Danish physicist, Niels Bohr reportedly quipped that “prediction is very difficult, especially about the future.” With that in mind, I’d like to share some of my personal thoughts on five important cloud computing trends and developments we can expect to see in 2013 and beyond.
1. The Cloud Condensed
If the past couple years can be summed up as an explosion in cloud-based platforms and service offerings catering to a steady increase in cloud deployments, 2013 will be a year of maturity and consolidation.
Currently there are dozens of vendors – from giant AWS to specialized CloudSigma or ProfitBricks – all offering up various flavors of public, private and hybrid IaaS. Third-party vendors such as Scalr, New Relic and Cloudyn (where I just so happen to be employed) respectively fill in the service gaps with deployment automation, performance monitoring, utilization analysis and cost optimization. System integrators with virtualization expertise design and build data-centers (on and off-site) for custom cloud deployments.
But like the B2B and B2C internet marketplaces that quickly ballooned and then burst at the turn of this century, there will be some serious cloud contractions with many players being acquired or going out of business. Without a question, the bigger players like Amazon will buy out competitors and complementary service providers, or just push the smaller players out of the market with their economies of scale.
2. Localized Cloud Services
This past October I had the pleasure of being on several panels at both the GigaOM Structure conference in Amsterdam and The Next Web’s Web Summit in Dublin. Talking with an array of European companies and executives made it clear to me that people still prefer dealing with local service providers. Whether this is due to language, culture or even if it makes sense from a financial perspective is immaterial. From what I have seen and heard, there is definitely a market for local IaaS.
Telefónica seems to thinks so too. Their recent announcement about their Instant Servers, an infrastructure-as-a-service, positions them as direct competitors with Amazon and other large cloud vendors. Based out of London and Madrid, the telecom giant should be able to easily leverage its existing customer relationships to power rapid growth.
Telecoms in general are better positioned than AWS to compete for enterprise business. With strong, established relationships and trust already in place, telecoms will leverage this advantage to win customers to their cloud offerings.
3. Replicating the MVNO Model in the Cloud to Penetrate Niche Markets
Continuing on to a related vein of Localized Cloud Services, there is also a challenge that all cloud vendors are experiencing – trying to get into niche markets. These niche markets could be a geographical area as mentioned above, or it could be a market segment such as gaming, content or even by customer size. A white-labeled solution with special prices and terms will likely be introduced soon by ‘cloud sub-contractors’ to cater to a focused niche market approach that meet geographical or market specific requirements. If there are virtual call centers and virtual mobile operators, why not do the same in the cloud?
4. Cloud Computing Tailored Just for You
Did you ever try squeezing a square peg into a round hole? It doesn’t really work so well. The same is true in cloud computing. Forcing customers to adapt their computing needs to standard offerings often leaves them with either too little or too much cloud resources. Although Amazon’s no-frills, self-serve offering appeals to many users, there is growing demand for custom-sized and optimized computing packages. Companies like CloudSigma allow you to build your own package with any “unbundled” combination of resources, so you can purchase CPU, RAM, Storage and bandwidth all independently.
On the service side of the cloud, the customer experience will definitely become a differentiator. Companies like Rackspace understand this with their “fanatical support” and usage of the Net Promoter System for maintaining high-levels of customer satisfaction.
5. The Expedia of Cloud Computing
When the cloud got started, the ease, agility and scalability were such a big draw that many companies jumped in feet first without deep analysis of costs and benefits. However, as the market grew up, real-world business concerns about cost and ROI began to surface.
Market leader Amazon quickly reacted to the cost concerns and began offering more cost-effective options such as their Reserved Instance pricing plans. Here at Cloudyn, we actually have seen a huge upswing of AWS customers moving towards Reserved Instance pricing from just 29% of customers using them in January to over 50% today.
As mentioned before, cloud users are looking for better service, and best-fit deployment in addition to optimal pricing and utilization. This has created an opportunity for a “cloud exchange” where compute units and other cloud resources can be bought and sold freely. Cloud consumers will be able to shop around for the best deal, akin to how travelers use sites like Expedia to build their perfect vacation package (I first made this analogy during the ‘Trading Computing Power as a Utility’ panel at the GigaOM Structure conference this past October). They also will be able to offload unneeded resources, helping to keep vendor pricing competitive with real market demand.
Challenges to Developing a True Cloud Marketplace/Exchange
There are many hurdles to make this a reality. While Amazon’s Reserved Instance market is an early attempt at this, it will likely take a few more years before materializing as a truly liquid market.
Some of the issues that need to be addressed include vendor standardization of cloud resources and compute units. This is necessary for buyers to accurately compare cloud packages from different providers with an ‘apples to apples’ clarity.
To be honest, I don’t see this happening from the vendor side (as they have an interest in keeping things opaque), but I do believe that we will soon see 3rd-party technology that enable cross-vendor evaluations and comparisons.
Another area that will affect market liquidity is the fact that data has gravity. The larger the volume, the more difficult it becomes to move data to a new location. Businesses managing more than 100 gigabytes of data will need to keep their data local (which might play nicely into the growing private and hybrid cloud market). Although bandwidth technology is growing, it will never match the rate at which data is being collected and created – creating a paralysis in moving data around to different vendors. However, there might be solutions where you can “bring the compute to the data.” If this works, than these companies managing petabytes of data will be able to significantly impact the liquidity of a cloud compute exchange.
Incidentally, this issue of data gravity is probably the answer to why many companies, despite repeated AWS outages, are still keeping their operations in Amazon’s problematic US-East region.
Towards a More Efficient Cloud
The main value propositions of the cloud have always included agility, scalability and affordability. But currently, the cloud is not nearly as efficient as it could be with customers getting stuck with resources they don’t need and bills they don’t expect. Trying to optimize one’s cloud consumption can also be very difficult as cloud vendors don’t provide adequate tools for visualizing and understand cloud cost and usage.
All of the trends I foresee will ultimately lead to a more efficient and transparent cloud where the consumer benefits from getting exactly what they need, when they need it, and at competitive market prices.
See how you can make your cloud more efficient and effective with our new Reserved Instance Calculator! Sign-Up for FREE Now!