This is a great article for any IAAS/PAAS provider who is poised to compete head to head with the 800 pound gorilla known as Amazon AWS. Full article on CBR here. Digest provided as a courtesy here:
Asking whether anyone can take on Amazon in the cloud would be like someone asking Larry Page and Sergey Brin whether anyone could take on InfoSeek back in the late 1990s.
The public cloud (IaaS) market is in its infancy. Originally used for software development projects and one-time batch computing projects, today's cloud provides the foundation for Web storefronts, social networks, mobile apps, online and mobile games, and increasingly critical business services.
In a short time, the cloud will handle the traffic and computing demands of billions of devices ranging from automotive sensors, appliances, building sensors and even medical implants.
This Internet of Things may crush providers like Amazon that feature infrastructures and architectures that were never created to handle the weight of data and processing demands emerging from machine-to-machine communications.
The total IaaS global market today is $11bn, depending on who is measuring the market. That's a fraction of the global IT industry, and barely a scratch on the surface of the global telecommunications industry that tops well over $1 trillion.
Yet today's cloud has also earned a reputation for trading flexibility and low up-front costs for uncertain performance and reliability.
Already, the demands of today's mobile and Web applications and services, from games to storefronts to social networks, are overwhelming dominant cloud providers like Amazon.
No one can dispute that Amazon is the largest Cloud provider, and they played a huge role in establishing the public cloud market. They made it easy, and inexpensive for developers to use their IT infrastructure, which is the main reason Amazon Web Services dominates the market today.
Despite this early lead, it is certainly possible to compete head-to-head with Amazon. Public cloud providers can beat Amazon by delivering better performance, better visibility into their cloud, and they can actually beat Amazon on price. And you can do that if you have a better infrastructure; one that is built on a strong, resilient architecture that is designed to be efficient, resilient, and can scale to meet tomorrow's computing requirements.
Beating Amazon Performance
To beat Amazon, it's important to deliver better performance. The truth is that Amazon Web Services was never designed to handle the high-speed, production business applications running on its cloud today.Amazon's cloud was designed to provide developers with cheap infrastructure for software development and testing, and for batch processing (such as loading a database of records into the cloud, and then running an analysis to draw conclusions from the data.
Then once you were done with the batch processing or development, you ended the contract and deleted your data.
Then developers and entrepreneurs began to build their businesses in the cloud, from e-commerce businesses, mobile and web applications and social networks. Suddenly cloud performance was directly linked to revenues.
As those cloud-based businesses grew, their traffic spiked, mobile and Web app responses began to slow because the underlying cloud infrastructure was becoming overloaded by computing and storage demand it was never designed to carry.
Providing Better Stability and Visibility into Cloud Performance
Amazon's four-day outage last year showed what can happen. If you were one of the businesses taken offline, that extended downtime meant lost revenues, and lost customers. Even worse, you had no idea why the cloud was down, because Amazon doesn't provide any way to see what's causing the outage.Therefore, by providing better performance, stability and reliability, you can compete with Amazon, especially if you can make it easy for customers to see what's going on in your cloud, by allowing them to measure latency at each level in your cloud infrastructure, and in their application.
After all, if you can spot a minor performance issue, then you can fix it, ideally before it affects your service to your end customers, whether they use your mobile app, or your Web storefront.
Competing with Amazon Pricing
Let's be clear. Amazon's low price played a major role in gaining their dominant market position.To beat Amazon on price, you need to have a more efficient infrastructure than Amazon's Elastic Compute Cloud (EC2). Your cloud architecture should operate efficiently using fewer servers and storage units. It should be less expensive to manage.
Scaling the Cloud to Infinity and Beyond
Customers should be able to add computing and storage capacity without multiplying their costs. With the right infrastructure, all of this is possible.Yet with Amazon - and cloud providers that use a similar architecture - adding capacity becomes more complex as the computing load and data volumes increase. Ideally, you should double computing and storage capacity by doubling processors and storage, but with many cloud infrastructures, including Amazon's, it doesn't work that way.
Instead, inefficiencies and system complexities begin to bleed computing power, leading to a need for far more servers and storage units then more efficient architectures.
As the tsunami of data and computing demands from the Internet of Things begins to sweep through the cloud market, providers that cannot scale efficiently will be overwhelmed.
