Category: Compute*


Next Generation Cluster Computing on Amazon EC2 – The CC2 Instance Type

 

You no longer need to build your own compute cluster in order to tackle your High Performance Computing (HPC) projects. By launching cloud-based compute instances on an as-needed basis, you can avoid waiting in lengthy queues for limited access to shared resources.

We’ve pushed the bounds of cloud-based HPC in the past with the introduction of our Cluster Compute and Cluster GPU instances. Both of these instance types have been used in a wide variety of High Performance Computing scenarios.

Today we are introducing a new member of the Cluster Compute Family, the Cluster Compute Eight Extra Large. The API name of this instance is cc2.8xlarge so we’ve taken to calling it the CC2 for short. This instance features some incredible specifications at a remarkably low price. Let’s take a look at the specs:

Processing – The CC2 instance type includes 2 Intel Xeon processors, each with 8 hardware cores. We’ve enabled Hyper-Threading, allowing each core to process a pair of instruction streams in parallel. Net-net, there are 32 hardware execution threads and you can expect 88 EC2 Compute Units (ECU’s) from this 64-bit instance type. That’s nearly 90x the rating of the original EC2 small instance, and almost 3x the rating of the first-generation Cluster Compute instance.

Storage – On the storage front, the CC2 instance type is packed with 60.5 GB of RAM and 3.37 TB of instance storage.

Networking – As a member of our Cluster Compute family, this instance is connected to a 10 Gigabit network and offers low latency connectivity with full bisection bandwidth to other CC2 instances within a Placement Group. You can create a Placement Group using the AWS Management Console:

Pricing – You can launch an On-Demand CC2 instance for just $2.40 per hour. You can buy Reserved Instances, and you can also bid for CC2 time on the EC2 Spot Market. We have also lowered the price of the existing CC1 instances to $1.30 per hour.

You have the flexibility to choose the pricing model that works for you based on your application, your budget, your deadlines, and your ability to utilize the instances. We believe that the price-performance of this new instance type, combined with the number of ways that you can choose to acquire it, will result in a compelling value for scientists, engineers, and researchers.

Operating Systems – This instance type uses hardware-assisted virtualization (HVM), so you’ll need to choose an AMI accordingly. You can use the Amazon Linux AMI or Windows 2008 R2. You can also install HPC Pack 2008 R2 Express (read Microsoft’s HPC Server FAQ for more info).

We have updated the Amazon EC2: Microsoft Windows Guide with instructions on setting up an HPC cluster complete with an Active Directory Domain Controller, a DNS server, a Head Node and one or more Compute Nodes.

Speed – We have submitted benchmark results for HPL to the Top500 site. The November list came out earlier today and we are ranked at position number 42, with a speed of 240.09 teraFLOPS. This result was obtained using a cluster of 1064 instances.

On a somewhat smaller scale, you can launch your own array of 290 CC2 instances and create a Top500 supercomputer (63.7 teraFLOPS) at a cost of less than $1000 per hour (perhaps a lot less, depending on conditions in the Spot Market).

Launch – My colleague Dr. Matt Wood cooked up a CloudFormation template to make it easy for you to get started with CC2 instances. The template uses MIT’s StarCluster to create a fully functioning cluster for loosely coupled or tightly parallel compute tasks with a single click. Matt says that the template will do the following:

  • Provision a new 2 node CC2 cluster with 32 hyperthreaded cores, into a new placement group.
  • Attach NFS storage, monitoring, and a 200 GB AWS Public Data Set.

The template creates a new t1.micro instance, which acts as a controller for the rest of the elastic cluster. From a basic Amazon Linux AMI, CloudFormation bootstraps all dependencies, installs and configures StarCluster and creates the necessary security credentials before provisioning the CC2 instances which spin up ready to accept jobs via Sun Grid Engine.

