Folsom, Calif.-based data center monitoring company Synapsense is looking to targeting computational fluid dynamics (CFD) software with its wireless data center monitoring system. The startup is two years old and has been shipping commercial products for six months. The company currently has 15 proof of concept projects at unnamed “Fortune 50” companies, plus support from the data center energy efficiency engineers at Lawrence Berkley National Labs (LBNL), according to Ray Pfeifer, vice president of business development at Synapsense.

Synapsense’s battery powered monitors track data center environmental conditions and use low power wireless [2.4 gigahertz] to communicate that data to a server. Synapsense’s software synthesizes that information and displays it as a live image [example below], which allows data center managers to look at real time maps of their data center and view air pressure distribution, humidity and temperature.

The following is an excerpt of a Q&A with Pfeifer:

Why wireless sensors? Why not wired?
Ray Pfeifer: The majority of the data centers are not new. The IT equipment refreshes every 3-5 years. The facility is there 15-20 years. Low power wireless becomes the first practical way to get in to monitor legacy data centers unobtrusively.

We can deploy in a 10,000 sq ft data center in a day to two days. It takes you weeks or months to do that with a wired solution. Because of the flexibility of wireless, as your data center changes, racks come and go, you can very easily reconfigure the wireless to match the environment. Some data center operators are moving 10-20% of their IT equipment in and out every month.

What specifically do the sensors monitor?
Pfeifer: Our standard installation measures the temperature on inlet and discharge of the racks, the temperature at the inlet and discharge of the Computer Room Air Coniditoner (CRAC) units, humidity at the CRAC units, and sub-floor air pressure.

But what we do with all this data is the interesting thing. A typical data center supplies anywhere two to three times the air that it needs because the majority of the air is being wasted. The strategic placement of the sensors allows you to adjust your airflow methodology. When you raise the return temps on the CRAH units, they become more efficient, you can shut units off.

The LiveImaging gives you a visual map: you can see where your hot spots are, you can see where you’re over-cooling. It enables a data center operator to reconfigure the data center and understand what they need to do to reconfigure it.

How does real-time imaging stack up against CFD analysis?
Pfeifer: There is some great CFD software out there, but the expertise to build a good model is pretty significant, so the people using CFD models are generally professional services folks. A full data center assessment is a $50,000 event for a 25,000 sq ft data center. The problem is, that’s a one time snapshot. They take your readings and they leave. In 90% of those instances, the facility guy looks at it, and it goes on a shelf. A year later, they hire someone else to do another one. If you put sensors cost effectively, do the analysis, collect data real time, you can present it visually to non-eningeering staff. Look at color. If it’s red, it’s hot. This is a dynamic tool that allows continuous commissioning the data center.

What’s the next step for Synapsense?
Pfeifer: The next piece that will complete the full solution will be energy metering, for both infrastructure and IT equipment, down to branch circuit level monitoring. We will also provide real-time DCIE/PUE and overlay it on your existing data center. We’re also working with the LBNL team as they develop DC Pro, the Department of Energy’s data center assessment tool. We’re putting those assessment tools into our software solution.

For data center managers looking for the best place to locate a data center, the price of power has been at the forefront of the discussion. Mega corporations are building data centers near dams to access three-cent per kWh utility rates, but fiber diversity is a growing concern, and is probably number two or three on the site selection criteria.

The Department of Homeland Security advises companies to seek diversity (i.e. multiple vendors) amongst fiber carriers, but the carriers don’t share maps — it’s part of their competitive advantage. And a lot of times, multiple carriers are in the same trench, which won’t provide resiliency in the event of a cable cut.

Also, with more companies locating data centers further outside city limits, out onto the prairies and Pacific Northwest, end users are running into latency tradeoffs. While some companies are dealing with latency problems with WAN optimization, others might only use the remote sites for disaster recovery or other instances where latency isn’t an issue.

Where do fiber diversity and latency issues rank in your data center site selection criteria? How are you working out fiber diversity with your developers and vendors? What are the tradeoffs you see? Please leave me your feedback in the comments.

For more info on site selection, check out the first two chapters of our Data Center Construction Runbook: Chapter one is data center site selection; Chapter two covers picking a data center design team.

