Thursday, February 04, 2010

Why are so many still using guesswork to determine their needs for power?

It is 2010 & so many data center & IT managers are still relying on manual derated name plate calculations to manage the power required throughout their power chain even though many of these data centers are on the verge of running out of power & many have experienced outages due to tripped circuits. So many data center & IT managers come to us looking for real-time monitoring of power, many solutions are evaluated, but few ever get implemented. I'm trying to figure out why many are not investing in real-time power management.

If you read the Green Grid's white paper "Proper Sizing of IT Power and Cooling Load" it discusses the fluctuations in IT power draw due to inlet temperature changes, server component changes, virtualization, etc. http://www.thegreengrid.org/en/Global/Content/white-papers/Proper-Sizing-of-IT-Power-and-Cooling-Loads

I don't think we can underestimate the potential danger in using derated nameplate information to calculate power requirements. Unvirtualized servers typically use 15% of the processing power, virtualized we see #'s in the 60-95% range of processing utilization, this directly correlates to #'s closer to nameplate values as the Green Grid pointed out in the white paper. Most IT organizations are rapidly adapting virtualization technology to consolidate and operate more efficiently at the same time, which is a good thing, but it is putting rapid pressure on previously underutilized power infrastructures in data centers.

With so many variables to account for how can one depend on derated calculation tools? With so many real-time tools available to measure & trend power accurately including; branch circuit monitoring, outlet level monitored power strips, in-line power meters, IPMI and extensive software options why are so many still trying to use derated calculations to guesstimate the power they'll need for higher density virtualized deployments? This guesswork leads to potential circuit breaker trips & designed inefficiencies throughout the entire power chain. I am amazed with rising power costs, less power capacity available and so many looking to operate a more efficient "greener" data center footprint that so few are investing in real-time power monitoring tools that will allow them to plan & manage capacity effectively.

Monday, January 18, 2010

What is your definition of a "Green" Data Center solution?

Is your organization looking for "Green" Data Center Solutions or are you looking to incorporate "Green" into your Data Center Design in 2010 or in this decade? Below are some thoughts on this important issue in regards to building "Green" Data Centers. We're interested in hearing your opinions & ideas as well.

For the most part “Green” Solutions for the Data Center, is in my opinion a bit of an oxymoron, because most supposed “Green” solutions still have a carbon footprint & typically use power generated by fossil fuels in the Data Center industry. We also find that rarely are Data Center owners & operators willing to reduce availability to improve the efficiency of their Data Center. That being said, our design philosophy is to design “Greener” Data Center Infrastructure technologies where possible into any proposed new builds, renovations and upgrades for Data Center facilities.

In our opinion the 1st step towards “Greener” Data Centers is collecting accurate measurements & trending the environmentals in your Data Center facilities so we can model proposed changes & fine tune the efficiency. PTS has been running several monitoring & management tools in our own Data Center facility as well as in our clients Data Centers for several years. We use this base knowledge as well as industry best practices & PTS’ proven trade secrets during an engagement for design to propose “Greener” solutions where applicable & in line with the rest of the key design criteria for a project. In our experience many “Green” solutions such as solar & hydro power rarely can make a impact to a Data Center Design, however using water or air side economizers to take advantage of the free cooling days available in an applicable climate can provide a reasonable ROI while “Greening” the Data Center.

In addition by eliminating air mixing in Data Centers, we reduce the power consumed by the HVAC systems supporting a Data Center, and we prove these savings in cooling through CFD modeling before making investments. ASHRAE has widened the temperature range in the new TC9.9 recommendations for Data Center operations, but before we embrace this “Greener” standard and go maximizing the set points for supply & return air; we first must make sure that air mixing has been eliminated as much as possible because as we raise set points, “hot spot” issues & inefficiencies will be amplified, second raising set points reduces the availability of the Data Center so we have to make sure that any proposed increases in set point to make a Data Center “Greener” are in line with the availability requirements established for the Key Design Criteria of a project, third many servers fans will spin faster as the intake temperature rise so there is an inflection point where raising set points further will not continue to lower power consumption.

Focusing on effective Data Center capacity management is key to any "Greening" initiative:
• Better predictability of space, power, and cooling capacity and redundancy limits means more time to plan on ways to mitigate their affect
• Increased real-time availability of IT operations as a result of an enhanced understanding of the present state of the power and cooling infrastructure and environment
• Reduced operating cost from energy usage effectiveness and efficiency as well as operator effectiveness from the use of automated tool sets

In our experience most facility oriented "Greener" solutions provide only a fraction of the efficiency gains found in IT focused solutions such as: server consolidation, virtualization & data deduplication. That's not saying we shouldn't consider the facility oriented "Greener" solutions, especially if they fall in line with our design criteria & ROI needs, but it is saying we should focus on the IT side 1st because of the greater savings & our capacity requirements will be appropriately defined if we become IT efficient first.