Once upon a time, it was easier for developers, entrepreneurs and IT managers to work with Amazon. That is no longer true today. While Amazon remains the dominant cloud provider, its customers are beginning to see cracks in the Amazon Cloud foundation.
For the right IaaS provider can beat Amazon head-to-head by delivering better performance, stability and scalability, and combining that with a way to see what's going on inside the cloud.
Amazon has played a critical role in building the public cloud market. Yet there is plenty of opportunity for the right provider to prosper and win as the market matures.
Steve Tuck, general manager of Joyent Cloud.
Read the full details on Quora here.
OpenSignalMaps recently released their fragmentation report and just for kicks, visualized all the phones on their network that run Android in a neat little heatmap, here:
The full site has much more data, which you can see here, but for what it's worth, is it "fragmentation" or just further proof that the truly value piece of the stack is the "software" and not the 400+ irrelevant pieces of hardware that can run the software? I think you know which way I would vote.
Get the whole story on Gizmodo here. As a former Flickr employee, it's sad to see my favorite on-line photo app go the way of the Dodo, MySpace and Friendster. I have over 24,000 images and affraid that I'm going to lose them all when Flickr is eventually folded by Yahoo corporate.
To summarize: I am increasingly cautioning folks thinking about taking the startups plunge to think twice. I’m concerned about 1. Founders who are not contrarian but following the masses, 2. The hightened practical challenges of starting a business due to the difficulty of standing out, and 3. That we are at a stage in the innovation cycle that is not favorable to less experienced and less technical founders.
via robgo.org
- The volume of all water would be about 332.5 million cubic miles (mi3), or 1,386 million cubic kilometers (km3). A cubic mile of water equals more than 1.1 trillion gallons. A cubic kilometer of water equals about 264 billion gallons.
- About 3,100 mi3 (12,900 km3) of water, mostly in the form of water vapor, is in the atmosphere at any one time. If it all fell as precipitation at once, the Earth would be covered with only about 1 inch of water. The 48 contiguous United States receives a total volume of about 4 mi3 (17.7 km3) of precipitation each day.
- Each day, 280 mi3 (1,170 km3)of water evaporate or transpire into the atmosphere.
- If all of the world's water was poured on the United States, it would cover the land to a depth of 90 miles (145 kilometers).
- Of the freshwater on Earth, much more is stored in the ground than is available in lakes and rivers. More than 2,000,000 mi3 (8,400,000 km3)of freshwater is stored in the Earth, most within one-half mile of the surface. But, if you really want to find freshwater, the most is stored in the 7,000,000 mi3 (29,200,000 km3) of water found in glaciers and icecaps, mainly in the polar regions and in Greenland.
from: Gizmodo
Cloud management isn’t just a pretty wrapper that you put on top of virtualization to make it easier to use, and it’s not a few scripts that automate builds or scaling functions. Cloud management is a platform that allows the cloud(s) owner to express their company’s directives and policies effectively and safely onto their myriad technology solutions and across international borders.
via gigaom.com
There is an inciteful article extrapolating the impact of the current Google/Oracle Java lawsuit on all programmatic APIs that got me thinking about how this would affect cloud infrastructure players (Joyent, Rackspace, Openstack, Eucalyptus, AWS, etc), orchestration providers (Rightscale, Enstratus) and any provider who clones APIs to be competible with the competition. If this lawsuit ends up giving the liked of AWS to effectively lock their APIs into a legally sanctioned walled garden, then all cloud customers will lose BIG because their apps/data are no longer portable to other platforms. Overnight, innovation and healthy competition is killed. Consumer choice is legislatively dismantled. And AWS will be free to expand their near monopoly on the cloud.
This harks back to Compaq's closing of the IBM PC chipset in the early 80's - a move that arguable democratized computing for the masses by lowering the cost and improving the access to previously unreachable and uber-expensive computers.
APIs and open frameworks for sharing information and control (CRUD) across platforms is going to be THE MOST IMPORTANT single factor enabling a connected life that crosses all physical and virtual boundaries, from our work and home, to our mobile life and even the cars we drive.
These thoughts are stewing in my head, and if I get time, I may write them down, but for now... PLEASE DO NOT LET THEM MAKE APIs PROPRIETARY.
Hey http://socialcam.com/ people... better fix your app before people get turned off and never use you again.





Recent Comments