Mapping Amazon Web Services

Chances are, your customers already know about Amazon Web Services do you? Our experts' overview can help you get up to speed.

Channel Partners

December 27, 2012

10 Min Read
Mapping Amazon Web Services

Jo Peterson By Jo Peterson and Manon Buettner

Industry-leading service providers such as Rackspace and Savvis took 12 years (and some acquisitions) to reach $1 billion in revenue. Yet in half that time, Amazon Web Services (AWS) has gone from zero to nearly $2 billion and more than a half-million customers an absolutely stunning rise especially considering the non-traditional sales and marketing tactics employed to launch an entirely new delivery model.

Chances are that most channel partners have at least one customer utilizing AWS. It has become so successful that one analyst pegs its market share in the infrastructure-as-a-service (IaaS) space at more than 70 percent, as compared with its nearest competitor Rackspace, which has slightly more than 10 percent of the market. But while weve all heard of AWS, do we really know where it plays? More importantly, how do we intelligently compete with this offering?

The fact of the matter is, our customers know what it is and are using it whether IT management knows it or not! Further, it is often considered the baseline by which other compute and storage alternatives are gauged. Our job as consultants and agents is to internalize the AWS offering and communicate to our clients where it makes sense to consider alternate private or public clouds based on their business, technical and cultural requirements.

In order to get up to speed on AWS, well highlight four key areas:

  • Product portfolio

  • Differentiators (strengths, weaknesses, potential traps)

  • Terms (pricing, contracts, support and SLAs)

  • Competitive landscape and futures

Product Portfolio

AWS has a rich IaaS product portfolio, with 11 distinct product groups: compute, content delivery, database, deployment/management, application services, software, networking, payments and billing, storage, Web traffic and workforce. The most well-known and referenced of these are the compute and storage offerings, which include Elastic Compute Cloud (EC2) and Simple Storage Service (S3).

EC2 first launched in 2006 as a commercial Web service allowing small companies and individuals to rent computers upon which to run their own computer applications. Similarly, Amazons S3 offers data storage for retrieval at any time and allows users to read, write and delete objects containing from 1B to 5TB of data each. The number of objects you can store is unlimited.  So simply put, EC2 replaces servers while S3 replaces non-production storage.

Other products are growing in lock step. In August 2012 Amazon announced Glacier, an off-site storage alternative, for merely 1 cent per gigabyte. A poor mans disaster recovery plan is solved for only $10 per month with 1000GB (1TB). This is an extremely attractive option for small businesses or those with no intention of retrieving the data short of a catastrophe.


While there are other companies with equally viable products and services, AWS has built a name for itself. Not only does everyone know of AWS, the brand is practically synonymous with The Cloud.” It is consistently recognized by analysts and media as a market leader and industry innovator. This, in and of itself, is a key differentiator for the company.

The AWS product portfolio has depth and the company quickly has been able to bring products to market and adapt those products based on customer feedback. AWS is thought to have the largest pool of capacity, which is key for short-term deployments of hundreds of servers at a time. Customers would describe AWS products as flexible and able to scale to their needs. Also, the go-to-market strategy of offering prebuilt Web services (e.g., billing, queue, database, etc.) accelerates efforts and reduces upfront investment.

AWS continues to add products and drive down prices. Many believe it is the least expensive cloud; however, this is not always the case once the connectivity, support and any third-party plug-ins are added in.

The product suite is suitable for bursty” workloads and short-term provisioning, so workloads in the test/development space, for example, work well in this model. AWS has colocated compute (EC2) and storage (S3) in the same environment recently to overcome issues around latency/bandwidth limits. AWS supports a wide variety of OS, apps and images.

There is a low barrier of entry for AWS in terms of cost and prebuilt services. AWS is known as a market price leader with a strong brand and partner ecosystem. This ecosystem has full API access to infrastructure and has built software specially packaged to run on EC2. There also are a growing number of third-party management tools, such as RightScale and VMLogix, that are designed to work with the product sets to solve concerns, such as portability, centralized billing and budget governance.


Every product and service has some inherent weaknesses, and AWS is no exception. While the virtualized product suite is robust, there are no brick-and-mortar products available.  Today, AWS does not offer colocation, managed services, dedicated non-virtualized servers and private, non-IP based connectivity. There is one basic class of service with limited security capabilities. The offering is developer-centric, not enterprise-oriented. The user interface tool can be cumbersome and a bit clumsy to use. While the feature suite is robust, it can be confusing and accessible only via proprietary API. 

In addtion, tiered phone support retainers come at a premium price and are not included. The pricing format can be confusing as its highly variable and unpredictable. While there are enterprise agreements available for large customers, there doesnt seem to be consistency. IP indemnification does not seem to be available, and terms will vary across services and even states.

One of the largest weaknesses is around SLAs, with credits in the 10 percent range and a requirement of at least two availability zones being down.

Consider transparency when you are comparing AWS against other providers. It’s often unclear where the AWS infrastructure physically sits and where the data actually resides. Typically the burden is on the customer to determine which AWS solution is the best fit for them. The online approach to customer fulfillment is usually not suitable for high-touch” enterprise clients. While most clients are using VMware, the back end of AWS is Zen-based.

