieve significant energy savings,” he explains.
ASHRAE guidelines also recommend a maximum hourly rate of change for
data center temperatures, pegging this point at no more than 9 degrees
per hour. “High rates of change, or constant temperature cycling, can
result in increased equipment failure rates,” Beaty says. Tape drives,
in particular, are susceptible to large changes in temperature or
relative humidity, he notes.
However, others argue that the higher the set point, the shorter the
window of opportunity to address cooling failures. A study by Opengate
Data Systems found that a typical data center running at 5kW per server
cabinet will experience a thermal shutdown within three minutes during a
power outage. Higher density cabinets with 10kW will shut down in less
than a minute. “Thermal runaways can wreak havoc on a data center,
causing instant data loss,” notes a spokesperson for Active Power, a
manufacturer of UPS flywheel systems.
Meanwhile, Mark Monroe, director of sustainable computing at Sun
Microsystems says companies can achieve an energy cost savings of 4% for
every degree of upward change in a data center’s set point. In a
recently conducted study of 14 of Sun’s data centers, Monroe discovered
that eight facilities had the temperature set at 68 F, five at 72 F, and
one at 74 F.
“If you’re running at 68 degrees, you’re running at the bottom level of
most of those ranges,” he notes. “There’s no reason why you can’t move
to 78. This is a really simple thing to do.”
Beaty agrees with Monroe, but not entirely. He claims raising a data
center’s set point can have a significant effect on energy usage, “but
there is no set percentage reduction in energy cost that can be stated
based on, say, a 1 degree Fahrenheit rise in temperature.” What can be
said, he reports, “is that chiller efficiency will increase, and
assuming that economizers have been installed, there will be a
significant increase in the number of hours per year that the
economizers can be used. This increase is specific to the type of
economizer (water-side, air-side, or evaporative) and the ambient
temperature distribution of each site.
Several companies have introduced products aimed at safely raising set
points. Examples of such products include Dynamic Smart Cooling from HP
and AdaptivCool from Degree Controls.
Humidity Levels Warrant Attention
When addressing the issue of data center thermostat settings, storage
administrators should also consider the impact of higher temperatures on
data center humidity. Excess humidity can cause condensation to form on
electronic components, leading to wasted cooling and, in turn, a higher
energy bill. Insufficient humidity can result in ESD (electrostatic
discharges), which can cause equipment to shut down and, quite possibly,
sustain damage in the process.
In data center environments, humidity is traditionally measured as
“relative.” Relative humidity is expressed as a percentage and measures
the amount of water in the air at a given temperature compared to the
maximum amount of water that air can hold. According to Beaty, an ASHRAE
technical committee that focuses on computer rooms recommends that
relative humidity be kept in the 40% to 55% range.
Some experts believe keeping humidity at the low end of the range cited
by Beaty won’t necessarily eliminate ESDs. “You can be fine in your
moisture range and still have an ESD event,” says Coy Stine, a
simulation engineer at Degree Controls.
Beaty advises storage administrators to carefully evaluate the pros and
cons of keeping data center humidity levels low. “The lower limit . . .
should probably be set based on a total cost of ownership analysis,” he
says. He believes the best course of action entails weighing the
operating cost of humidification against the cost of increased equipment
failure due to a lower relative humidity limit.
by Julie Ritzer Ross
|
Don't put your
hardware in danger! |
A Kevinism is a funny or intriguing statement or idea from our Vice President of Sales, Kevin Hunt. Kevin is a big fan of Sandra Bullock, Pizza Inn black olive pizza, The Washington Redskins, and Dr. Pepper (not necessarily in that order). He is not a big fan of Chinese food, seafood, or other types of "dead" stuff.
Last month we talked about Kevin’s theory for why the Redskins can’t win. One of the undercurrents in that theory is that Jerry Jones (the owner of the Cowboys) will coerce and threaten his players if they don’t perform. Kevin says the Redskins would never do something like this, as they are just being altruistic, and their motives are pure.
When questioned further, the Jerry Jones angle turned into its own theory. To sum it up, Jerry Jones is responsible for everything that happens in the NFL, such as:
- Jerry bribes all the officials to ensure the Cowboys win.
- Jerry has bought off all the other owners, so he can determine who wins and loses in every game, every week.
- Jerry has bought off all the analysts and commentators on TV, radio, and the press.
- Jerry lets some teams win, and even allows the Cowboys to lose (2007 record of 13-3, although they lost in the first round of the playoffs), so as not to make his total control too obvious.
- Jerry is not above any devious and/or criminal action that helps him totally control the league, and the outcome of every game.
Of course the only team that is not under Jerry’s control is the Washington Redskins. They are too “good” to be part of Jerry’s plans. They are the only team to resist being subverted, and therefore can still follow their grand strategy.