You can spin up the stack, log in to the controller instance and hop onto the cluster master to submit jobs, or scale the ad-hoc cluster up and down in just a few clicks. That’s pretty cool for just 169 lines of declarative JSON. You can get started very quickly with this friendly button:

We are making the CC2 instance available as a public beta so a few caveats apply:

  • The instances are available in a single Availability Zone in the US East (Northern Virginia) Region. We plan to add capacity in other EC2 Regions throughout 2012. Please feel free to contact us if you are interested in CC2 support in other Regions.
  • You can run 2 CC2 instances by default. If you would like to run larger jobs, please submit an EC2 instance increase request.
  • You cannot currently launch instances of this type within a Virtual Private Cloud (VPC).

Supercomputing 11 – The AWS team will be out in force at the Supercomputing 11 conference (November 12-18 in Seattle); here’s a summary of AWS activities at SC11. Our booth (#6202) will be open from the 14th to the 17th (come by and say hello). My colleague Dr. Deepak Singh, will participate in a panel on SaaS-Based Research. Deepak will host a BoF session on HPC in the Cloud ; Matt Wood will host a session on Genomics in the Cloud.

What can you do with a supercomputer of your very own?

— Jeff;

Now Open – US West (Oregon) Region

We have just opened up another AWS Region!

This one is located in the Pacific Northwest area of the United States, in the beautiful state of Oregon, and offers low-cost, low-latency access to our services from the Western portion of the US. Were always trying to provide you with services at the lowest possible price. Pricing for this new region is currently the same as for the US East (Northern Virginia) region.

The new US West (Oregon) Region supports the following services:

This is our seventh Region, and our fourth in North America. You can see the full list in the Region menu of the AWS Management Console:

You can launch EC2 instances or store data in the new Region by simply making the appropriate selection from the menu.

— Jeff;

PS – There are really four US Regions even though the menu shows just three. The fourth region is the ITAR-Compliant AWS GovCloud (US) Region, which I wrote about earlier this year.

New – Use Your Microsoft BizSpark Licenses on AWS

It is a great time to be a software developer with an entrepreneurial mindset. You can get access to development tools, startup resources, and much more at no charge or at a very low cost these days. You don’t need to get venture capital or max out your credit cards in order to get started.

Microsoft, an AWS Solution Provider, offers the Microsoft BizSpark program to help software startups succeed by giving them access to Microsoft software development tools, connecting them with key industry players, including investors, and providing marketing visibility to help entrepreneurs who are starting a business. As part of this offer, BizSpark offers a select list of Microsoft licenses, including Microsoft Windows Server 2008 R2 and Microsoft SQL Server 2008 R2. 

We are pleased to announce that you can now import your BizSpark licenses for Windows Server and SQL Server products to AWS and use them to launch Elastic Compute Cloud (EC2) instances. You can then run these products on EC2 instances at an hourly cost that has been adjusted to reflect the removal of the license charges. In other words, Windows instances will be billed at the same rate as Linux/Unix under this plan.

If you are an existing BizSpark member, click here and enter your Microsoft BizSpark Subscriber ID. You can also learn more about Using BizSpark Licenses on AWS.

Be sure to check out the AWS Toolkit for Visual Studio (read my review here).

— Jeff;

IT-Lifeline – Disaster Recovery Using AWS

Matt Gerber,CEO of managed service provider IT-Lifeline spent some time talking about AWS on camera. Here’s the full video:

Matt enumerates the three main benefits that AWS provides him: cost, elasticity, and time to market (sometimes saving 6 months to a year for a new offering). He also talks about security and compliance, and about working with the AWS team. Matt notes that traditional disaster recovery (DR) is a capital-intensive business, and that they’ll be able to pass savings of 25-50% on to their customers. He says that the carrying cost for disaster recovery is effectively reduced to zero — “It doesn’t cost us anything to know that that AWS infrastructure is there for when our customers need it for a disaster.”

We’ve pulled together some additional DR and storage resources on our new Backup and Storage page.