Beltway data center managers, listen up! Data Center Decisions is hosting a data center energy efficiency seminar at the Capital Hilton, Washington, DC.

According to Public Law 109-431, the Environmental Protection Agency has been tasked with studying data center energy efficiency, and as part of that law, governmental agencies are required to show leadership through implementation of best practices for their own facilities.

This one-day event is a free opportunity for data center pros in the public sector to get in front of this issue before it becomes a regulatory requirement. The speakers for this event include Ken Brill, Founder and Executive Director of The Uptime Institute, and Joe Clabby, President of Clabby Analytics. The speakers will address best practices for energy efficiency across both the IT and Facilities aspects of the data center. Click here for online registration.

This post was contributed by regular columnist, Robert Crawford. On the opposite end of our normal content, which provides helpful tips to smooth out your operations, this post is offered up for a bit of comedy. We encourage you to share your comments or your own practical jokes in the comments section at the end.

[Update: For those of you who missed the tongue in cheek nature of this post, it is in fact, a joke. Please don’t try this at work. — Matt Stansberry, Editor]

Work is getting to be too serious. Between the demand for 100% availability and doing more with fewer people, there is little room for those “water cooler” moments we used to enjoy. I say it’s time to revive joy in the workplace and build team spirit. This column suggests some practical jokes that will engender mirth and leave everyone in stitches.

Remap the 3270 emulator keyboard

When a colleague leaves his or her workstation unlocked our first temptation is to send a scathing e-mail to the CEO. But some situations call for something more subtle. Instead of a career-ending missive simply reprogram the victim’s 3270 keyboard mapping.

There are a lot of creative ways to do this. My favorite is to shift every key one to the right so that typing, “logon” will actually be “;phpm” on the screen. Keyboard macros also offer possibilities where a joker could map“left-shift-F1” to type, “tso delete sys1.linklib” and hit enter.

Rename ISPF profile dataset

The lowly ISPF profile dataset (ISPPROF) looks innocuous enough as a simple, 80 byte record library. However, over years of use it accumulates a user’s preferences, job card text, PF key assignments and favorite datasets. Most of us don’t realize how lost we would be without this saved information until it’s gone.

Make someone’s day by deleting his or her personal ISPPROF library. Be sure to stand near your victim’s cubicle so you can hear the hilarious exclamations of disbelief as he or she scrambles to recover. For extra points, be sure to delete any backups in Hierarchical Storage Manager (DFHSM).

DD Dummy a DBMS archive file

Each database management system (DBMS) from IMS to DB2 has an archival process that copies active log records onto tape or other media for safekeeping. Normally these archives aren’t needed except for database recoveries.

Someone with access to the archival JCL could alter the archive output dataset to DUMMY. The records are then copied off of the active logs, but are not saved. The records will soon be gone for good when, in a few hours, the DBMS comes around and overwrites the allegedly archived log.

Imagine how hard your database administrators (DBA’s) will laugh when they have to do a production database recovery only to find they have no records to do so. This also works in cases where a DBMS has to go into the archival logs to emergency restart.

Replace stand-alone dump (SAD)

SAD is the failure data capture tool of last resort designed to gather storage dumps when the system itself is so damaged it can’t recover. System-wide failures are rare nowadays but strict uptime requirements mean any full LPAR failures must be diagnosed on the first occurrence.

Messing with SAD isn’t easy as not everyone is prepared to do the type of primitive, low level programming required. However, a study of SAD’s structure and flow may allow a lesser programmer to cause mischief by replacing a CSECT or two.

There are a couple of variations on this joke. One tactic might be to replace the SAD prompts with questions such as, “What was that IPL volume again?” Another opportunity is to leave the prompts alone and execute enough code to keep the processor busy for the amount of time a SAD usually takes. But, instead of writing storage contents, fill the dump dataset with cheerful speculations about your coworkers’ personal lives. I’m sure IBM support will also appreciate your wit after a long day of looking at normal dumps.

Reinitialize database volumes

This practical joke comes with plausible deniability. After all, anyone could understand why a simple finger check or brain freeze might cause someone to reinitialize the wrong set of volumes.

