PUE is defined by the total data center energy consumption or power divided by the IT energy consumption or power. So, the lower the resulting variable, the higher the energy efficiency of the data center because more of the total energy is being used for its intended function – to power the IT equipment – rather than to cool the servers, provide lighting, support infrastructure, etc.

The PUE standards have been updated recently to account for the  loopholes that several companies had exploited to lower their PUE value (these abuses were reported on Forbes.com and InfoWorld.com, among other places). As a result of these considerations, new standards such as including a weighted measure of different types of source energy (not just electricity but natural gas, fuel oil, other fuels, district chilled water, district hot water, and district steam) and a requirement to take readings from energy meters over a 12-month period (rather than over one 24-hour duration and extrapolating for the entire year).

The new recommendations are definitely worth a read for those involved with data centers. There are a few important points to keep in mind about Version 1 of the new PUE standards:

1. Dedicated data centers – the new standards are limited to data centers which are located in a building where the power, cooling, lighting, and support infrastructure are devoted exclusively to them. An update to the current document will include guidelines for measuring PUE in mixed-use facilities.
2. Report PUE with subscript – there are four categories of PUE (0-3), depending on how energy consumption or power is measured. PUE₃ is rated the highest standard because the energy consumption reading is taken at the point of connection of the IT devices to the electrical system, not at less accurate locations. PUE₀, the most generic category, can not be used when source energy includes types other than electricity. PUE must always be reported with the subscript of the category used, and the eventual goal is for PUE₃ to become the industry standard of data center efficiency measurement.
3. Must subtract non-IT equipment power use – Categories 0 and 1 measure IT energy at the UPS output location. According to the white paper, “If there is non-IT equipment supported by the UPS system (e.g., CRAC, CRAH, In-row coolers, etc.) it must be metered out and subtracted from the UPS Output reading”.
4. Energy sources are weighted – See the white paper for a table of the different factors by which source energy types are weighted.
5. Heating nearby homes doesn’t count – A trend in data centers has been to use excess energy generated from cooling the server rooms to heat other buildings (for example, a data center in Zurich heats nearby homes). These efforts to improve overall energy efficiency do not lower the PUE score in any of the categories because PUE is defined as an efficiency measurement of only the dedicated data center itself. According to the report, though, “currently there are on-going industry efforts to define a metric that could be used to account for this beneficial use, but it is specifically excluded from PUE.”

According to an article on GreenBiz.com, billions of dollars are wasted a year worldwide to run and maintain servers that sit on racks unused. Hopefully increased standards for energy management in data centers, such as the PUE metric, will help us be more mindful of how energy is used (or misused) in our IT hardware hubs.

While this blog was operational, it was an emerging resource for IT people (and also those not so knowledgeable in electronics) to learn more about computers, have their questions answered, and their server problems solved.

I have taken over for the former blogmaster, Alex Pyatetsky, and will continue the good work that he has begun here. My name is Alex as well (confusing, yes).

I’ll begin by linking to a few posts in our other blog, the International Computer Concepts (ICC) company blog. The idea behind running two blogs simultaneously is that our company blog will be for general current events pertaining to servers, HPC, new technologies and the like. This blog, ServerProblemsSolved.com, will be more for introductions to new technology (like the introduction to IPMI post below) and issues related to, well, solving server problems.

So, without further adieu, I invite you to check out our introduction to InfiniBand post on the ICC blog. Also, if you are interested about how GPU (graphics-processing unit) servers are making supercomputers affordable for everybody, see that post as well.

Let’s begin.

Brief History of [Painful] Server Diagnostic Tools

For some time, motherboard diagnostics were obtained by checking the server BIOS settings or using software diagnostic tools installed on each server’s hard drive. However, diagnostic needs have become more complex over the years. First, companies (and individuals) have more servers than ever before. Checking the diagnostics for each server individually is a hassle. In addition to the sheer volume, the distance between people and their servers has increased, making remote access to diagnostic systems vitally important. As such, the world’s demand for server diagnostic systems is changing. Traditional software solutions might be sufficient for a small company with a few servers that within easy access, but sys admins in a larger corporation require the latest diagnostic technologies to manage their hordes of distant servers.