To me, the grand strategy (beating only “quality” teams, while losing to anyone else so as to boost their self-confidence), cannot be executed together with the second theory (Jerry Jones runs everything), but for some reason Kevin thinks it can.
If Jerry is able to select who wins and who loses in any given game, then it stands to reason he can affect the Redskins, whether they want to be affected or not. So if the Redskins lose, it is because Jerry wanted them to lose, not because they are executing the grand strategy of building up the other team for a future fall. Not so says Kevin. The Redskins will win if they want, or they will lose if they want.
I doubt if there is any truth to this at all. If there were, we would be constantly presented with the spectacle of two teams trying to lose. The Redskins’ motive would be altruism, and the opponents’ motive would be because Jerry says so.
Wait a minute! Didn’t Washington blow a lead to the Eagles in the last minute of their game? Didn’t they throw an interception with 30 seconds left to ensure they lost to Tampa Bay? Didn't they call two consecutive time-outs and incur a 15 yard penalty that helped Buffalo kick a last second field goal to win? Didn't the Redskins beat the Cowboys in the last game of the season? Is it possible Kevin could be right? They made it to the playoffs, but are the Redskins playing in the Super Bowl this month?
Nope. If their opponent in each case really wanted to lose, they would not have made the plays they did. Interception in the end zone? How about a dropped interception? 50 yard screen pass to Michael Westbrook? How about stepping out of bounds, or allowing himself to be tripped up? What about missing the field goal? If Jerry were really in control and he wanted the Redskins to win, then the Redskins would win. An even though the last game of the season was meaningless for the Cowboy's playoff picture, it was crucial for getting the Redskins into the playoffs.
I think Kevin needs a better theory. Maybe recognizing the truth would be an innovative one. The Redskins are just like any other NFL team. They want to win every game. They made it to the playoffs, but could have had a better seeding if they had won more games. Look at the Giants, who were the other wild-card team. The Giants had a better seeding than the Redskins, and combined with some excellent play on the road, won three straight playoff games to get into the Super Bowl. Neither the "Jerry Jones controls everything" nor the "altruistic" Redskins theories have any merit or any effect on why the Redskins lose. Kevin just needs to accept this.
|
SIGN UP TO RECEIVE The ISC Newsletter |
||
 |
Subscribe! | |
 |
Subscribe a Friend! | |
Q: What is a Coryphée?
All correct answers will be placed into a pool for a random drawing at the end of the month. The winner will receive two free LBP-4A rackmount cable lacing bars, plus free ground shipping. Send your answers to: Jack Burlin
See next month's newsletter for the winner and the correct answer.
Answer from January's Newsletter.
Q: How many planets in the Solar System have a ring system similar to the more familiar one on Saturn?
A: Although Saturn has the most impressive rings, satellites and telescopes like the Hubble space telescope have shown us that Jupiter, Uranus, and Neptune have ring systems as well.
The winner was John Bojorquez. Congratulations!
|
|
|
On another of the infamous T-38 (see photo below) solo out-and-back missions, a flight of ten student pilots were enroute from Willie (near Phoenix) to Holloman Air Force Base in Alamogordo, NM. This would have been a fairly uneventful mission, were it not for the fact there was a foreign student involved. If you have read my previous story about the Indonesian student trying to get to Vandenberg Air Force Base (see last month's newsletter), you can skip the next paragraph.
Foreign students were officers from friendly nations who came to the US to learn how to fly. After completing pilot training they would normally go back to their country of origin to be trained in the particular aircraft in their nations’ inventory. Sometimes there was advanced training in the US, but mostly they just went through “undergraduate” pilot training, meaning T-37 and T-38. At the time I was serving as a T-38 instructor, we trained Danes, Norwegians, Indonesians, Saudi Arabians, Iranians, Nigerians, and many others. The US trained so many Germans they designated a single Air Force Base as the home of that operation.
On this particular mission, there was a Nigerian student named Lieutenant Matt Osume. Matt was a decent pilot and did pretty well in pilot training. However, he was not that familiar with the day-to-day operations on an Air Force Base, or perhaps he was just not very observant.
There are a lot of different vehicles operating around a base. At our home base of Willie, they had a tram that you got on to ride out to your airplane. There were more than 80 aircraft on the flight line, and those closest to our squadron building were right outside the door. However, the other end of the flight line was about ½ a mile away, and to ensure everyone got to their airplanes as expeditiously as possible, we had a tram. This tram was pulled by a little tractor like vehicle, and usually had two cars. If you have ever been to Disneyland and ridden the tram from the parking lot (many years ago), or the parking garage (most recently) the trams in both cases are very similar.