What I like about this joke is the slow buildup. You can reinitialize the volumes at 08:00 in the morning. Nothing bad will happen immediately as the DBMS usually has the database datasets along with the necessary extent information. However, once the volumes are reinitialized they become available for allocation and the wide open spaces are too tempting of candidate volumes to System Managed Storage (SMS).

Accordingly, by 10:00 or so other datasets will start to pop up on the volumes. As they appear DBMS reports of missing records and I/O errors will trickle in. Soon the errors will become a flood just about the time the DBA’s realize what’s going on.

At that point you can explain your joke to the merriment of all. Better yet, you have your own topper when the DBA’s realize they can’t recover the databases because you dummied out the DBMS log archive a couple of days ago.

ABOUT THE AUTHOR: For 24 years, Robert Crawford has worked off and on as a CICS systems programmer. He is experienced in debugging and tuning applications and has written in COBOL, Assembler and C++ using VSAM, DLI and DB2.

IBM has been pushing mainframe specialty processors — the IFL, zAAP, and zIIP — as a way for mainframers to save money. But do they really?

Trevor Eddolls at the Mainframe Update blog writes about this very topic, calling it a “bit of a swings and roundabouts situation:”

An organization needs to pay for a specialty processor, and it needs to buy software that can make use of the specialty processor, and then it can start saving money – I hear the sound of Excel Pivot tables being used to present the result of what can be quite a complicated calculation. But is there any software that can make use of this hardware?

Eddolls goes on to mention DB2 as one prime workload able to be moved to the zIIP, and there are some other major software vendors out there — CA, BMC, DataDirect and Neon Enterprise Software — that are looking to offer customers the opportunity of running their software on zIIP or zAAP.

IFL, the Linux processor, has been a success for the most part, mainly because IBM has sold it — and customers have bought it — as a way to consolidate hundreds, if not thousands, of Linux instances onto a single box. That has potential savings implications in floor space, power, systems management, and staff.

Selling the zAAP for Java and zIIP for data applications hasn’t been as easy. Why? The consolidation attraction isn’t there for one, and the benefits are more subtle and indirect. The case for the zAAP and zIIP is that you can save software licensing costs by getting workloads off the central processors to the zIIP and zAAP, where workloads don’t count against the million service units (MSU) rating of the box that is used to calculate software licensing costs.

So you might be able to save in software licensing costs, but the processors themselves cost about $100,000. In talking to users, it seemed to me that the benefit of the zIIP and zAAP was more indirect. By taking workloads to those processors, you can free up room on the central processors. That’s what might matter the most.

So instead of having to buy a new mainframe, you can just buy one or two of these zAAPs and/or zIIPs. The real savings comes not from the reduced software licensing costs as much as it does from being able to buy a six-figure specialty engine instead of a new seven- or eight-figure mainframe.

Climbing on early to the “next big thing,” major vendors and a lot of start-ups are staking out their claim in the cloud computing space. The basic idea for these cloud-based applications is that users sign on and do all of their work through the platform rather than logging into their PC and working on the desktop. However, according to eXpresso CEO George Langan, there is a disconnect between the way people work and what these companies are trying to do.

“You never worked in a single application,” explained Langan. So, to bridge the gap his company is helping develop software that will do what users want to do in any of the platforms, like Salesforce.com. But beyond developing working applications, they are focused on making their product accessible by all of the different platforms for easier sharing between users.

There are around 25 platforms right now and they’re proliferating. The major vendors include

• Salesforce.com
• WebEx Connect
• Amazon
• OpSource
• Yahoo! (Zimbra)
• Google
• Microsoft

Along with start-ups such as: Coghead, Mosso, and Joynet.

An example of eXpresso’s integration works is given in the company’s briefing packet, which includes screen shots of how eXpresso works in the Salesforce.com platform. The initial goal of the company was to integrate Excel documents across platforms, but now it is expanding to other file types.

The problem, explained Langan, is that there needs to be portability between application and files between business applications. If the WebEx Connect platform is your platform of choice and you need to work with someone on a different platform or no platform at all, you can invite them through eXpresso to view the document or file. Additionally, eXpresso allows you to put permissions on the document to control change, and the shared document is stored on an eXpresso server.

