A successful dinner party needs two things: food and seating. In our first scenario, you are hosting this dinner party at your home. Because you are responsible for purchasing enough food and having enough seats on hand, you’re going to have to make some important decisions and purchases well in advance. To do this, you’ll have to estimate to the best of your ability.

One forecasting strategy is to rely on RSVPs. Using that data, you are able to determine an approximate amount of guests attending your party, which can help determine the amount of resources needed (food and chairs) for a successful evening.

While this strategy would work in a perfect world, I have found it to be unreliable. I often have to plan for excess, having extra chairs and food for the attendees. This results in waste because I have cooked extra food for my party. However, the opposite could happen; I could have an influx of guests who didn’t RSVP and I could not have enough seating or dinner. This scenario is analogous to dedicated hosting in the sense that you have to use a lot of estimation to plan for the traffic to your server. Conversely, if you don’t have a lot of “guests” show up, you could have spent an excess of money on your hosting solution by having all that gear up.

For our second scenario, let’s look at hosting your dinner party at a restaurant; this is analogous to cloud hosting. While you would still want to get guests to RSVP to get a general idea of the attendance, having an exact number isn’t as crucial. There is a fully staffed kitchen that can cook to order for the partygoers, so you don’t have to worry about how much food is needed. Furthermore, instead of serving an average amount of food, guests are able to order different plates and quantities of food directly off the menu, satisfying their exact appetite.

Not only can the restaurant’s kitchen produce food for extra guests, the same goes with ensuring that they have a place to sit as the restaurant has many chairs. This on-demand support of the restaurant is equivalent to the resources of the cloud that can be deployed at a moment’s notice to handle the exact traffic to your site. This is one of the true powers of the cloud – the ability to scale for your given level of Internet traffic, while ensuring that you pay just for the resources needed to serve up that traffic.

Developers who already had an app in the cloud were very interested about how they could scale it. The key for doing this is having an application with modular design, where you can split different functions (web, app and database servers) across multiple nodes. Furthermore, having the app integrated with a hosting provider’s API is a way that it can programmatically make changes to the infrastructure, like creating new servers. This takes a certain amount of planning as the app is being built, but is the best way to leverage the true power of the cloud.

For developers who were bringing their application to the cloud for the first time, it was more about their initial deployment rather than scale. They wanted to figure out how to get their application to deploy in a way that was going to be as agile as possible and gain them as many options for growth and iteration. I talked to them about the first five things that they should do when starting a cloud project, but also was able to reiterate those same tenets of modular design and utilization of the API as I discussed with devs who were interested just in scaling.

The final thing that most developers talked about was feature requests. They were very clear about letting us know what we could do to make their lives easier by adding in certain features and functions. This was a great opportunity because we got to show them that some of these features exist in our architecture. In other cases, we were able to show them how to use a combination of existing features in order to get that same functionality. For those things that didn’t exist in the Rackspace Cloud ecosystem, we were able to bring back a lot of great feedback to our product team.

After visiting with all these developers, the one thing that I learned was that we get awesome results when we combine the expertise of our customers who host on the Rackspace Cloud with Rackspace architects. I would encourage more customers to call in and talk to us about how to host their application on our infrastructure.

What are mobile-ready cloud stacks?

These stacks will allow developers to start building mobile apps on pre-configured cloud backend. This will save deployment time by not having to reinvent the wheel every time they start a mobile project

The stacks also provide scalable reference architectures based on our experience as a backend provider of complex applications.

Finally, we want to provide a developer experience that is beyond just deployment and includes day-to-day management and operations of the backend including upgrading, scaling, patching, monitoring and troubleshooting the cloud services infrastructure where your app is hosted.

What’s included?

The first pre-installed stack available today is a PHP-based backend, comprising LAMP and open source caching software. The exact configurations are:

• Linux (Ubuntu)
• Apache web server
• MySQL
• PHP
• Varnishd  (HTTP accelerator)
• Memcache and the PHP-memcache extensions
• The Alternative PHP Cache (APC)

We are working on a similar offering for the Ruby on Rails and Node.js backends

How do I get it?

