With infrastructure services like Amazon EC2, Rackspace, and VMware making it easy to take advantage of the flexibility, portability, and reduced costs of cloud computing, it seems obvious to jump on the cloud bandwagon for new IT projects.
But, developers are generally left on their own to deal with the pain of deploying their apps to the cloud: configuring application servers, libraries, disk partitions, networking, clustering, service connections, and virtual private networks. After they get their app installed they also need to install management agents that run on top of the application layer.
If you really want to take advantage of the cloud and optimize return on investment, you'll want the on-boarding process to be easy and fast and you won't install that agent. Agent-based solutions are inherently inflexible. Deploying agent-based solutions in a cloud-based environment, which is, by definition, highly flexible, is often like trying to fit a square peg in a round hole. In agent-based solutions, hard-coded agents are installed on every machine to monitor the application. If a change to the application configuration occurs--such as the IT department adds a node or upgrades a component--the agents must be updated as well.
Each agent and management server must be configured separately with management and monitoring solutions generally not portable. When every change to an environment requires installation of multiple agents on each server and configuration of multiple management servers, it becomes a tall order to move an application from a traditional infrastructure to the cloud, or from one cloud infrastructure to another: private to public, public to hybrid, or hybrid to private.
How do you get around this so you can actually capitalize on the benefits of cloud computing? Go virtual. Move application management, including easy on-boarding, from above the application stack into the underlying virtual layer, along with the rest of the cloud infrastructure.
I was recently briefed by webappVM CEO Isaac Roth on how the company is pioneering this new approach. He said the virtual path allows you to actually realize all of the flexibility, portability, and reduced costs that come with the promise of cloud computing.
... Read more
Amazon Associates is the partner program the company uses as part of its affiliate advertising programs, allowing customers to make money advertising Amazon products.
Associates can now simply click a link in the toolbar to send a link (replete with sales-y text) to Twitter as part of their shopping and selling experience. Amazon gets a sale, Twitter gets traffic, and the associate gets revenue share. What could possibly go wrong?
Linking to Amazon or other online retailers is obviously nothing new, though Amazon has been particularly successful in using its link networks for both sales and to garner higher Google rankings for organic advertising.
This new program does introduce an issue related to link fraud, where spammers and scammers leverage URL-shortening services for spam links. Currently there is no way to verify that the link you click actually goes to Amazon. It's a bit surprising that it decided to use an URL-shortener that it doesn't own, though I suppose the network effect of the URLs helps perpetuate the life of the links.
There is also a risk of nondisclosure wherein in Twitter users attempt to push products that offer some kind of gain to them that they don't clearly state to you. While I understand the argument for disclosure on blogs and media in general, Twitter remains a playground for people to post whatever they want. I highly doubt all the celebrities with accounts would bother wasting their precious time if they weren't posting for their own gain.
Interestingly, there is no mention of whether Twitter is an Amazon Associate, suggesting that Twitter won't see any of the revenue share. I'd like to think that they cut a deal that gives them a piece of the pie, but to date we haven't seen Twitter monetize itself too effectively.
Twitter is quickly becoming the flash news vehicle for everything from news alerts to product placement. And based on a very quick review of my Bitly account, Twitter users just love to click on links. But, I still have to wonder if Twitter will ever get beyond its current role as a marketing tool?
If recent research is any indication, Amazon.com and Google are winning the cloud game.
Evans Data on Tuesday released a report (registration required) on how developers perceive cloud service providers related to cloud services offerings, including their completeness and the companies' ability to execute on the vision.
Janel Garvin, the founder of Evans Data and the author of the report, provides excellent insight into the current state of the market and how quickly things could change, if certain large vendors (notably AT&T and Microsoft) got their acts together more quickly.
Given their robust services, it isn't surprising that Amazon and Google top the list. And although IBM, VMware, and Microsoft trail, each offers important components of cloud infrastructure.
... Read moreAmazon Payments today launched a new service that brings the company's payment processing tools to mobile devices. Amazon Mobile Payments Service (MPS) includes a set of APIs (application programming interfaces) that allow mobile developers and merchants to provide payment options to their customers within mobile Web sites and applications--including the convenience of Amazon's 1-Click checkout system.
There are already a number of mobile payment providers, but Amazon is the big dog of the e-commerce world with an enormous amount of customer accounts already in use. This could be an excellent option for companies that offer mass-market mobile applications and are looking for ways to easily accept payments.