This is not to say that the old ways are useless. They are great for troubleshooting and monitoring a specific server, especially if you are standing right in front of it. But wouldn’t it be nice if all of the server diagnostic systems were routed into one terminal and given a KVM (Keyboard, Video adapter, and Mouse) interface? Imagine a world where interfaces from multiple servers could be directed into a single terminal. It sounds good, but if the software across your servers is not compatible, the odds are good that when routed into a single terminal, these tools will have to be accessed separately. That sucks.

How can IPMI help?

AOC-SIMLC(+) (top) & AOC-SIMLP-3(+) (bottom) Examples of IPMI cards, provided by SuperMicro

IPMI, short for Intelligent Platform Management Interface, provides a standard for server diagnostic systems, thus enabling one application to access diagnostic data from any IPMI-compatible motherboard.  IPMI was first standardized at the Intel Developer’s Conference in August 1999.  Version 2.0 was standardized in 2004.  The IPMI system consists of a Baseboard Management Controller (BMC), which interfaces with other controllers and sensors on the motherboard via the Intelligent Platform Management Bus/Bridge.

The BMC can either be on a separate IPMI card, or it can be directly integrated into the motherboard as a chip.  IPMI cards are most common now, but onboard IPMI is likely to surpass their popularity over the next few years.  Because it is on or attached to the motherboard, the BMC can still be accessed even if the server is off or damaged and unable to boot.  So if a server is locked up, a sys admin can check the IPMI diagnostics from a remote system, whereas traditional diagnostic software would probably be inaccessible until the server is rebooted.  Likewise, an IPMI-enabled server does not even need to be on to provide diagnostic data.  This feature is particularly useful for server inventory management and on/off monitoring.  Even if the server is off, you can bring it back up remotely.

Not bad, right?  But the benefits of IPMI do not end there!

IPMI—The Sys Admin’s Swiss Army Knife

As an industry standard, multiple IPMI-equipped servers can all be connected to a common terminal over IP, eliminating the compatibility issues born by previous diagnostic tools. This technology is most useful to data centers with large numbers and varieties of servers, but small businesses with only a few servers make good use of IPMI as well. While a large business may put fifteen server racks easily into fifteen (or less) terminals, a small business can still output a single rack into a single terminal. Bottom line: regardless of business size, you can still gather data on multiple servers from a single point, and that’s pretty convenient.

Now wiring up an IPMI system with Ethernet cables, server to terminal, is most useful for a business with a dedicated operations team that can check up on the servers from a terminal on the floor. But better yet, IPMI over IP allows any IPMI-compatible software to be interfaced with the BMC and sensors via the internet. Because it is a published standard, IPMI has numerous open-source client programs that can be used for monitoring servers over the web. A quick glance at Google or Wikipedia will unearth many such apps, meaning there is no shortage of software to help to check on your server babies while you are out on the town.

An example of an IPMI web interface tool (provided by SuperMicro)

Another feature of IPMI is the Sensor Data Records, a small table which holds information about each sensor on the motherboard.  This table functions as an inventory of the components on the motherboard being monitored.  Its primary use is internally for the BMC and software to determine what data should be getting retrieved, but it is also provides a useful component inventory.

A Sensor Data Records Screen from SuperMicro’s IPMI web interface tool

When a Man Loves a Diagnostic Tool

By now, you might be falling for IPMI.  She can display data for multiple servers via the KVM switch and terminal, give you a full inventory of monitored components, and enable you to check the state of the components regardless of what the server is doing (or not doing) at the time.  She sounds like an easy way to get all the data you want when you want it, but don’t get hitched yet.  IPMI has a few little quirks that you’ll have to overcome to make this relationship last. The most annoying aspect of IPMI is that each IPMI card (or IPMI-equipped board) requires an IP address.  This might seem like a small price to pay, but to a company trying to avoid the costs of registering more addresses in the stagnant economy, IPMI could look like another gold-digger.  With that said, the cost for this IP is the same as any other IP address, which is only a few dollars a month, so if you have the money, this is an investment I would recommend.