To start using this stack as the backend to your mobile app, follow these steps:

1. Sign up for a Managed Cloud account at: https://cart.rackspace.com/cloud/
2. You will receive a call from our Launch Specialists. Tell them that you are interested in our mobile stacks, or reach out to our team at (800) 480-3163. They will stand up the backend for your app in minutes, so you can start coding!

Resources for more information:

Mobile cloud page: http://www.rackspace.com/mobile-cloud

Mobile Blog: http://www.rackspace.com/blog/mobile

In some instances you might need to scale your configuration extremely large, exceeding your wildest dreams. While the cloud can facilitate this growth, sometimes developers could begin to see build failures or receive errors as they try to grow so dramatically. It’s possible that they hit an API limit.

What’s an API limit? One of our jobs at Rackspace is to ensure that all of our customers are protected in a shared environment. This means that all customers have the right performance and stability to host their sites and applications. To help do this, we want to make sure that we don’t have too many customers dramatically scaling up to a very large configuration at any given point in time. This API limit helps us do this.

Any time you plan to have a large event, always call your Fanatical Support Team to let them know about it. They can then make allowances for what your API limit should be set to in order to supply you with all of the requisite resources that you need to drive your application. Furthermore, they have tons of cloud experience and can also give you some other advice to make sure that this event goes flawlessly.

As you begin developing in the cloud, it is common to start with just a single Cloud Server. However, over time as you get more visitors to your site or using your application, you will need to add more resources. In the cloud, you always want to scale your environment horizontally, which means adding more nodes to your configuration. To route traffic to each of these servers, you will need to make sure that you have a Cloud Load Balancer in front of the configuration.

If your website or application has to store all the information from transactions occurring on your site, you’ll need to determine what you are going to do about your database. Cloud Databases can connect to your web servers to help store this transaction data.

Similar to a home computer’s hard drive, there is a certain amount of cloud storage associated with each Cloud Server. As you begin to fill that storage space, you have a couple options to expand your storage to keep your files and media. Cloud Block Storage is a way to bolt on additional storage to your Cloud Server, allowing you to increase your storage capacity without having to increase your compute power.

Another way to increase your data storage is to use Cloud Files, which gives you several different advantages. First, this is an excellent place to store your backups (which you should always be taking) because it is a location that is off of your server. Second, by storing frequently accessed static content in Cloud Files, you can take advantage of the Content Delivery Network (CDN). The CDN is a group of geographically dispersed servers that delivers your static content to the end user from an endpoint that is physically closer to them, resulting in faster times to serve the content.

Finally, we have an option that is a big differentiator to round out the complete Rackspace Cloud solution, and allows you to connect a dedicated server configuration to a cloud configuration: RackConnect. Whether you are growing and would like to have a dedicated server presence, or you must have dedicated servers due to compliance requirements for your particular business, RackConnect is a way to have that layer of your configuration on dedicated gear while taking advantage of the elasticity of the cloud.

The best part of the Rackspace Cloud ecosystem is that it is all built on top of our world class Fanatical Support. That’s what Rackspace is all about; we are here support you and your company so you can focus on your business. You can see that the cloud is truly a complete ecosystem. With some of the products we have here powered by OpenStack, we hope to provide our customers the power to build complete solutions for their site or application.

Check out Joseph’s previous post and video where he talked about the power of the Rackspace Cloud API. Find out more about the API documentation at docs.rackspace.com and check out the Rackspace DevOps Blog at devops.rackspace.com.

So what exactly can you do with the Cloud API? Anything you can dream up. By stringing together commands, you can have your application dynamically modify your cloud environment in many different ways. Here are three of the most popular things that customers do with our API.

