3 Keys to Data Center Energy Efficiency

Posted on

Data Center Knowledge published a recent article recapping three guidelines from an electric utility on how to best attain energy efficiency in your data center. Let’s take a look:

  • Create a comprehensive inventory model of the data center. Not only does this allow you to see the existing capacity of your power and cooling systems, but it can also bring to light dependencies and allow you to better match resources to devices.
  • Make the power chain visible. This requires a “big picture” of the power system – dependencies, active power pathes and current status for each device.  You will also need a unified view of real-time energy consumption and current and historical PUE and DCIE stats.
  • Continuously monitor the data center and compare against historic trend information. An alarm or event management tool is also useful for effective monitoring.

The main point is to achieve energy efficiency you must understand your data center’s power chain. Can you:

  • Align power delivered to SLA performance
  • Know which rack has the required amount of power to handle new equipment
  • Identify trapped capacity along the power chain
  • Identify unnecessarily power idle equipment
  • Have nameplate, derated and actual power consumption documented
  • Receive real-time alerts when capacity thresholds are exceeded.

With summer here in Minnesota, it is a great time to think about your data center’s power consumption and how you measure and monitor.

Enterprise Data-Driven Transformation

Posted on

“Companies will spend an average of $7.4M on data-related initiatives over the next twelve months , with enterprises investing $13.8M, and small & medium businesses (SMBs) investing $1.6M. 80% of enterprises and 63% of small & medium businesses (SMBs) already have deployed or are planning to deploy big data projects in the next twelve months,” reports Louis Columbus for Forbes.  Data is dominating discussions in our organizations and our data centers are working overtime. How does an enterprise best prepare to utilize the data it has and the data to come? How can enterprises truly transform their businesses with data-driven actions and decisions?

Thor Olavsrud of CIO outlines tips to pave the way for enterprise data-driven transformation. Here are some of his insights:

  • Recognize that becoming a data-driven organization is “no simple task.” Set expectations that this will be a comprehensive and challenging task
  • Understand your business objectives and use that as a starting point to outline your roadmap – what new data sources, technology is needed to achieve
  • Prepare your data landscape for more data coming from connected people and things (IoT). Mr. Olavsrud’s article points out that silos that have been previously broken down could appear again when new data floods into your environment
  • Work to change your data culture – get people to view data not as a “by product” but as an “asset”
  •  Start working towards data- and insights-as-a-service
  • Measure your data! Make evaluating the insights you gain a KPI that is measured and reviewed for its effectiveness just like you would with other enterprise programs and initiatives

In closing, don’t forget that security and governance of your enterprise data is still very essential for success. Unreliable data or data used or accessed improperly only has the potential to hurt the enterprise as well as its goal of being more data-driven.

Time for Money (For Time)

Posted on

by Michael Downs, Solution Architect, Evolving Solutions

I read a post on Wired today where the author said the value of the the Apple Watch will be measured in time gained. Or regained. To quote the original source, “There are very few products that allow you to hand someone cash and be given back time. This will be the Apple Watch metric to track: time saved.”

Well, maybe. Isn’t that the genesis of every technology ever produced by human beings? The wheel, the printing press, the cotton gin, the steam engine, they all gave us back time.

Initially, the learning curve, building processes and procedures and integration, take time and temporarily reduce efficiency. But, we quickly adapt and become accustomed to the new technology, it becomes as integrated as what we replaced and the expected time gains occur. Then something interesting happens. The technology we’ve adapted to becomes the right answer to everything. We extend, configure, customize and tweak to make it work, and it does. But every time we stretch the original to fit the new, it becomes more fragile, more inflexible and requires more people, process and money and time to hold it together. Worst case, we lose opportunities believing it’s impossible to change. It’s the germ of what’s commonly referred to as Technological Debt.

Introduction of new technologies threatens the stability and efficiency we’ve worked so hard to gain, but extending what we know to solve what we don’t slowly destroys those gains regardless. So what’s a modern technology organization to do? The guidance I give people is to identify the technology that addresses the challenge (or opportunity) you’re facing and apply it. It sounds obvious, but most organizations are filled with technologies stretched to their limits.

Here are two examples: terabytes of unstructured data sitting on conventional storage arrays or maybe ebusiness intelligence solutions on top of a conventional SQL database. Sure, they both work, but they are stealing time and money.

There’s an evil twin to this theory, where the new technology tries to do what the old technology already solved, but that’s for the next post.

What do you think? How do you decide when to introduce a new technology into your environment? What cost or time or opportunity constraints factor into your choice? I’d love to hear your perspective in the comments.