The service will automatically detect the request origin, meaning a Web or mobile browser, or a mobile application so that developers don't need to re-work their applications.
... Read moreArguments for and against the cloud are starting to calm down a bit, and most people agree that the cloud is somewhere in your future, if not in your present.
Instead of arguing semantics of application development and delivery, the discussion should really be around how to deal with a mix of on-premise and on-demand, a combination that is unlikely to change in the foreseeable future.
I spent the first half of this week in Las Vegas at a nontech trade show, and missed both VMworld and the Red Hat Summit. However, watching and reading from afar, I noticed two major themes in discussion around both cloud computing and virtualization: cloud interoperability and the lack of application management tools.
Cloud interoperability--the ability to abstract the programmatic differences from one cloud to another--is a key to adoption. If we assume that some percentage of private compute clouds will be based on virtualization, and we know that a large percentage of public clouds already are, then the ability to move among virtual machines is a critical function in this regard.
Red Hat is obviously taking interoperability seriously, with Thursday's launch of Deltacloud, a new open-source project "designed to enable an ecosystem of developers, tools, scripts, and applications that can interoperate across the public and private clouds."
Let's remember that right now, there is a difference between managing applications that are in your own data center and managing those at a cloud provider. Missing here are new management tools that cross borders in a seamless manner and don't discriminate against different hypervisors or application platforms.
Cloud application management isn't so much about workloads as it is the ability to move applications and associated data from cloud to cloud and system to system with no interference. This new realm of internal-external systems management opens up a world of opportunities but faces some significant speed bumps.
I noted last week that Amazon's announcement of virtual private clouds presents a challenge for many cloud-oriented start-ups. The issue is that Amazon calls the shots on the cloud and VMware on virtualization. And while both companies have done fairly well by their users (let's say better than we would expect from Microsoft or Oracle), innovation is stuck within their respective ways of doing things.
Regardless, there is a cloud management opportunity, with open-source projects like Puppet, as well as Red Hat's new release of Network Satellite 5.3. While neither is cloud-specific, applications that support large-scale infrastructure management are perhaps the first step in harnessing the computing power inside and attached to your data center.
Arguments for and against the cloud are starting to calm down a bit--and most agree that the cloud is somewhere in your future. The discussion should really be around how to deal with a mix of on-premise and on-demand, a combination that is unlikely to change in the foreseeable future.
Follow me on Twitter @daveofdoom.
A huge amount of digital ink has been spilled trying to define "the cloud" and "cloud computing" and now Amazon Web Services has once again upped the ante with its latest Virtual Private Cloud (VPC) services.
The VPC, outlined here by fellow CNET blogger James Urquhart, provides a way for companies to create a logically separated set of Elastic Compute Cloud (EC2) instances and a secure VPN connection to their own networks. Effectively, it takes a chunk of Amazon's services and makes it private. Still on the Internet and still on shared hardware, but private.
The VPC use case is for enterprises to use cloud services outside their firewall, connected in a secure manner. This is very compelling to IT shops that need more resources but have to date been concerned about security. And while this approach can certainly be called a "private cloud," I believe the more likely private use case is that of an "enterprise cloud" that is behind the firewall and adheres to compute cloud principles (elasticity, seamless scaling, etc.).
Where the computational resources live seems to be the crux of the private vs. public cloud discussion, and now that Amazon has defined private clouds as part of an Internet-hosted infrastructure I expect we'll see more references to "enterprise clouds" as the software matures.
The enterprise cloud is really just an actualized version of the "compute cloud" concept that we've seen for the last 10 years (and I've written about repeatedly) with a deployment model that mirrors Internet-based cloud services.
A "compute cloud" is a different animal, according to the developers of Eucalyptus, an open-source, EC2-compatible infrastructure-as-a-service. Typically based on virtual machines, "cloud computing allows users to dynamically provision processing time and storage space from a ubiquitous 'cloud' of computational resources."
Incidentally, Amazon's new offering not only resets the semantics of how we talk about cloud computing but also puts serious pressure on other cloud providers to offer feature parity. Few other providers offer the full breadth of AWS services and those that do are constantly playing catch-up.
Another aspect of the new offering is the impact it has on a variety of start-ups that have attempted to do similar things, or to augment AWS in ways that Amazon hasn't yet tried. As with so many other platforms, the risk of building on top of something you don't own (i.e. EC2) is significant. And while AWS continues to be very innovative, start-ups are going to have to be very agile to outfox Amazon.