1. Auto Scale Your Configuration

By integrating your application with a monitoring platform, when traffic to your site changes, you can call the Cloud API to grow or shrink your configuration. Instead of attempting to predict the traffic demands, your application can actually become responsive to the traffic on your site. You can do this with Cloud Servers, Cloud Databases, Cloud Load Balancers or any product in the OpenStack suite.

2. Create New Cloud Environments

The API is effective at creating new environments in the cloud. This could be used when you want to create new environments for a side project or a development sandbox. However, a broader use of this function could be to create new server environments for customers that signup for your SaaS application. For security purposes, you may want them sandboxed in their own environment. Your application can do this by simply making an API call, eliminating the need for you to manually provision out a new environment for each new customer.

3. Integrate With Partner Solutions

Our Cloud API gives you a complete connection to a variety of solutions that have already been created. Many of our Cloud Tools partners can integrate with your configuration, as well as other cloud vendors, via our API. There is a diverse set of solutions, so it may be worth investigating what tools have already been developed in the cloud ecosystem before programing it on your own. This can save you, and your business, a lot of time.

These are the top three uses that I see for the Cloud API. How have you been using it with your application and configuration? Be sure to let me know in the comments below.

Check out Joseph’s previous post and video where he discussed developing on the cloud. Next week, he looks at how Rackspace has a complete cloud solution. Find out more about the API documentation at docs.rackspace.com and check out the Rackspace DevOps Blog at devops.rackspace.com.

The API, fundamentally, allows for code to perform basic functions within your environment, such as creating, deleting or changing virtual machines. In the hands of skilled developers, the API becomes the connective tissue that allows our cloud environment to react intelligently based on your code and the code of Cloud Tools partners. As variables and traffic change in your environment, your code can control you cloud configuration in a moment’s notice.

We are living in a new age of hosting. Instead of thinking of the physical hardware that your code sits on, your cloud architecture and configuration can be boiled down to lines of code. Today, you can actually save your configuration specifications and deploy them in an instant. Developers understand the importance of good, clean design; now, they can take those design principles and apply them to their architecture.

Since code can now control the environment that it runs in, it makes sense to create a self-healing environment. If something goes wrong, using a single call via the API to our platform can completely change how your configuration is working. This allows your application to automatically manage the resources to help ensure that you have enough compute power to handle your traffic.

Success in the cloud is about coding to scale. Putting your code over multiple nodes is an excellent way to make sure that there’s not a single point of failure, ensuring the performance and predictability that you are looking for. However, this takes practice. In order to code to scale and not code to fail, think about having a development sandbox made up of multiple nodes to test and develop your code. Doing so is a way to guarantee that your application, or site, is able to run over a distributed architecture, an architecture that makes the cloud incredibly powerful.

Check out Joseph’s previous post and video where he discussed how to prepare your site for a blitz of online traffic. Next week, he discusses the power of the Rackspace Cloud API. Find out more about the API documentation at docs.rackspace.com and check out the Rackspace DevOps Blog at devops.rackspace.com.

Whether it is the big game or another high traffic event, you can take steps to help ensure your website can survive a crushing amount of traffic. After all, no one likes going to a website that is down. Here are some tips to keep your site up and running in the cloud through a huge traffic spike.

1. Cache Content On Your Site

Most content management systems, such as WordPress, render a unique page for every visitor to your site. Rather than having to build a page a million times, caching lets you build it once and save it, taking a significant load off of your configuration.

There are different strategies for caching content whether you are running a Windows or Linux server. Additionally, plugins on popular CMS platforms can help you cache as well. The point to takeaway is that it is always more efficient to serve cached content instead of having to go to the database each time, and Rackspace can help suggest strategies to best suite your needs.

2. Set Up A Static Landing Page

If millions of people will be hitting your site at the same time, one way to make sure that they see something is to have a static landing page. By dynamically generating content for your homepage, you run into the situation of a server getting overwhelmed. However, if you take advantage of a static homepage and upload it to Cloud Files, it will be distributed over our Content Delivery Network (CDN). This results in your content being distributed on servers around the globe, allowing it to be served up to your visitor on a server that is physically closer to them and resulting in faster load times.