And finally, the proprietary technology AWS uses is fine for greenfield environments, but is often difficult for a client who is migrating from one cloud provider to AWS, since apps need to be rewritten to mesh with the AWS APIs. Without the assistance of RightScale or a similar management tool, instances would be unable to port to another provider, which results in vendor lock-in.”


“Terms” include contractual components, such as pricing, support and service level agreements (SLAs. Most service providers have either a master service agreement (MSA) plus relevant order forms or a simple self-service ordering interface with inclusive terms. The reason AWS has until now been predominantly adopted by users more so than IT executives is because of the modularity and elasticity of its model with no need for a contract a truly self-service, usage-based delivery model.

Pricing. From a pricing perspective, the true cost is highly variable and seems to depend upon location, server OS and processing demand. Pricing includes a one-time fee, plus service fees (instances) that are volume based. Data transfer fees and features, such as load balancing and redundant architecture, are additional as is technical support.

The pricing model includes an estimating calculator, but it is often challenging for customers to ascertain the price correctly since so many components are variable. A true comparison would need to include storage upgrades, as persistent storage is often needed, bandwidth between servers (depending on the design) and bandwidth costs in and out of Amazons network. These costs are not generally part of the model your client sees. If your client is architecting across availability zones” to take advantage of more robust SLAs, they will need to factor additional costs and plan to pay for data transfer between zones. Pricing also is delivered as an hourly number, so longer terms need to be specially requested. 

Billing. Billing is done via a credit card. Traditional corporate invoicing must be negotiated as a special request (or via cloud service broker.)

On the compute side, customers will pay for what they use with no minimum fee. Amazon EC2 is currently available in eight regions: U.S. East (Northern Virginia), U.S. West (Oregon), U.S. West (Northern California), EU (Ireland), Asia Pacific (Singapore), Asia Pacific (Sydney), Asia Pacific (Tokyo), South America (Sao Paulo) and AWS GovCloud. Pricing varies across regions with the U.S. being the least expensive, followed by EU, APAC and South America. 

Pricing elements include size, OS and type of instance (on-demand, reserved and spot being the options), data transfer and storage. Additional pricing options include server imaging, backup, monitoring, custom stats/metrics, load balancing and virtual private cloud.

Support. From a support perspective, AWS has recently added multitier support levels beyond user forums to address gaps.  Even so, business and enterprise-level support can run 10 percent of usage and is geared toward the technically knowledgeable, expert-level users. Consider this when comparing like services which on the surface appear more expensive but include comprehensive support and expert guidance.

SLAs. Amazons SLA is perhaps their single biggest weakness, with 99.95 percent availability for compute and 99.9 percent for storage based on monthly uptime. AWS puts a heavy burden on the client to prove the outage and the terms are difficult at best to understand. To further complicate the matter, the calculation is based on 365 days, so new instances will figure 100 percent availability for the balance of installed days.  It is recommended to clients to have instances in more than one availability zone a decision that would have spared many notable companies exposure in recent outages. 

Competitive Landscape and Future

The competitive landscape is evolving for IaaS at a mind-blowing speed. It has become a full-time job to keep abreast of the providers and offerings a good thing for agents. No longer an SMB play, the AWS product portfolio is quickly closing the gap between its legacy small business/startup focus and is taking on the traditional, enterprise Gartner Magic Quadrant leaders. 

AWS offers bespoke solutions to address dedicated and secure private cloud standards required by large enterprises and the U.S. federal government sector. This year AWS has begun adding territory-based salespeople, some with product specific focus, outside of Washington in a definite move to reach enterprise IT management teams and prove itself amongst other industry leaders.

The future for AWS looks promising, too. Expect more boutique providers to standardize on AWS APIs. Anticipate most comprehensive service providers to offer AWS API support in addition to OpenStack and other cloud APIs both open and proprietary. Watch for AWS to move closer to the top right corner of the Magic Quadrant in enterprise-class cloud categories.

A leader” in the most recent Gartner Magic Quadrant for public cloud IaaS, AWS continues to dominate the market as a price leader, innovator and the standard by which other public clouds are judged. It continues to respond to market demand and is moving aggressively into the enterprise by offering the terms and corporate visibility IT management seeks.

Our intent with this article is not only to educate, but also to make the point that AWS cannot be overlooked. AWS does not have an agency program in the traditional telecom sense. Instead it focuses on solution providers that wrap development, integration and management services around its infrastructure. However, for those agents moving into the cloud computing space, it is imperative to keep abreast of the AWS offering and how it compares with others you represent. Dismissing or omitting AWS from relevant conversations will damage your credibility, considering that our role as consultants is to lead with market intelligence and industry perspective.

Jo Peterson is director of Converged Cloud and Data Services for telecom agency

. Manon Buettner is principal at


, a data center and managed services consulting and brokerage firm. Both are members of the

Cloud Computing Forum (aka Cloud Girls)

, a women’s networking group for channel partners and suppliers delivering cloud solutions. Cloud Girls is a standing member of the Channel Partners Advisory Board.

Read more about:

Free Newsletters for the Channel
Register for Your Free Newsletter Now

You May Also Like