Follow me on Twitter @daveofdoom.
According to a new report by researchers in Australia, stress tests have revealed that the "infrastructure-on-demand services offered by Amazon, Google and Microsoft suffer from regular performance and availability issues."
The seven-month study of Amazon's EC2, Google's App Engine, and Microsoft's Azure cloud computing services simulated 2,000 concurrent users connecting to services from each of the three providers, with researchers measuring response times and other performance indicators.
The results were at best mixed, and at worst, severely dysfunctional. For example, I'd never heard that when using Google App Engine, none of your data-processing tasks can last longer than 30 seconds, lest the service throw an exception back at you.
Researchers found that the three platforms "delivered wildly variable performance results as Amazon, Google and Microsoft trialled, added and dropped new features."
... Read more
One of the big takeaways from the RFQ is the reminder of how immature the market and usage of the cloud really is. The GSA is basically asking for hosting, with the ability to use APIs that manipulate data and services. They do have the right vision for how the cloud can change infrastructure, but they aren't quite there yet.
While there are a growing number of cloud service providers, few can meet the full set of requirements the RFQ outlines. For example bothAmazon Web Services and GoGrid claim to meet the GSA's 99.95 percent service up-time requirement, but do so via service credits if they experience downtime.
The obvious vendor candidates like IBM, EDS, and Sun might be able to meet the networking and infrastructure requirements, but don't have offerings that address cloud storage and interaction functions--essentially the APIs that cloud services rely on.
Generally speaking, most data center providers don't have the ability to address the functional specifications for cloud functions--i.e. PUT, GET, DELETE, etc. that move and manage data in a scalable manner, one of the main promises of the cloud.
... Read moreAs the world pushes ahead with cloud computing and business users demand software-as-a-service (SaaS) applications, many IT departments are struggling to keep legacy applications on life support. Many of these zombie applications are there only for storage and audit purposes, not for real-time data interaction.
Even if applications have been "turned off" the data continues to live on in databases and file stores, continuing to take up storage space and software licenses. The result is a state of paralysis, with application retirement merely a dream.
U.K.-based Clearpace recently unveiled a cloud-based data archiving service called RainStor. RainStor's technology is being used to solve a completely overlooked problem domain: application retirement. I spoke via e-mail with RainStor CTO Andy Ben-Dyke to understand how the service works and why it makes sense.
RainStor's Instant Application Retirement service works in 3 steps:
1. Send--Structured data from any RDBMS is automatically compressed by 40x or more, encrypted and sent to the cloud using a client-side software appliance. The extreme compression that is applied significantly reduces the time to transfer large volumes of data to the cloud.
2. Store--The encrypted data is stored in a private archive on Amazon's highly available and secure storage cloud (S3). Though compressed, the original schema format is preserved and RainStor is able to layer on additional archives which reflect any schema changes (e.g. add or delete of columns).
3. Search--Running on Amazon's highly scalable compute cloud (EC2) RainStor allows you to query data through any industry-standard reporting or BI tools over ODBC or JDBC with lightning speed. Providing "point-in-time" query capability based on its ability to store schema evolution changes.
The RainStor service can be had for as little as $1 per GB of data stored per month with no commitments, including Amazon storage and resource costs. Clearpace is also offering a 90 day free trial.
Given that there is a untold fortune of hardware and software tied up in legacy apps waiting to freed up, turning off those apps and sending the data to "heaven" in the clouds just seems like a no-brainer.
Follow me on Twitter @daveofdoom
Amazon.com's legal team is "investigating" open-sourcing their various Web services API's including those for EC2, and S3, Amazon's main cloud computing interfaces, according to Enomaly founder Reuven Cohen.
Amazon's APIs are already the de-facto standard used by thousands of Amazon Web Services (AWS) customers and arguably the best designed method of interaction for various cloud services. By releasing the APIs as open source (creative commons, or no-sue covenant or whatever), Amazon cements its way of doing things into the greater marketplace, while opening up an even broader ecosystem of applications to interact with AWS.
If other companies, such as Eucalyptus can be assured that they won't fall on the wrong side of the legal fence, it only ensures that they will continue to grow their usage of the Amazon systems and methods.
This would be a win for everyone interested in cloud computing. Let's hope it's true.
Follow me on Twitter @daveofdoom