Wednesday, December 16, 2009

The Devils in the Details - Enhanced SAN & Switching Solutions for Next Gen Data Centers

PTS is pleased to announce a new educational event, The Devils in the Details - Enhanced SAN & Switching Solutions for Next Generation Data Centers, in which we will introduce several new technology-based solutions, which will enhance data center optimization, consolidation, virtualization, and disaster recovery.

Prior to an upcoming New Jersey Devils versus Philadelphia Flyers hockey game we will leverage our understanding of the synergies between Facility and IT Infrastructure and introduce two highly efficient and cost effective solutions. These solutions can dramatically reduce the costs and complexity of your IT environment while increasing your ability to adapt, manage, and grow your storage and computing infrastructure. Learn about flexible, scalable solutions that will meet your business and security challenges and better understand how IT requirements drive new needs for your data center support infrastructure.

February 10, 2010
NJ Devils vs. Philadelphia Flyers
Prudential Center, Newark
Presentation with dinner/drinks starting at 5:00 PM
Game time 7:00 PM

Join us for an informative discussion and learn about:
  • PTS Data Center Solutions' strategic data center design approach combining both IT and support infrastructure expertise to design, manage and operate a superior data center.
  • Dell EqualLogic PS Series SANs designed to cost-effectively integrate advanced data and disaster protection features directly with VMware virtual infrastructure to help provide seamless data protection and disaster recovery management.
  • Enterasys S-Series® enterprise switching and routing solutions specifically designed for high speed core and SAN deployments.


Please RSVP by 1/5/2010. Tickets are limited and available on a first-come basis.

Data Center World, Spring 2010


PTS Data Center Solutions will be presenting and exhibiting at this spring’s Data Center World Event, held in Nashville from March 7-11. Data Center World is the largest global event of its kind and has been named one of the 50 fastest growing tradeshows in the U.S. It is the leading educational conference for data center professionals.

Our team will host roundtable discussion on Information Technology Infrastructure Library (ITIL) & ITSM Metrics Programs for the data center. This presentation will take a nuts and bolts approach to setting up an ITSM metrics program and will discuss how this process will allow IT to present data to senior management.

We’re also hosting a product information session, titled “Data Center Maintenance Management Software - Computerized Maintenance Management for the Data Center”, during which we’ll demonstrate how you can use best-in-class solutions to more effectively manage support infrastructure. The presentation will discuss Computerized Maintenance Management Systems (CMMS) and present our new Data Center Maintenance Management Software (DCMMS) Solution. This innovative software application from PTS Data Center Solutions allows the user to manage assets and parts, estimate and manage maintenance costs, track recurring problems to pinpoint those that may lead to more critical issues, and generate work orders with the details needed to properly perform preventative maintenance.

In addition, I’d like to invite you to visit us at booth #739 where you can get a first-hand look at our specially designed DCMMS solution. To learn more, please contact Amy Yencer at AYencer@PTSdcs.com (201-337-3833 x128).

To register for the event, please visit http://www.datacenterworld.com/. See you in Nashville!

Monday, December 14, 2009

PTS Announces a Strategic Distribution Relationship with Dell Corporation

I’m excited to announce that PTS has launched a strategic distribution relationship with Dell Corporation which includes the full breadth of Dell products targeted for the small to mid-size business segment.

As a leading data center design and turnkey solutions provider, we’ve been approached by many clients asking us to help them reduce overall data center operational costs through power efficiency analysis and improvements. The relationship with Dell allows us to provide consultative support by focusing upon key technology energy drains in the data center, namely routing, server processing, storage and security-based infrastructure products.

By partnering with Dell, we see ourselves as partnering with a best-of-breed solutions provider for our mid-market clients. Depending upon client applications, a host of solutions such as the Dell EqualLogic iSCSI storage family and PowerEdge blade and rack servers can improve power efficiencies, support growth within the data center and provide superior price / performance returns.

To learn more, please contact us today.

Tuesday, December 08, 2009

‘Lights Out’ Data Center Management

In a recent post at The Data Center Journal, titled “Save some money – work with outsiders,” Rakesh Dogra discusses the new trend to minimize power bills using Lights-Out data center and remote management. [As a side note, way back in 2006 we blogged about how “dim” data center designs are a realistic goal for most companies. You can read that post here.]

Dogra explains that the use of these tactics can lead to major cost savings. He suggests that, looking at your IT, security and facilities staff, it is unwise to cut back on security personnel but it may be prudent to use remote management to replace portions of the IT staff. Additional benefits may include:
  • A lesser possibility of accidents and security breaches since fewer people will have physical access to a computer room.
  • Response time is boosted with remote bios level access to a data center’s servers.
  • Geographical independence can also be achieved through this system.