Another vehicle you frequently see on an Air Force Base is the “follow me” truck. This is usually a pick up truck, and it is so named because there is a big sign in the bed of the truck with flashing lights around it that reads “follow me.” This truck is operated by the transient support people, who are the ones that work with visiting aircraft. When a visiting aircraft arrives at the base, it is usually met by the follow me truck at the end of the runway. The aircraft is inspected by the transient ground crew, the landing gear pins are installed, and then the aircraft follows the truck to the Base Operations (base ops) building where it is parked. The crew then shuts down the engines, and goes into base ops. If they are not staying, the aircraft is refueled while they are inside base ops getting an update on the weather and planning the next leg of their mission.
Another vehicle is the sweeper truck (see photo below). This vehicle looks just like the one you might see once in a while in your neighborhood. It has the rotating brushes and the big vacuum cleaner. Its purpose on an Air Force Base is to constantly patrol the taxiways and runways (when they are not active, or in periods where no aircraft arrivals or departures are anticipated). They are the primary weapon against FOD (Foreign Object Damage). Since we are talking about foreign students, I need to clarify a little bit. FOD is not property that belongs to foreign students. In this case the F in FOD applies to objects that are foreign to jet engines. Things like rocks, pencils, paper, nuts/bolts/washers/screws, or any other small object that could be ingested by the aircraft engines as the aircraft moves over or adjacent to them. FOD damage to aircraft engines causes millions of dollars in losses each year. So bases work very hard to ensure that there is no FOD where aircraft are present. To supplement the sweepers, there are often “parties” of pilots and ground crew walking around the ramp to pick up FOD.
Now you would not imagine that anyone could confuse a follow me truck and a sweeper. The truck has the big sign, and of course it is really a truck, which does not look anything like the sweeper vehicle. However this is exactly what happened.
At Holloman, all the T-38s were landing at about five minute intervals. Since there are usually only two or three follow me trucks at any Air Force Base, this quickly maxed out the trucks. As soon as they brought one aircraft in to base ops, they had to drive out to retrieve another one. Sometimes it took a while for the follow me truck to reach the aircraft. The number of planes arriving was exacerbated by the layout of the runways at Holloman.
Holloman has three runways, none shorter that 10,000 feet, and the two crossing runways (crossing at a 90 degree angle (runway 07/25 and runway 16/34)) are both longer than 12,000 feet. The third runway (04/22) crosses the X made by the other two at the 07-16 points, so the layout almost looks like the roman numeral ten with only the top crossbar. The taxiway system to support this layout is enormous. Taxiing around Holloman can make you feel like you are in the middle of nowhere, especially when out beyond runway 04/22 (about 4 or 5 driving miles from base ops).
Lt. Osume landed safely and taxied off the runway. Not being in site of any buildings at the base, he started down the taxiway looking for the follow me truck.
At this time of day, near sunset, the only other vehicle in sight was a sweeper, so Lt. Osume, evidently unable to distinguish it from the standard blue Air Force pick up truck, started following it. Since the sweeper was pursuing his regular route, and not trying to go back to base ops, Lt. Osume was led on a merry chase around the taxiways at Holloman. Evidently the sweeper did not know he was being followed. This would be bad enough except that another T-38 was taxiing in, headed in the opposite direction (head-on) with the sweeper.
Once the sweeper realized he was in the way of an aircraft, he was going to turn around and get out of the way. Once he turned around he found another T-38 behind him. At this point, the sweeper turned off his vacuum and brushes and moved off the paved taxiway into the desert. He had effectively gotten out of the way, but there were still two T-38s pointed at each other, with no way to turn around on the narrow taxiway.
It was fortunate that the other T-38 was being flown by a more experienced student. As I recall it was Capt. Bass. He and Lt. Osume were on the same ground control radio frequency, so they started talking to each other. Capt. Bass told Lt. Osume to taxi slowly and carefully over to the right edge of the taxiway, while he provided feedback on where Lt. Osume’s right wheel was (since you can’t see your own wheels from the cockpit). Lt. Osume got over as far as he could. Then Capt. Bass slowly and carefully moved over as far as he could to the right, with Lt. Osume providing feedback on the position of his wheel relative to the edge of the taxiway. Once this was accomplished, it looked like they had enough room for the aircraft to pass (the T-38 has a wingspan of about 25 feet). So with Lt. Osume giving feedback, Capt. Bass taxied by him with a couple feet of wingtip clearance.
While all of this was going on, the other aircraft had all made it to base ops, and the follow me trucks could now go out and pick up the stragglers. Each aircraft was able to make it in to base ops with no further problems.
I think the person who suffered the most in this whole episode was the poor sweeper driver. I doubt if any other aircraft had ever started following him around the base, or had him penned in on a taxiway. It must have been a frightening experience for him.
Thanks for reading, and I hope you enjoyed this
month's newsletter.
Please direct your comments to
Jack Burlin.
 |
|
© 1998-2006 All Rights Reserved - But feel free to forward
this or email it to all of your friends.
For reprint permission,
please call 800-458-6255
Copyright ISCDFW.COM, 1998-2008
Information Support Concepts, Inc.
Mansfield, Texas
