Tag Info

Hot answers tagged

31

Before the outage: Power everything off - workstations, servers, printers, switches, the works. Turn off your UPS' so they don't panic when power is lost. After outage in this order: Turn on UPS Turn on networking (router, switches etc) Turn on servers Turn on workstations Turn on everything else Have a test plan ready so you can test important ...


22

There are 3 Main BIOS settings in the Dell R710 that control this under Power Management: OS Control sets the CPU power to OS DBPM, the fan power to Minimum Power, and the memory power to Maximum Performance. In this setting, all processor performance information is passed from the system BIOS to the operating system for control. The operating system ...


21

The console-tools package allows console options to be controlled. To turn off screen blanking and powerdown, set BLANK_TIME and POWERDOWN_TIME to 0 in /etc/console-tools/config. If you'd prefer not to modify the config file, the same effect can be achieved by creating a new file in /etc/console-tools/config.d containing the following: BLANK_TIME=0 ...


17

A couple of jobs ago, one of the datacenters for the place I was working for was one floor below a very large areal. This large, thin, metal item was the tallest thing in the area and was hit by lightning every 18 months or so. The datacenter itself was built around 1980, so I wouldn't call it the most modern thing around, but they had long experience ...


15

Or you use /etc/kbd/config to set up (depends on your system, what is installed) BLANK_TIME=0 BLANK_DPMS=off


13

It allows the raid card to remember what is in its buffers ( that hasnt been sync'd to disk ) Its very important for people who need high data integrity.. Or to save your DB from certain types of corruption.. (Basically whats on disk, is on disk - so thats safe.. The problem is when the OS thinks its on disk but its actually not and in a RAID card buffer) ...


12

For posterity, here are the settings needed for HP servers (as of the DL360G6 I just checked) The main one: Power Management Options HP Power Regulator HP Dynamic Power Savings Mode Vary CPU frequency and power setting based on load. Ignores OS settings. Default HP Static Low Performance Mode Tunes CPU frequency and power settings to their lowest ...


12

I've answered a similar question elsewhere. Constant cycling of the power on regular drives actually leads to premature drive failure. The drives are rated for only so many spinups/spindowns, and if you cycle them too frequently, the drive will start to exceed its rated capacity for this. Note that this assumes you're using desktop or server-style 3.5" ...


12

Horizontal PDUs are a mess -- In my experience there's nothing that can be done to make these neat: You can bundle your cables neatly down the side of your rack, but when you get to the PDU they fan out into a rat's nest. For vertical ("Zero-U") PDUs you can acquire custom-length power cables (they're available from various suppliers, usually in the same ...


11

Buy an Uninterruptible Power Supply - put it between your server and the idiots around you :)


11

Having just done through a datacenter shutdown in the last week, this is fresh on my mind ;). Yes, shutting everything down needs to be done. Some things can tolerate having the power yanked out from underneath them, and they typically can be identified by not having a power switch on them. Depending on what the heck the facilities people are doing, you may ...


11

This is a pretty huge question. There are alot of different ways people are saving energy and then in turn saving $$ by greening their datacenters. I think you are asking about the ROI on such efforts. Virtualization is certainly a great example - purchase some big physical servers (2 or 4 sockets) with a ton of memory (32 - 128 GB) and then convert ...


11

The brief interruption to someone's work is more likely going to cost less than buying and maintaining a desktop UPS. That's one of many reasons why many professional solutions are server based and centralized, as the service level can be focused there. If there is a business justification with Return On Investment (ROI) in your environment, I would ...


11

I'm not sure about servers, but the current thinking in embedded devices is not to bother with steps between low-power and flat-out because the extra time involved will eat your power savings, so basically they run low power until they get any real amount of cpu load at which point they flip over to fastest-possible so they can finish the job and get back ...


9

I have always turned off any type of power management on servers. I am curious to what others have experienced, but I always assumed that if the server is under-clocking, there will always be some delay to 'step up' the CPU to 100%, and in a data-center setting any delay like this is unacceptable. The data you provided seems to support this assumption. So, ...