_____________________

Michael Downs is a Solution Architect with Evolving Solutions.  He uses broad experience from infrastructure to application life cycle to business services, across a variety of industries to create solutions for clients. He is particularly focused on business outcomes and creating opportunity for transformational innovation.

Technology Spring Cleaning

Posted on

April is here and with that spring weather and plans!  It is also a time to refresh and gear up your operations for the rest of the year – perhaps a “spring cleaning” is needed.

Kristopher Spadea of CIO points out that a good “spring cleaning” is an important part of leading business transformation.  He writes, “many companies underestimate the importance of spring cleaning, failing to take a close look at what is out of order, cluttered, dusty, or dysfunctional within their own house. But if efficiency, productivity, and cost are three primary drivers for your business transformation (as they pretty much always are), then you should take a long, hard look at what you are doing and how you are doing it – and get ready to clean up the problem areas you are inevitably going to find.”

We gathered his thoughts and recommendations for you to consider:

  • Access your capabilities. List out the functions that you deliver and make an honest assessment of each.  Which functions do you deliver best, which ones are a struggle and, finally, which ones are failing.
  • Decide what stays and what goes. Mr. Spadea helpfully points out, “don’t assume that you always outsource what you currently don’t do well, or that you always keep what you’re good at. You may decide that it is worth your while to learn to do a job better yourself, rather than turn it over to a vendor; or that you should outsource a function even though you’re good at it to free up resources to better meet business goals.”

Last but not least, make encouraging a mental shift within your team part of your spring cleaning efforts. Mr. Spadea writes, “business transformation includes evolving from a tasks completed mindset to a services delivered mindset. Tasks completed has a narrow, departmental focus: “Can we check this off our list?” Services delivered has a broad, holistic perspective, “Did we help our company/vendors/clients achieve their goals?”

Take time today to get started on a little technology “spring cleaning” to gear up your company for success this year.

A Desktop in the Clouds

Posted on

by James Keating III, Business Technology Architect, Evolving Solutions

As part of my job I routinely look into and evaluate technologies.  This takes two paths, new or emerging technologies, and also changes to economics and trending that make existing technology more suitable for business use today.  One of these latter case technologies is virtual desktop infrastructure or VDI.

I can vividly remember the first time I came into contact with the concept of VDI.  At the time I was a manager of a large storage team in a large enterprise setting.  We had been approached by the Windows group and Desktop group about VDI as they wanted a way to control the sales  and offshore teams who at the time were issued company owned laptops.

The idea was simple, provide a desktop that lives on some central compute and storage and allow them to access it via whatever device they had (preferable for this case a non-company provided system).  They saw this as a way to simplify administration and a way to control the data, and  to prevent the lost or stolen laptop issues that was prevalent.  So we looked at what it would take in terms of infrastructure to support this.  Many storage folks know where this story is going, the performance required of the storage to support such an endeavor quickly made the idea seem less than appealing from a storage manager standpoint.  Basically, I would have to provide some fast and expensive disk to support a laptop workload.  This didn’t sit well, and further since we didn’t want to mix the workload with other production stuff, this created an island storage array, dedicated to VDI and only VDI sitting in the middle of my other production systems.  Add in backups and disaster recovery if required and the whole thing collapsed in a poof of infrastructure cost  uncertainty.

Fast forward 6 years and a lot has changed.  The cloud is now something that one could utilize (not just on paper, but for real) to support the same use case described above.  The storage is no longer an issue due to both software and hardware advances over the 6 years (FLASH, software VDI acceleration).  The island array is also no longer a factor if one consumes the VDI in a cloud as a service model, since the storage is part of the service bundle that comes with the VDI service.   Basically, the same use case I ran into 6 years ago that became a 9 month journey into all the issues and why VDI while nice was just not for the company, could now be solved in a simple manner using VDI as a service bundles.

So as I said/typed in my first sentence of this blog, I look at technologies as part of my day to day job, and have been looking and testing various VDI as a service options and offerings.  An inexpensive option is Amazon Web Services Workspaces.  This option provides Windows 7 desktops and access to them from most Windows, Mac, iOS and Android devices.   I have tested it, my team has tested it.  However, coming from an operations background I want field data, to know really how these things will perform under actual “in the wild” use cases.  Basically since the price point is so attractive, I want to find all the limits, the rails, the gotchas so I can work with customers to understand the real world outside of the marketing brochure.  So to get this field data I am prepared to offer up to the first 10 folks who contact me and meet some requirements, one month of the base Amazon Workspaces virtual desktop to use.  As with all things some restrictions apply, but the goal is to find out from a group of real users how a desktop in the cloud works.