I was curious as to the reason behind the concept of cloud computing, and Langan shared that essentially, “What everyone wants to do is replace the desktop. They want to take away the Microsoft dominance of the desktop and desktop applications. Salesforce.com was the pioneer, and they’re starting to gain a tremendous amount of footprint. And Google Docs is making tremendous inroads. When Yahoo bought Zimbra, they claimed it had 600-700,000 users. Amazon claims they’ll be deriving more business from computer applications than from their retail sales.” Langan sees quite a future in this type of platform computing. He even predicts that email may be soon be obsolete. When asked what he thought would replace email, he answered “instant messaging.” At his own company, he guessed that they have cut down use of email by 40-50%, simply using the instant messaging in WebEx Connect.

“You end up sharing things in spaces instead of sending them around,” said Langan. “This generation of Web 2.0 platforms allows us to replicate the workspace we had 20 years ago. I used to work at ATT in Times Square. There were 1800 people in the building I worked in, with 125 people on the floor I was on. We walked around to talk to people about projects. But now at eXpresso, we do our developing in Vietnam and Romania and 10 people scattered around the United States in five states. Instant messaging is equivalent to walking around the floor in Times Square, but now done on the desktop.”

Langan predicts that “Businesses are going to start realizing the value of the technology — and then there will start being mandates to use the technology.” He explained that the real value is in how work is changing. “I think this is as exciting as the shift from corporate systems that were mainframes and client server based to PCs.” And while the eXpresso offering is currently the only platform integrator on the market, Langan predicts that more companies will appear in the next six to nine months, as it fills a gap in the increasingly cloudy skies.

Microsoft announced the release of Windows HPC Server 2008 Beta 2 today.
Windows HPC Server 2008 is used for cluster based supercomputing based on x64 versions of Windows Server 2008. Windows HPC Server can be used for massively parallel programs (computational fluid dynamics, reservoir simulation) as well as embarrassingly parallel programs (BLAST, Monte Carlo simulations), according to Microsoft.

The Windows HPC Server is a cluster of servers that includes a single head node, and one or more compute nodes. The head node controls and mediates all access to the cluster resources and is the single point of management, deployment and job scheduling for the compute cluster. It uses an existing Active Directory infrastructure for security and account management.

This offering lets users design their applications to work with Windows HPC Server 2008 so that users can submit and monitor jobs from within familiar applications without having to learn new interfaces.

The Evaluation copy of Windows HPC Server 2008 Beta 2 is now available on Microsoft’s Beta site.

Yesterday I went to an Aperture-sponsored event in downtown Chicago that Andrew Fanara from the federal Environmental Protection Agency spoke at. Much of it was information that he has spoken about before and that we’ve reported, all of it around the data center energy efficiency issue that the EPA has gotten more involved with in the past couple years.

A major focus of the event was measurement. Leaders in the industry say that data centers must learn to measure how much power they’re consuming in order to reduce it. Then they can have before-and-after accounts of their Power Usage Effectiveness number, which is an efficiency metric dividing your total facility load by the IT load.

Your PUE number is like golf — the closer to 1, the better. At least that has always been the common wisdom. The goal, says experts, is to reduce your PUE. But sometimes an IT energy efficiency project can play games with that number.

Steve Yellen from Aperture said a virtualization project can temporarily hurt your PUE number. Take this example: You have 10 megawatts coming into a facility, and 5 of them are taken up by the IT load. You virtualize and consolidate servers, thereby reducing your server footprint, and thereby reducing your IT load. So now your IT load is only 4 megawatts even though your facility load is still 10 megawatts. So your PUE would go from 2 to 2.5.

Presumably there would be an adjustment. You would see that the IT load had decreased, and so you would adjust your facility load accordingly. According to Yellen, everything would be hunky dory again, right? Wrong. Your PUE would still take a hit. Let’s take the same example:

• Your facility load is 10 megawatts and your IT load is 5 megawatts, so your PUE is 2.
• You virtualize and consolidate so that your IT load becomes 4 megawatts, a one-megawatt reduction. Your PUE is now 2.5. Uh-oh.
• So you adjust, reducing your facility load by one megawatt to match with the IT load reduction. So now your facility load is 9 megawatts while your IT load is 4 megawatts. Your PUE is now 2.25, which is still worse than the PUE of 2 you had before you virtualized and consolidated. Still uh-oh.

