Selection of Hybrid Cloud Vendors

February 22, 2012

How do you evaluate and select a hybrid cloud vendor; it really depends on the problem(s) that you need to address and solve. If data transfer and storage are critical then the most important issues are those of bandwidth and data transferring. If the system needs to support bursting, or spikes in web traffic and/or computation loads then price may be more important.

What follows is a description of our evaluation and comparison of vendors based on our needs for a hybrid cloud infrastructure that provides a private cloud (more like managed hosting) and a public cloud (that provided elastic/on-demand computing resources).
Please bear in mind that your needs are likely to be different.

Cloud computing vendors were evaluated using the follow criteria:
Completeness of Hybrid Offering – By vendor or in combination with 3rd party.
Maturity of Offering(s) – Relative length of time vendor has been providing hybrid offerings.
Cost – Total costs.
Reliability – SLAs for the private and public clouds
Bandwidth and Data Transfer – Maximum bandwidths for data transfers between clouds.
Self-service Support –
Developer Support – How ‘developer friendly’ is the infrastructure (and vendor).
Portability of Deployments – How easy is it to move deployment from one vendor to another.
Integration Support – Support for open standards or public APIs for integration.
Security – Tools and capabilities.
Management – Management tools for both public and private clouds.

1) As portability of deployment is a critical requirement, no PaaS solutions were considered as those solutions, by their design and implementation, are not portable.
2) Computing and storage costs are highly dependent on configurations.
3) Portability of IaaS cloud implementation can be heavily dependent on how the systems are configured and deployed.

The following vendors were considered as we believe that their current offerings could address most of our evaluation criterion: AWS, ATT, Datapipe, Go Grid, IBM, RackSpace, Terrmark.

Potential candidates
Datapipe – Has strong manage hosting and ability to hybridize Amazons solutions with its own. Claims seamless integration between AWS and Datapipe environments, high I/O performance, and integrated support and management.
Go Grid – Smaller, independent provider of public and private clouds. Very high SLAs. Competitive pricing. All APIs are proprietary, portability may be an issue.
RackSpace – Strong managed hosting. Open source development via OpenStack project. Offers some hybrid configuration. Is moving quickly to provide fully featured, hybrid offerings.

Vendors that are lacking in one/more critical areas
ATT – Very strong in managed hosting. And, Synaptic Compute is an ambitious offering. However, the services appears to still be in beta (not fully released)
AWS – Amazon does not provide a native, hybrid cloud offerings and they do not provide non-virtualized servers. They do provide hybrid offering in partnership with 3rd party vendors (e.g. Equinix) via Direct Connect. However, that would require us to provision two separate clouds with two different vendors.
CSC – Nascent hybrid solutions.
IBM – Strong managed offerings. Complex contracts and pricing structures. Focused on large enterprises. Level of commitment to full set of hybrid offerings is unclear at this time.
Terrmark – Moving quickly into the hybrid cloud space with the acquisition of CloudSwitch. However, their hybrid offering are relatively new.


The Cost of Cloud Computing

May 30, 2010

Yes, I know, price isn’t everything, it’s not the only thing, but enterprise computing costs do matter!  If you don’t think so, then just imagine the conversation you will have with the C-level/VP/Director/Manager person when you go in and ask for X percent (say 10, 20, 30 or more) of this year’s Cap-X budget to fund the hardware  for your latest and greatest project (and don’t forget to include the support costs).  Oh yea, that will be a ‘fun’ conversation.

Cost of Cloud Computing Resources
Late last year (2009) Amazon, Google and Azure lowered their published pricing for reserved computing instances (computing cores).  Amazon’s rate for a single CPU, continuously available cloud computing instance was little as 4 cents an hour (effective hourly rate based on 7×24 usage) for customers that sign up for a three year contract.
Single year contract rates were about 20% higher.  Pricing for on-demand instances (no upfront payments or long-term commitments) was about two and a half to three times the three year contract rate.
A rough calculation says that a cloud data center of 10, single core servers (at the three year contract rates) could be operated around the clock under $0.50 an hour, or just under $3,500 a year (that includes servers, data center facilities, power, cooling, and basic operations). That’s about $350 per server per year – pretty impressive!

Commoditization of Cloud Computing
And if the costs of cloud computing weren’t low enough Amazon announced pricing for EC2 ‘spot instances’.  This pricing model will usher in the beginnings of a trading market for many types of cloud computing resources: support services, storage, computing power, and data management.
Under the old model you had to pay a fixed price that you negotiated with a bulk vendor or a private supplier.  Now in the new spot market you can look that the latest price of available cloud capacity and place a bid for it.  It your bid is the highest, then the capacity is yours. Currently this is available from Amazon’s EC2 Cloud Exchange.

TAG Enterprise 2.0 Society – November 5, 2008 Meeting

October 8, 2008

TAG Enterprise 2.0 Society – November 5, 2008 Meeting

Cloud Computing – Amazon Web Services

In this session, Seattle-based Jinesh Varia, Evangelist for Amazon Web Services, will discuss the latest innovations and new technology trends like Utility computing (Paying by the hour, paying by the Gigabyte usage), Virtualization and Web Services in the Cloud and most importantly, discuss some of the innovative business models for Start-Ups and Enterprise companies.

In this session, we will learn how aspiring entrepreneurs and enterprises can take advantage of these technologies to quickly scale up their infrastructure programmatically without any upfront heavy infrastructure investment. Often termed as Cloud Computing, we will see how these technologies are changing the way we do business today.
Amazon Web Services provides Amazon Elastic Compute Cloud (Amazon EC2) that allows requisition of machines on-demand using simple web service call and paying for computation by the hour. Amazon Simple Storage Service (Amazon S3) which is infinite storage in the cloud and Amazon SimpleDB which is the Database in the cloud and how these services can help local companies to scale-out and go live quickly. Also, we will see some exciting apps and some unique business models that are built on AWS that have become profitable businesses and others that are just simply cool to see.

As a Technology Evangelist at Amazon, Jinesh Varia helps developers take advantage of disruptive technologies that are going to change the way we think about computer applications, and the way businesses compete in the new web world. Jinesh has spoken at more than 50 conferences/User Groups. He is focused on furthering awareness of web services and often helps developers on 1:1 basis in implementing their own ideas using Amazon’s innovative services.
Jinesh has over 9 years experience in XML and Web services and has worked with standards-based working groups in XBRL. Prior to joining Amazon as an evangelist, he held several positions in UBmatrix including Solutions Architect, Enterprise Team Lead and Software engineer, working on various financial services projects including Call Modernization Project at FDIC. He was also lead developer at Penn State Data Center, Institute of Regional Affairs. Jinesh’s publications have been published in ACM and IEEE. Jinesh is originally from India and holds a Master’s degree in Information Systems from Penn State University