A potential downside of this system is that a “data center will need people within its premises too to fire fight something going wrong like outages. Also, a data center manager may not find someone with the required amount of experience and expertise to fend off crisis when it happens.”

It is surely a best practice to consider operating as ‘lights out’ a data center as possible, as the author suggests.

For PTS, the real secret to realizing operational costs savings from reduced energy consumption has less to do with facility based solutions than it does with IT. Our position is that there is far more operational cost savings potential coming from virtualizing servers and storage.

To prove the point, in 2010, PTS will perform a network re-design effort of our own operations and provide detailed documentation and analysis of the before and after conditions of our data center energy usage. So, stay tuned...

Friday, November 20, 2009

Data Center Education Series Expands to More Dates, Cities

I'm pleased to announce the expansion of our Data Center Education Series to include more dates and cities.

If you're not already familiar with the program, our Data Center Education Series provides students with comprehensive, vendor-neutral, module based training led by the data center design experts from PTS. The training series discusses the most pertinent topics in the data center industry, tying in case studies and real world situations to provide the knowledge IT professionals need to understand, operate, manage, and improve their data centers – ultimately reducing operating costs and improving service delivery to users.

For instance, the Data Center Infrastructure Management course will show attendees:
  • Power and cooling infrastructure in the data center and how hardware and configuration impact energy efficiency and availability
  • Methods to improve data center energy efficiency
  • Management tools available to help you optimize data center performance and availability
  • Practical steps to implement ITIL
  • How to measure the IT Service Management metrics that really matter
  • How to monitor your data center to optimize performance and availability
  • What impacts data center availability and how you can improve it
The course schedule for the first half of 2010 is as follows:
  • Jan 17 - 19, 2010 in San Francisco, CA
  • Jan 25 - 27, 2010 in Washington, DC
  • Feb 8 - 10, 2010 in Chicago (Schaumburg), IL
  • Feb 22 - 24, 2010 in Dallas, TX
  • Mar 15 - 17, 2010 in Ottawa, ON
  • Mar 22 - 24, 2010 in San Jose, CA
  • Apr 19 - 21, 2010 in Washington, DC
  • Apr 26 - 28, 2010 in New York, NY
  • May 3 - 5, 2010 in Chicago (Schaumburg), IL
  • May 10 - 12, 2010 in Atlanta, GA
  • May 17 - 19, 2010 in Dallas, TX
To learn more about the Data Center Infrastructure Management course and to register, visit http://www.data-center-education.com/DataCenterInfrastructureManagement.asp.

Related courses, taught by experts in each field, are also available and include:
  • How to Get Started with ITIL (Information Technology Infrastructure Library)
  • ITIL Service Capability: Planning, Protection and Optimization
  • ITIL Service Capability: Service Offerings and Agreements
  • ITIL Service Catalog
  • ITIL Service Lifecycle: Service Strategy
  • ITIL v3 Foundation
  • Understanding Networking Fundamentals
  • TCP/IP Networking
  • Telecommunications Fundamentals
  • Voice over IP Foundations
For more information regarding each of the courses including costs and the dates and cities where they are available, visit our Data Center Education website.

Friday, November 06, 2009

PTS Data Center Solutions Showcase

PTS' growth of solutions to design, build and manage the data center has never been stronger.This post showcases two industry-leading solutions that you may want to consider for your own data center.

Energy Monitoring Systems

Would device-level power consumption monitoring help you manage costs for effectively?

Working in conjunction with Packet Power, PTS is pleased to announce a cost effective per-device energy monitoring system which is easy to deploy and highly accurate. The system provides device level monitoring & trending without having to change out power supplies or PDU's.

Features include:
  • All billing-quality power monitoring hardware is built into standard equipment power supply cables.
  • All standard cable connector con-figurations (C13/C14, C19/C20 etc.) as well as voltages and current loads are available. PP Monitoring Nodes
  • Data collection network automatically supports thousands of devices in a single facility, is configuration-free, entirely wireless, secure and operates independently of any Wi-Fi or other networking infrastructure.
  • All information gathered by our system and all advanced monitoring, billing and management functions are accessible via the web & e-mail.
  • All available without any additional hardware or software.
  • All information generated by the system can be integrated with your existing operations management and billing systems.
Learn More...

Air Curtains - A Green Alternative

Reduce data center cooling costs by directing cold air where it is needed most - through the computer racks! Air Curtains and strip doors separate cold air and warm air aisles, maximizing the dynamics of air flow to cool your data center. A system can pay for itself in months!

  • Save energy on both air conditioning and fan systems - 15% and 67% respectively (according to a study by the Lawrence Berkeley National Laboratory)
  • The Air Curtain product line includes transparent curtains, strip doors, panels and patented hardware; create a solution specific to your needs.
  • Specially formulated vinyls are low-outgassing and anti-static while meeting ASTM and NFPA fire retardancy requirements.
  • Hardware is also designed so curtains fall away in the case of fire, allowing fire sprinklers full operating range.
Learn More...

