Support Costs: The Elephant in the White Box Room

Arthur Cole
Slide Show

Embracing the Software-Defined Future: Looking Ahead at 2015

Whether you are planning a traditional data center build-out or all-new cloud infrastructure, the appeal of white box hardware is difficult to resist.

Provided you need enough of a particular device to benefit from economies of scale, and you have a plan to layer all the functionality you need via software, white box infrastructure can do wonders to reduce the capital costs of any project. Plus, you always have the option to rework the software should data requirements change.

But it isn’t all wine and roses in the white box universe. As IT consultant Keith Townsend noted to Tech Republic recently, white box support costs often emerge as a fly in the ointment. Large organizations like Facebook and Google have the in-house knowledge to deploy, configure and optimize legions of white boxes, but the typical data center does not. It takes a specialized set of skills to implement software-defined server, storage and networking environments, and white box providers as a rule do not offer much support other than to replace entire units, even if only a single component has gone bad. There is also the added cost of implementing highly granular management and monitoring tools to provide the level of visibility needed to gauge a device’s operational status to begin with.


Of course, white box is not the only alternative to souped-up proprietary platforms. It turns out that many consumer, prosumer and SOHO devices are sporting advanced enterprise capability as well, according to InfoWorld’s Paul Venezia. A $60 wireless access point, for example, can now provide WPA2 Enterprise access and LDAP authentication, while new low-cost firewall technologies like pfSense offer advanced functions like load-balancing and QoS on multiple links, with full failover redundancy to boot. And in many cases, the management interfaces of these devices are far less complicated than their higher-end brethren since they are aimed at non-tech-savvy operators.

But even if the enterprise sticks with traditional platforms for legacy environments, the growth area going forward is cloud infrastructure, which is characterized by massive scale-out architectures that hold particular appeal for commodity solutions. This is why we are seeing the rise of the branded open hardware model, says Big Switch Networks’ Kyle Forster. As he explained to SearchSDN, top vendors like Dell and Juniper are hoping to thread the open source needle by implementing open computing on key product lines. This would allow users to implement alternative operating environments, like Big Switch’s Switch Light OS or Cumulus’ Linux distribution, while still offering key benefits on the hardware side. This is a risky strategy because it is based on the idea that these benefits can be implemented better or more cheaply in hardware than software. So far, this has not caught on with the top hyperscale consumers but is still largely up in the air when it comes to the broader enterprise market.

ROI Value

IBM, in fact, has taken this concept, also known as bright box, to a new level by opening up its Power architecture and then forming a development community to support it. The OpenPower Foundation just celebrated its one-year anniversary and counts about 80 members ranging from chip-level solutions to those working on higher-stack software development. The company is banking on the notion that advanced capabilities on the Power chip, such as the Coherent Accelerator Processor Interface, will enable a high level of customization within the platform, enabling developers to provide a “buffet-style” approach to memory, I/O and other vital functions.

The main advantage that white box and bright box solutions have is their ease of deployment and lower capital costs when implemented on a broad scale. But this should not obscure the fact that what is gained in the purchasing and deployment phase tends to ebb away on the integration and operational side.

Open solutions do not necessarily produce plug-and-play, set-it-and-forget-it infrastructure – it takes a lot of skill, and a fair bit of trial and error, to get it right.

Arthur Cole writes about infrastructure for IT Business Edge. Cole has been covering the high-tech media and computing industries for more than 20 years, having served as editor of TV Technology, Video Technology News, Internet News and Multimedia Weekly. His contributions have appeared in Communications Today and Enterprise Networking Planet and as web content for numerous high-tech clients like TwinStrata, Carpathia and NetMagic.



Add Comment      Leave a comment on this blog post

Post a comment

 

 

 

 


(Maximum characters: 1200). You have 1200 characters left.

 

null
null

 

Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.