FOCUS OPINION
Issue 2, February 2009
SPEED UP, SPEND WISELY This year’s solutions must focus on consolidation and efficiency, explains Carrie Higbie
As the new year began and I turned (a very young) 48, I was thinking about the changes in our industry over the now 28-plus years I’ve been involved in it. For those of us old enough to remember punchcards and baud rates, we see the same principles come full circle. We have gone from centralised to de-centralised, back to centralised data centers; virtualised (mainframe) to individual servers and back to virtualised (various platforms); into the Windows boom and now back to various other operating systems; single storage towers to separate storage, then back to SANs, NAS and shared storage; outsourced to in-house to outsourced again, etc. There are only a handful of computing demands that seem to grow in a single, uphill linear direction and keep going. On the other hand, spending is changing, with a more watchful eye on return. In short, speed up, spend wisely.
On the incline, data centers today are ramping up to support 10GbE copper and fiber applications; 40Gb and 100Gb are right around the corner. When I speak at conferences around the globe, it never ceases to amaze me how many times I still hear: “I don’t need to go any faster on my LAN, my internet connections are still T-1/E-1.” Now, granted, if all packets used that link, that bottleneck would kill a network in short order. But in data centers in particular, most of your traffic doesn’t use that link. Redundant offsite storage uses its own link in most cases, and vendors have done a great job of transmitting only necessary bits and bytes. It’s time to get rid of old-school thinking!
Telecommunications carriers aggregate and trunk services on wide- area circuits to allow for higher and higher speeds at more reasonable pricing than ever. In fact, many companies don’t even have a data center at their location. With the increase in overall wide-area bandwidth, it isn’t necessary for your data center to be similarly located anymore. You can put it where power is cheaper, skilled labour pools are larger, and cooling comes from nature.
Storage has to be a close second to the ever-increasing need for bandwidth. Not only do companies see data as a competitive advantage, but increasing legislation regarding data retention has put storage demands at an all-time high. Not only do we need the data, we need it redundant and accessible in real time. It’s a good thing that storage is getting smaller in physical dimensions, or we wouldn’t have room for desks!
While not a technology per se, the third expected increase has to be value to each monetary unit spent on technology. This is the single most critical factor for data centers. In the tech boom, people bought new toys and technology just for technology’s sake. With recent turns in the economies of many countries, companies are learning to spend
52
www.datacenterdynamics.com
their money more wisely, with a more scrutinising eye to payback ratios. This year I predict that companies will focus on tuning spending in the following areas.
Improvements in facilities is tops, with increasing power costs and lack of available power in many areas. Government mandates in the US and by the EU are contributing factors, but the bottom-line number for power is the largest catalyst; being greener is the byproduct. Granted, facilities is a big category and the term may mean different things to different people. So to narrow this down, decommissioning old servers, turning off non-essential consuming devices, improved CRACs and air handlers, free cooling, removal of old cabling and obstructions, and redistribution of equipment to eliminate hot spots will be the focus.
Next, standardisation projects will increase. This will include everything from infrastructure standards to platform standards. I work with many G500 companies, and many are completing or looking towards standardisation projects. VoIP started many of them as they found out that letting each site do its own thing was not a good thing. Expensive systems deployed globally over poor physical infrastructures were a costly endeavour.
There is a reason why some vendor’s products cost more. They provide value-added services and education to assist end users, pay attention to
quality for installation companies and will be a business partner, not just a part vendor. This often overlooked and relatively free addition to your IT staff should be utilised. While the products may cost slightly more, the paybacks over time are immense. Being a value-added resource myself, I can tell you the companies that I work with globally take advantage of services that make their lives easier. There is no need to reinvent the wheel, and there is benefit in learning what has worked for others and what hasn’t before you jump into a project.
Also on the consolidation and spending watch list is getting away from having multiple platforms to support. No company is telling their IT department today, “Here is more. Please do less.” Companies buying the cheapest here or there – one of these and one of those, and myriad vendor products – suffer from interoperability problems, training and maintenance nightmares, huge wastes of resource time and eventually lose the economy of scale.
So, as the new year starts, it’s time to upgrade wisely. Keep an eye on long-term technology investments and maximise the free vendor resources that are available to you, even if it means changing vendors. ROI calculations should be based on your actual expenses, not on what- ifs. And, most importantly, have a wonderful and safe 2009!
Carrie Higbie is Siemon’s global director of data center solutions and services
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56