Ptak, Noel & Associates. Here, he takes a look at how going green applies in network operations.
First, let's make it perfectly clear this isn't about saving the Earth. The world's climate is dynamic. At one time, vast parts of the world were steamy, festering swamps that eventually were covered with miles of thick glaciers. Over eons of time and climate, we have what we have today, a world that swings between colder and hotter temperatures. Mankind and animals have displayed a remarkable adaptability to climate and weather. Today, the world's humans (and an array of other species) live comfortably in places with average annual temperatures that differ by some 30° F -- e.g., Helsinki, Finland, and Singapore.
The focus here is on efficient network operations because it is part of a network professional's job to contribute to the success of his/her company -- and one way to do this is to cut waste and inefficiency. This is becoming more important as energy costs increase.
The cost of power will not stop rising. The laws of economics are as fixed as any laws can be. Rising demand for a scarce commodity drives up its price. Thus, using less energy will become more and more important as energy costs continue to climb.
Now, what's to be done with networks?
First, have a plan. You can't just jump in and start unplugging devices and dialing back services. Think about what you want to do, what can be done, and how you will go about moving forward. Identify inefficiency, waste and over-provisioning, understand and document business impact, and prioritize your actions.
A lot of companies have delayed upgrading infrastructure for the past few years. If things are really starting to fray, there is a good case to be made for a budget to upgrade the network infrastructure refresh. So start with an inventory of network devices, servers and switches. Find out what gear is and is not being used. Identify candidates for consolidation, replacement, upgrade and removal. Understand how things are being used. Know what is most important from the perspective of business operation.
To bolster your case, do an energy profile of your network operations center (if you have not already done one) to monitor, track and report how resources are being used and how energy is being consumed, and where and by whom. Identify the most inefficient energy hogs. You need to understand what is happening before you can decide what you need to do and what can be done. Then, research and evaluate alternatives. You need to know not just what can be done but also the cost, savings and time involved. You will need such documentation for management support.
A "refresh" doesn't have to mean buying the latest technologies. Before purchasing or reconditioning, consider virtualization and consolidation. For years, servers have been underutilized, running at 10% to 18%. Virtualization, properly implemented, can raise that to the 70% to 80% range (considered a best practice today).
Also, consider reconditioned and refurbished equipment. Major vendors such as IBM, as well as specialist vendors such as Network Hardware Resale, World Data Products Inc. and Sapia Networks, have large stocks for purchase. Reconditioned or recycled gear comes at a lower price, some (check this out) with warranty and support rivaling new gear. Such equipment saves resources by extending the life of existing products and reducing resources consumed in buying new. However, be sure to do your homework to avoid counterfeit gear or support problems.
If the budget is there, look at devices utilizing the latest energy-management and savings technologies. Chipmakers like Intel, as well as box suppliers like IBM, HP and the rest, are doing remarkable things to dynamically reconfigure their processors to provide maximum utilization with minimal power consumption based on the use of workload priorities and business policies to determine action.
In addition, there are intelligent management tools that use a range of data sources -- from workload run-times to time-of-day to business impact and priorities -- to make configuration changes, adjust workloads and scheduling to maximize server utilizations, and reduce power demands.
From a general perspective, educate yourself about intelligent power management embedded in hardware. Talk to peers to find out what they are doing. Read the literature to search out articles and information about industry best practices. Manual processes and workflows are inefficient; aggressively pursue workflow and process automation for management, problem resolution and root-cause analysis.
Ignore ranting about biofuel-based independence. There just isn't enough arable land, and there is too much competition (human and animal) for foodstuff raw materials such as corn, sugar beets and roughage. Also, these form diet staples (e.g., some 300 different kinds of tacos in Central and South America) for a great many of the world's people who can't afford Starbucks, arugula or even Twinkies! How much sense does it make to burn foodstuffs (which are 20% less efficient and consume 20% more resources to produce) for transportation and energy instead of inedible, more efficient oil and gas?
Comments? Questions? Disagreements? Corrections? Send your thoughts and comments to: email@example.com
About the author:
Richard Ptak is founder and partner at Ptak, Noel & Associates. He has more than 30 years of experience in systems product management. He was VP at Hurwitz Group and D.H. Brown Associates and worked at Western Electric's Electronic Switch Manufacturing Division and Digital Equipment Corporation. He is frequently quoted in the trade press and is the author of Manager's Guide to Distributed Environments.
This was first published in April 2008