If you are interested in participating, don’t hesitate to contact Evolving Solutions to see if you qualify for this and let’s get started testing VDI.

Enterprise Software Trends – 2015 and Beyond

Posted on

Joab Jackson writes for Network World, “These days, IT’s basic firmament is giving way to a more breathtaking geography that the IT pro must traverse, based on pay-as-you-go cloud computing, building applications and performing deep data analysis. Perhaps more fundamentally, IT operations are moving from merely supporting the business to driving the business itself, which requires agility and making the most of resources.”  Mr. Jackson sites the following as the most impactful changes to the platform, software and data:

  • Platform. Cloud is not new but expect it to continue to be front and center especially as concerns about security lessen and as more enterprises face the mounting costs of maintaining and upgrading internal data bases to meet new customer demands
  • Software.  The pressure to move fast and to have software that works well and provides innovation in the marketplace has placed more light on the benefits of DevOps.  Mr. Jackson reports, “60 percent of CIOs plan to use DevOps to manage their software” according to an IDC study
  • Data.  Data not so much for reporting but for making real-time business decisions at the customer level

In another article about enterprise software trends Jennifer Lonoff Schiff gathers the following predictions for 2015:

  • Hybrid cloud goes mainstream
  • Subscription pricing for enterprise software
  • More mobile enterprise apps
  • ERP value rises and with it more integration with ERP and CRM
  • Open source software continues to take hold in the enterprise
  • Business intelligence software will become more visual and easier to use
  • Social intelligence will progress further in enterprise decision making

All these changes may at times seem a bit unsettling especially when it comes to shifting away from solely a business supporter to a business driver.  As 2015 kicks off embrace the changes to come, many of these predictions will allow your company to deliver better customer experiences and in-turn increase revenue.

Power8: Improve Your Virtualization Skills

Posted on

Alex Woodie of IT Jungle recently interviewed Tim Klubertanz, Senior IBM Power Systems Architect at Evolving Solutions, about the need to improve your virtualization skills to better handle the new IBM Power8 environment.

Power8 relies on virtualization even more than Power7 systems, but Mr. Woodie writes, “On the plus side, the massive compute and I/O capacities of Power8 open up new possibilities in the analytics world. For IBM i shops, that means taking seriously the potential to run new analytics workloads right next to your transactional loads.”

In terms of virtualization skills, Tim calls out knowing how to use Virtual I/O Server (VIOS).  Power8 has more threads per core and more cores per socket and no I/O expansion enclosures so most workloads will need to virtualize to utilize. He also adds, “there’s been a reluctance to use VIOS because it’s based on AIX. That’s completely different for the iSeries administrator. I [Tim] would say that most customers overestimate some of that involvement. There is quite a lot of automation and graphical interfaces to the VIOS and the virtualization, so they can do a lot of the management from the HMC now without having to log in and run AIX commands.”

Mr. Woodie and Tim highlight, “Just as logical partitioning took some getting used to when it was introduced so long ago, VIOS is finding some resistance in the IBM i installed base. But shops are starting to warm up to VIOS, especially since IBM has started supporting the Storwize storage arrays there. The Storwize products are proving hot sellers for IBM and its channel; in particular, the midrange Storwize V5000 is finding a lot of buyers among the IBM i faithful.”

Evolving Solutions became the first IBM Business Partner to install the new Power8 server technology in our on-site technology lab. Big Data is an ever-recurring theme in the current IT landscape. The new IBM Power8 servers will be a significant help to companies who are faced with managing the continuous explosion of data generated from social and mobile computing.

The Problem with Enterprise Monitoring

Posted on

By James Keating III, Business Technology Architect, Evolving Solutions

Enterprise monitoring, can there be anything that on the surface appears so straightforward but is indeed so vastly complex?  I have spent years working in various capacities with enterprise monitoring from an active administrator of the monitoring software, to an architect of alerts, process and procedure around creating enterprise monitoring.  In all of these endeavors it has become apparent that monitoring is something everyone does, but rarely is it done with all of the following in mind:

  • Truly adding business value
  • Using a cost effect methodology
  • Providing data that is useful
  • Improving IT, not just adding complexity

Sure many IT groups do a good job with monitoring, but they are usually an island and are good at monitoring the stuff they know well, care about or are directly on the hook for.  Anything that falls outside that realm is somebody else’s problem.  I am referring to the situation that if IT were to be willing to admit it, is happening often, and that is monitoring broken apart by application, team, group etc.  The backup guys use tool X, and the UNIX guys use tool Y, the monitoring group uses tool A but really needs to see the data from what the DBA group is doing with tool B.  What is lost in this all too common equation is business value and cost effectiveness as the approach focuses on reaction, not planning.  Good monitoring should not only have a reactive alert based approach, but also have a planning component.  After all, you are gathering data from your entire infrastructure, shouldn’t this data be put to use to drive the business forward?