Also, IPMI cards can be a hassle because they are not interchangeable with other motherboards.  That is, a BMC is designed for a specific motherboard or type of motherboard.  The sensors on the board, and therefore the data being sent across the Bridge between these sensors and the BMC, will be different on different motherboards.  So while the application-interfacing side of IPMI meets a standard and allows for all kinds of wondrous implementations, the motherboard-interfacing side is very specific.  Depending on your needs, this may or may not matter.  If you change servers often and feel you will be swapping your IPMI card around often, bear this caveat in mind.  Onboard IPMI eliminates the issue of swapping cards, but the motherboard itself will likely be more expensive in that case.

Most board manufacturers are beginning to incorporate IPMI on or with their boards, so this technology is now accessible to anyone looking for it.  Intel and SuperMicro have both been pushing heavily on IPMI technology, with other board manufacturers such as Tyan following suit. Most recently, Supermicro released a line of Supermicro Superservers for Intel’s Nehelem processor that all include onboard IPMI.

Love/hate IPMI compared to other diagnostic tools?  Have other questions?  Let us know in the comments section below!

PR: wait… I: wait… L: wait… LD: wait… I: wait… wait… Rank: wait… Traffic: wait… Price: wait… C: wait…

Inside the New Platform

In the previous article, we explored just how fast the i7 is and realized that it will not be a cheap upgrade. Today, we will see what Supermicro has brought to the table with its motherboard line-up for the i7.

To this day, there is only one chipset that supports the Core i7 processor (Intel x58), which Supermicro has introduced on eight new motherboards. In addition to the x58 chipset, the new boards will include an improved SATA Raid controller (Intel ICH10R) with at least six SATA ports. All boards will also share up to 24GB of DDR3 memory support and two Intel Gigabit LAN ports. Below, we discuss many of the other model-specific features that Supermicro brings us with this line.

Supermicro i7 Compatible Motherboard Highlights

Below are some highlighted features of Supermicro’s new line of i7 compatible motherboards.

• C7X58 will be the board for the heavy gamers out there. Using two x16 PCIe 2.0 slots (also single x4 PCIe slot), it supports both SLI and Crossfire configurations. This board will also feature a Realtek 7.1 high def. audio on board for those high end surround sound systems.
• X8SAX is very similar to the previous board, except with an addition of two PCI-X 64-bit slots for many server add-in cards. This model will not support SLI or Crossfire (still has same PCIe ports as C7X58 though) and will target a more server oriented client.
• X8ST3-F is packed with many different features that are designed for a server environment. It will have an onboard LSI 8 port SAS controller, onboard Matrox VGA, and an Integrated IPMI management card with KVM support (allows access of the system through LAN anywhere in the world). As for the PCIe slots, it will have three of them: x16, 2x x8, and a x4
• X8STi-LN4 will be the only board that packs four Intel Gigabit LAN ports, (all other boards have two). It also has Matrox onboard VGA and x16 and x8 PCIe ports.

These are just four out of the eight new boards introduced. The other models (X8STE, X8STi-3F, S8STi-F, X8STi) are just simplified versions of those ones described here.

Our Outlook

There is no doubt that Intel has created a monster processor that not only dominates its predecessor, Core2 Duo, speed wise, but also introduces a new chipset platform. The introduction of triple channel DDR3 and the return of hyper-threading will allow users to break performance records in multi-tasking with servers and graphic workstations. Supermicro has sweetened the deal by integrating new features, such as IPMI and SAS controllers, into its new i7 motherboards.

In short, the i7 is packed with features and very attractive to any type of Supermicro Superserver where performance matters. The real question that we always find ourselves asking whenever a new product arrives,  “is it really worth upgrading?” Unfortunately, the i7 is not compatible with older platforms and is certainly priced higher as well. There’s no doubt, that this will be an expensive upgrade. But, all things considered, we think it’s worth it. Whether you are using the an AMD solution or Intel’s older Core2 Duo, the i7’s performance leaves them both in the dust. For those users who are satisfied with their current system’s performance, there is really no reason to upgrade, however I don’t see the prices on the i7 coming down anytime soon, so if you are looking for some serious performance gains, there is no reason to hesitate.

Are you ready to make the jump to i7 for your Supermicro machines? Have you already done it? Love it? Hate it? Let us know in the comments section.