9

Flywheels are used in large data centers to cover short power gaps like yours, or to allow time for a backup generator to come on line. They are (typically) only good for 45-60 seconds, and they are expensive, but they have a longer lifetime than UPS batteries, and can make sense for a very large data center. Another approach was Google's, who integrated a ...


9

I'd personally let the server run or move it to a better running environment. This may not be worth automating. If you must shut down the entire host, please make sure that your VMs have the VMWare tools installed (to allow graceful shutdown and power up) and that they're set to stop and start with the host. You can use the vCli and schedule a cron job to ...


9

On the server version of Windows, you can't. Once the Hyper-V role loads, hibernate and sleep are disabled. You can sleep on Windows 8 Hyper-V because it is "Client Hyper-V" in which sleep states remain enabled. (Windows IT Pro has a list of differences between Server and Client Hyper-V, and this appears among them.) If you really need Hyper-V installed ...


8

CPUs (or individual cores) can be put into "deep" C-states by an OS kernel. You're thinking that the processor "puts itself to sleep" when it "detects" inactivity. This isn't how it works. The OS scheduler determines system idle percentage based on the amount of time spent in the system idle loop to detect system inactivity. The CPU itself (in x86 land, at ...


8

I work for a large state government where we recently began implementing power management. According to our calculations based on metering a sample population shutting off PCs during idle periods saves about $35/PC per year for a mainstream business desktop with LCD monitor. Your mileage will vary, so do some testing yourself. Laptops are generally provide ...


8

The parameter is controlled via the kernel command line, using setterm merely alters the runtime settings. To disable it system wide you can alter your kernel boot command line by appending it with "consoleblank=0" in your boot configuration (grub/lilo). If something during boot setterm's it then it will override the value.


7

Your options are: UPSes laptops Google-style batteries and custom motherboards[1] Crazy flywheels[2] Basically you'll need batteries somewhere to cover even a couple of seconds of power outage, or a huge spinning slab of something heavy which you can generate energy from in a hurry. UPSes aren't that expensive, and look even cheaper when you consider ...


7

I find UPS batteries fail after 2 to 3 years of continual use. If it's an APC UPS they have a battery test function and I use this about once a month to make sure the battery is OK. JR


7

The efficiency improvements you get from moving to 110v to 220v are not that great. Yes, there is less heat, but we're talking a few percentage points. The few things I've seen, generally in commodity hardware not server hardware, show efficiencies moving from 83% to 86% efficient, or +3% efficiency. That can add up if you have a 10K sq/ft room full of ...


7

I would take it down to single user mode, unmount any non-OS filesystems and remount any existing ones as readonly. Then run your firmware update tool and 'ipmitool chassis power cycle'. It's not pretty or clever but should be safe enough


6

Buy short power cables. In the US, I buy 1-foot and 2-foot cables and run to the vertical Zero-U PDUs located on the left and right sides of the rear of the cabinet, since the power supplies are usually situation on the left or right side of the server chassis (never center). I don't bother with server cable management brackets since they restrict airflow ...


6

DL360's should handle 208v with no problems. They have auto-switching power supplies which'll handle the switch for you. Also, running at 208v is slightly more efficient than running at 120v. A few percentage points gained, which can add up for a server that routinely burns 300+ watts.


6

You can get amp meters to mesure the power. or you could take meter readings from your electric meter. THe best way to get god readings is to do this for 24 hours. Not just a few hours. (You mentioned the weekend) and that obviously is a good time. But if you are able to get 2000 servers avabile for downtime at a weekend, do you need all of those servers? ...


6

With X11 you want to disable DPMS in /etc/X11/xorg.conf. #In the "Monitor" section, you need a line like: Option "DPMS" #Then, in the "ServerLayout" section (for Xorg 7.2 and later, make a #separate ServerFlags section instead), include lines like this: Option "BlankTime" "0" Option "StandbyTime" ...


6

Your UPS vendor should have guidelines. Our APC UPSes tell us when they think the batteries need to be replaced - from memory, it's about 2 years. How heavily the UPS is loaded has a big impact on battery life. APC recommends that you never load them more than 1/2 way.



Only top voted, non community-wiki answers of a minimum length are eligible