In fact, the more energy you save with your virtualization/consolidation project, the worse it could be for your PUE. Say your project reduced your IT load by two megawatts instead of one. So you reduce your facility load by two megawatts as well. That means the facility load is 8 megawatts and the IT load is 3 megawatts, yielding a PUE of 2.67. Uh-oh.

Taking it a step further, any project that improves your IT load alone will yield a worse PUE. If you buy those new super-duper efficient servers, that could make your PUE worse. If you install blanking panels and move perf tiles around the right way, that will improve your PUE.

Let’s not panic here, because there is a good side to this. If I consolidate servers, I have fewer servers to cool. That presumably means that I’ll be able to reduce my facility load further because I might be able to shut down one of the cooling units. And maybe fewer servers means I can get rid of one of my uninterruptible power systems (UPS) units. In the end, it might all even out, but it may just leave you with a zero-sum game instead of an improved PUE number, which is what you think it would do.

In the end, what’s most important is reducing your overall power load, and if you can document how it all happened, all the better.

As a managed hosting company, Contegix by necessity has to contend with every major operating system. But to run its own business, Contegix is an open source shop. “We run Red Hat Enterprise Linux,” said CEO Matthew Porter, “and we use Hyperic as the core of our management and monitoring systems.”

From its St. Louis data center, Contegix uses Hyperic’s open source HQ management tool to monitor its own applications, and customers can use the system to get their own metrics, even if “they run any operating system on the planet,” Porter said.

Initially, Porter balked at the prospect of installing Hyperic because he didn’t want to put an agent on every machine. However, Porter also wanted the ability to do in-depth monitoring of applications, and not just take stock of the network. The other tools that did both application-performance and network monitoring were SNMP-based, Porter said. “There’s absolutely nothing wrong with SNMP, but our network engineers would have to become experts in it,” Porter said. “We made a financial decision to go with Hyperic and install agents because it was expensive to train our engineers.”

Some customers with servers that were three-to-four years old needed more RAM to run agents, and Contegix provided them with the memory they needed. Contegix now collects 25,000 metrics per minute via Hyperic and provides its managed services customers with the ability to set their own thresholds and parameters.

 Trends in data center power was the topic at AFCOM New England Chapter’s meeting this week, and apparently it’s a subject that resonates with members—at least judging by the nearly 100 attendees who showed up. (As the New England chapter enters its third year, President Rocko Graziano, whose real job is manager of infrastructure operations and services at L.L. Bean, said this was the largest meeting yet). Two speakers gave the audience their take on the some emerging trends they see taking shape.

Rudy Kraus, CEO of Validus DC Systems, a provider of direct current (DC) power infrastructure for data centers and telecommunications facilities, naturally sees a bright future for data centers powered by DC rather than AC-based electricity. Kraus cited a number of statistics from the likes of the Uptime Institute and McKinsey outlining just how much power data centers can save by switching to more efficient DC power. If data centers in the United States converted only 10% of their capacity to DC power, that would eliminate $1 billion in electric bills. The co2 emissions for a 10 megawatt data center with 17,500 servers would drop from 99,776,400 pounds to 59,865,840 pounds. Kraus invited members of the audience to do their own comparison by visiting an online calculator offered by Intel that analyzes facility-level efficiency of AC and DC servers.

The other speaker, Brian Ouellette, of J.S. Fleming Associates, a provider of power and cooling systems, spoke about the five power trends heading to a data center near you. The top trend, that energy efficiency is gaining importance, is pretty self-evident. The other four trends centered on ways to make data centers more efficient: New ways to scale UPS architectures into adaptive models that can adjust to changing power requirements; two-stage power distribution that reduces restrictions to cooling air flow, among other benefits; increasing use of monitoring with tools such as smart power strips (that monitor in-rack power) and branch circuit monitoring (that monitor each PDU output circuit). Ouellette also pointed out that data centers don’t have to go high-tech in order to become more efficient. When Ouellette asked the audience whether they use blanking panels in their data centers, only five people raised their hands. “Blanking panels are a great way to get the air where you need it,” he said. “Otherwise, you’ll get cross-contamination of air from your hot aisle and cold aisle.”