I believe monitoring falls into the following categories:

  • Break/Fix or operational monitoring
  • Business planning and baseline performance concerns
  • Change control and governance requirements

All of the above areas have different needs and requirements and most tools available only address a small portion of each.  That is why the various groups choose different tools and monitoring becomes a complicated web of intermingled data with little analysis overall of the data generated.  If we take a step back and choose to  address the business performance and governance angles, and move the reactive monitoring (operational monitoring from above) into the various layers that would need to respond to alerts we can create a better monitoring plan and picture that does address business needs.  Basically operations groups usually have the skills to perform good break/fix monitoring (that is why we get the tool sprawl from above), but often planning and performance are not the focus, it is that old saying about how hard it is to remembe to drain a swamp when it is full of alligators.  The tools used for operational monitoring are reactive in nature, they are really not meant to provide a good method to analyze performance over time, let alone any forecasting ability.

The key to truly monitoring the enterprise is to realize the competing priorities of the groups involved, but still find a way to gather a baseline across all the disciplines that allows for forecasting and real insight into the infrastructure layer.

If you would like to find a way to get better performance and planning data, it may be time to look at a different way to monitor.  At Evolving Solutions we can help you build a forward looking monitoring solution for areas 2 and 3, that doesn’t complicate operational monitoring.  Further, we can help to design a solution that is not labor or cost intensive.  Contact us to find out more about monitoring for the business.

________________________________

James Keating III is a Business Technology Architect for Evolving Solutions. James is a technology leader in the IT community and brings a combination of excellent technical skills and business acumen which allows him to guide customers in developing IT solutions to meet business requirements.

BYOD Tips: Discarded Devices

Posted on

Not all bring-your-own-devices (BYOD) provide the same level of ease for lock-down or remote wipe if lost, stolen or discarded reports David Taber for Computerworld. For example, devices with Flash RAM which is soldered into the device can hold on to content long after the device is no longer in use. When these devices are discarded by the owner, sold or even donated to charity, this presents the possibility of outsiders accessing company information. Mr. Taber provides the following tips to help boost your mobile security:

  • Always encrypt and password-protect data anywhere you can on the device
  • When you can, require passwords on every sign-in
  • Use the latest tools and tech for information leakage prevention, data loss protection, auto-erase and call-home software on the device
  • Enforce updates by the user of protection tools. Registering devices to the network will allow you to monitor for lapses in use. Mr. Taber recommends, “once the device is registered on the network, the device owner should be contacted via phone if the device has been off the air for more than a month”
  • Establish your BYOD policy and allow only users to add their own devices if they are willing to comply. In your policy layout the rules or policies for selling, recycling or donating devices as well as what to do if lost or stolen
  • Develop your protocol for locking down data or changing passwords inside and outside the network. Establish a hotline or clear procedure for users to follow if a device becomes lost or stolen
  • Mr. Taber also recommends creating a BYOD FAQ that employees must read before connecting their device to company data

What BYOD procedures do you have in place to protect data on discard devices?

Ghost Servers: Unplug

Posted on

Patrick Thibodeau reported on Network World that the Uptime Institute estimates “20% of all racked IT equipment has no earthly function.” These are sometimes called ghost servers caused by expansion from new applications that over the years have either run their course or have very little use, but the servers, equipment, still remain – taking up power and data center space.   The Uptime Institute has found that unplugging or decommissioning these servers can save businesses real dollars and the institute holds a contest to draw attention to this issue.

This year’s top winner, Barclays, removed 9,124 physical servers in 2013. Those servers consumed 2.5 megawatts of power and Mr. Thibodeau reports, “could fill 588 server racks.” Barclays estimates its power bill would be $4.5 million higher if the systems were left running. The article further reports that Barclays saved $1.3 million on legacy hardware maintenance costs and  freed 20,000+ network ports and 3,000 SAN ports.

Why is decommissioning not done more regularly?  Mr. Thibodeau’s article quotes Uptime Institute vice president of energy programs, “decommissioning can be a lengthy process involving both IT and the business interests, which may be focused on developing new products — not on getting rid of old equipment. Moreover, decommissioning may involve porting an application to a virtual platform and private cloud environment.”

When was the last time you reviewed your server utilization? Perhaps there are a few “ghosts” in your data center.