3. Know The Different Types Of Scaling

One of the amazing aspects of the cloud is how it can scale. But to prepare for the onslaught of traffic, you need to know the difference between vertical and horizontal scaling. Think of vertical scaling as simply clicking “resize server” in our control panel and selecting the next size up. This increases the allocated amount of virtual CPU, RAM and disk. The benefit is that it can be done with a click of a mouse, but the drawback is that there is downtime while the server resizes, something you probably don’t want during your high traffic event.

The other type, horizontal scaling, helps mitigate this by essentially adding additional servers to your configuration. By doing this, you increase the available pool of resources to handle your high traffic event, without having downtime to add another server to your line. One thing to consider is that you need to prepare your application to take advantage of this type of scaling in advance.

4. Have A Load Balancer Out Front

If you do plan on distributing your load across multiple servers, or plan to take advantage of horizontal scaling, you will want to set up a load balancer to help distribute the traffic to different nodes. You will want to do this in advance, because you will have to point the IP to the load balancer, which could result in some downtime while the DNS records update. The upside to a load balancer is huge, and you will definitely need one to help route the high volume of traffic.

5. Database Replication

When you have a high traffic event, there are going to be lots of reads from your database. If that database was to topple over, you risk losing functionality to your entire site. To prevent this from happening, you will want to consider a database replication strategy.

To do this, you have one “master” database and a pool of “slave” databases. All the write requests are sent to the master database and are then replicated out to the slave databases. The read requests have the ability to happen on any of the database nodes in your configuration. With database replication, if your master database begins getting overtaxed, you will want to promote a slave to become a new master. There are several Cloud Tools partners that can help you and your company with database replication, as well as the Rackspace Managed Databases team.

Those tips should help your website defend against the rush of traffic during a major event, like the Super Bowl, and enable you to sit back and enjoy the game instead of worrying about your site being sacked by a traffic blitz.

Want to learn more from the Cloud Zen Master? Check out Joseph Palumbo’s author page for all of his posts!

Most people think about their cloud configuration having a load balancer up front with the web and database servers below it. Sometimes a spike of traffic might not let it perform the way you want it to. One way to mitigate this risk is to have a static home page served off the CDN.

Doing this leverages all the geographically dispersed servers across the globe, to make sure that when someone comes to your website, they are going to get that web presence of your landing page. Now, this page is not going to be dynamic, but your customers will be able to click through to access your shopping cart or application. Having a landing page on the CDN is a safe guard to ensure that you always have a landing page for your visitors to access.

Interested in finding out more to help you prep for the holidays? Check out Joseph’s white paper now. You can also view his previous post where he gives tips on load balancing for the holidays and check out his next post where he talks about how to prep your site for a high traffic event.

The holidays seem to have the same effect on our server configuration as it does with our households. For example, for most of the year my kitchen table and chairs work to serve dinner to my family. However, when the holidays come and my house is jam packed with guests, we have to set up additional tables and chairs (some in the living room) to accommodate all the guests. Furthermore, you want to make sure certain family members sit together and that the kids have their own table. The load balancer does something similar for your server configuration.

As traffic flows into your site, the load balancer is really good at making sure that it goes to the right place. For example, if you have two web servers, the load balancer will ensure that one of them doesn’t get overwhelmed with traffic. It will route the traffic to the appropriate web server to make sure that the visitor gets the content as fast as possible. Much like separating the kids out to their own table, you can also set up rules to route different types of traffic to different servers.

Interested in finding out more to help you prep for the holidays? Check out Joseph’s white paper now. You can also view his previous post where he gives tips on tuning and testing for the holidays. In his next (and final) post, he talks about serving a static homepage off of the Content Delivery Network (CDN).