And remember, PTS typically designs these solutions and performs Computational Fluid Dynamics (CFD) modeling prior to deployment to guarantee the results, savings & performance. To learn more, please contact us today.

Wednesday, November 04, 2009

Introducing PTS' Information Technology Solutions Group

For years our team has provided exceptional service to analyze, survey, design, plan, commission and manage Data Centers for our clients. We are now pleased to leverage our expertise in All Things Data Center to launch an exciting new division, PTS Information Technology Solutions Group (ITSG).

ITSG provides information technology based consulting, design, implementation/integration, and ongoing support services as well as IT infrastructure solutions to companies nationwide. ITSG extends PTS' data center expertise beyond facility planning, design, engineering, construction, and maintenance to include service and solutions pertaining to:
  • LAN/WAN Networking
  • Information/Network Security
  • Servers & Systems
  • Virtualization Technologies
  • Enterprise Storage
  • Unified Communications
  • Software
  • Application Development

ITSG's services and solutions are tailored specifically to the needs of our client's project needs, including:
  • Technology Roadmaps
  • Data Center Relocation
  • Consolidation
  • Technology Refresh

ITSG follows our proven project delivery process:



PTS' goal is to provide our clients with 100% turnkey, people, process, and technology solutions from data center facility to IT operations.

ITSG will be led by Rich Horowitz, an industry veteran, who has been involved in all facets of the technology industry for more than 20 years. Rich is actively involved in business development, operations, Channel Partner development, Mergers & Acquisitions, and services delivery. Rich has been involved in approx $700 million in technology hardware sales, software sales and technical services engagements. Rich will be responsible for establishing and strengthening the PTS IT Solutions Group brand, and working with our clients to understand their needs and how we can provide value to them.

To learn more, please contact us today.

Monday, November 02, 2009

Intel's Active Management Technology {AMT} provides KVM access & console access eliminating the need for external KVMs or console servers?

There has been a lot of talk in the industry on how Intels new onboard AMT could replace service processors, such as; ILO, DRAC, RSA & ILOM.

http://sof​tware.inte​l.com/en-u​s/blogs/20​09/10/18/i​ntel-kvm-t​his-is-you​r-sol-on-s​teroids/

According to the Blog the local user has to allow the remote user in so I’m not sure this is a valid KVM or ILO replacement as much as a replacement for desktop tools like PCAnywhere & GoToMyPC?

If Intel does have a strategy to lead server remote access & control with AMT, I don't believe it will work. 1st of all, I'd like to point out AMT is an Intel product so it isn't an open standard for a management console. What about those who are buying AMD Opteron processors and/or Sun UltraSparc?

Secondly an open standard for server managment is already well underway from 1998 with IPMI & I think we need to look at what has transpired with IPMI to see what if any support will be given to AMT at the server level. IPMI was originally proposed in 1998 and driven by market leaders Intel, Dell, HP and NEC. Since then IPMI has been adopted by more than 150 other companies, including IBM, Sun, and every major server platform vendor. IPMI is now on its third major release. A significant percentage of rackoptimized servers and most blade computing platforms now include some form of built-in server processor technology that can work with IPMI. Obviously, IPMI data from across the enterprise can only be useful if management teams can view it from a common console. Otherwise, it would offer no advantages over a fragmented, vendor-specific management architecture. Thus, to take full advantage of IPMI, management teams need a solution that 1) delivers aggregated IPMI data to a single application, and 2) supports the IPMI implementations of different vendors.

This second point is critical. While most server vendors include the IPMI protocol in their platforms, they often hide it behind proprietary software/firmware extensions and/or bundled management solutions. An effective server management solution must be able to handle these variations in IPMI implementation in order to provide a unified view into the computing environment. My point is if AMT is to be successful like IPMI the Server OEM's are going to build their own management tools around it to differentiate themselves. Then there will also be 3rd party vendors that build central management tools to centralize access to the different Server OEM's tools that leverage AMT just as there was for IPMI. However, I'm not sure I see all of this happening for AMT because it is proprietary to Intel. IPMI is already included on most systems for these system mangement & diagnostic purposes. The Server OEM's have invested heavily in tools like ILO, DRAC, RSA & ILOM to take advantage of the IPMI chipset. Unlike AMT, IPMI is independent of the CPU and thus independent of a CPU chip failure and can be run on most systems out of band on a separate NIC. Although a few years old. Here is a good whitepaper that covers the development of IPMI and what has occured with its development.

http://www.avocent.de/web/de.nsf/AttachmentsByTitle/IPMI+WP_5+Reasons+to+Cap_0406.pdf/$FILE/IPMI+WP_5+Reasons+to+Cap_0406.pdf