In the world of micro processors, we consumers don’t have many options to choose from because two companies have dominated this market since 1969. In the competition between Intel and AMD, the two competing chip giants, Intel has been in the lead since 2006 when it created the Core 2 Duo platform. Today, I’d like to introduce you to Intel’s new platform, the Intel Core i7 on the X58 chipset, and discuss what exactly is different, why it’s the fastest commercial processor to date, and whether it’s worth the upgrade or not.

What’s New in i7

Core i7 (or just “i7” if you’re a true Nerd) has been greeted with excitement by the technology community as Intel’s first major innovation since Core 2 Duo, which was released over two years ago. The new processors will have 1366 pins, making them incompatible with Core 2 Duo and older platforms. Although the i7 is manufactured with the same 45nm technology as the Core 2 Duo, it has undergone many architectural changes.

According to Intel, one of the improvements is the introduction of an integrated memory controller. This will have a significant increase in the processors’ performance, especially in memory access.  Also, this will allow the use of DDR3 memory that will run up to 1066MHz in triple channel. One unexpected change in the i7 is the return of Hyper-threading technology. Hyper-threading died with the Pentium 4, but its back as in the Core i7 line. Hyper-Threading will allow each core to work as two virtual cores on the software level, which will allow for more parallelism and an improvement in multitasking performance. This new platform has already been tested by on various occasions and according to tomshardware.com, the new processors are 16 percent faster than their Core2 Duo counterparts in a same clock speed comparison and 22 percent faster than AMD’s new Phenom II processors. This is a single processor comparison. When Intel releases a two processor i7 platform, we expect the already impressive performance difference to become even more substantial.

i7, to Buy or Not to Buy?

It is evident that Intel has created an incredible product with great selling potential. For all practical purposes, the i7 is far superior to the Core2 Duo, boasting significant performance improvements. However, there is only one chipset that supports it today, Intel’s x58, making i7 compatible motherboards expensive. If you need the horsepower, it makes sense to upgrade for the additional performance, but the extra cost will leave some consumers disinterested.

The next article in this series will discuss Supermicro’s motherboard solutions for the i7. We will talk about the new features available on all the new boards and why you may not want to wait to invest in the i7.

What are your biggest questions and concerns regarding the i7? Let us know in the comments section and we’ll do our best to help you out.

This is how we pick teams at ICC. Image by Groovyguru @ Flickr

The Official Launch

I know you’ve all been waiting for a long time, asking yourselves, “when will this damn blog be up? I have soooo many server problems!” Well, I have good news for you, fellow Nerds – today is your day.

Today, we celebrate the launch of Sever Problems Solved, your new favorite publication dedicated to finding practical solutions to frustrating server problems!

Who are you guys? What are you doing here?

We’re glad you asked.

This blog is presented on behalf of the good folks at International Computer Concepts (aka. ICC). Since 1993, we’ve been building servers and supporting a wide variety of clients including hosters, government, research/universities, start ups, end ups, and everybody in between.

This blog is our opportunity to hit many birds with 1 stone:

1. Give the IT community a reliable and thorough resource for answers relating to the most frustrating and relevant hardware problems.
2. Engage our readers to uncover and resolve community wide concerns as they crop up.
3. And hopefully, by using this resource, you’ll come to like us, and will consider us for future IT hardware purchases. If not, that’s ok too. We appreciate your eyeballs.

You Mentioned an Internship?

You bet. And its paid!

In order to deliver quality content on a consistent basis, we want to build a team of trusty interns who will help synthesize the collective wisdom of our company into a stream of articles.

Are you always tinkering with computers and technology? Are you the go-to guy/girl for all of your friends’ computer problems? Do you find yourself thinking “Damn, that’s dumb,” when you see technology being misused? If so, we may need to talk.

See the complete job description and instructions on how to apply.

Before you leave!

Thanks for checking out the very first post on this blog. We’re really excited to help you solve your server problems.

Please subscribe to our RSS feed or follow us on Twitter to keep up with our articles. Also, if you click the “Tweet This” button below and help us spread the word, we would be unspeakably thankful.

Having server problems? Want to request an article? Let us know in the comments below. If not, just say “hi” so we know you’re out there.