— Jeff;

Run Microsoft SQL Server 2012 (\Denali\) on AWS Now

Microsoft SQL Server 2012 (code named “Denali”) has a lot of really intriguing features including high availability, columnstore indexes, self-service Business Intelligence, faster full-text search, data visualization, and improved security and compliance. Consult the feature guide for more information.

In order to make it as easy as possible for you to test your applications with this new database, we’ve worked with Microsoft to create a new EC2 AMI that is preloaded with Windows Server 2008 R2 and the Community Technology Preview 3 (CTP3) of SQL Server 2012. The new AMI is available in all five of the public AWS regions. The AMI IDs are listed here; you can also find them using the AWS Management Console:

Instead of locating some suitable hardware and spending your time downloading and installing nearly 3 GB of code, you can simply launch the AMI and start exploring within five minutes. You will pay only for the EC2 instances that you launch and run; there are no additional software licensing costs.

To get started, visit the Denali Test Drive page. If you are new to AWS and are looking for some step-by-step directions, I recommend our new Microsoft Windows Guide.

— Jeff;

 

Launch EC2 Spot Instances in a Virtual Private Cloud

Over the past two months, we have had the opportunity to share several exciting developments regarding Spot Instances. We have told you how to Run Amazon Elastic MapReduce on EC2 Spot Instances, we published Four New Amazon EC2 Spot Instance Videos, and we outlined the excitement around Scientific Computing with EC2 Spot Instances. Others in the community have also shared their experiences with Spot Instances. You may have read about Cycle Computing running a 30,000 core molecular modeling workload on Spot for $1279/hour or Harvard Medical School moving some of their workload to Spot after a day of engineering, saving roughly 50% in cost.

In typical Amazon fashion, we like to keep the momentum going. We’ve combined two popular AWS features, Amazon EC2 Spot Instances and the Amazon Virtual Private Cloud (Amazon VPC). You can now create a private, isolated section of the AWS cloud and make requests for Spot Instances to be launched within it. With this new feature, you get the flexibility and cost benefits of Spot Instances along with the control and advanced security options of the VPC.

Based on feedback from customers in the community, we believe this feature will be ideal for use cases like scientific computing, financial services, media encoding, and “big data.” As an example, we have received a number of requests from members the scientific community who have been mining petabytes of confidential (e.g. human genome or sensitive customer data) and/or proprietary data (e.g. patentable). Traditionally they have set up their own software VPN connection and launch Spot Instances. Now they can leverage all of the security and flexibility benefits associated with VPC.

We have also heard a number of customers looking for ways to integrate on-premise and cloud solutions, and to burst into the cloud. These customers now can leverage VPC and Spot for a great low cost solution to this “computation gap.”

Launching Spot instances into an Amazon VPC is similar to launching Spot instances, but you need to specify the VPC you would like to run your Spot Instances within. Launching Spot Instances into Amazon VPC requires special capacity behind the scenes, which means that the Spot Price for Spot Instances in an Amazon VPC will float independently from those launched outside of Amazon VPC.

The AWS Management Console includes complete support for this new feature combo. You can examine the spot price history for EC2 instances launched within a VPC:

You can use the console’s Request Instances Wizard to make a request to launch Spot Instances in any of your VPCs at the maximum price of your choice (just be sure to choose VPC for the Launch Into option):

For more information on using Spot Instances in VPC, please visit the EC2 User’s Guide.

If you want to learn more about the ins and outs of Spot Instances, I recommend that you spend a few minutes watching the following videos:

Getting Started With Spot Instances

Deciding on Your Spot Bidding Strategy

How to Manage Spot Instance Interruption

There is also a video coming soon on the how to launch a Spot Instance within VPC, so check back at the Spot Instance web page again soon. I will tweet when it becomes available (please follow me (@jeffbarr) for more details).

As I mentioned, the Spot service has been rapidly evolving, and we would love to get your feedback on the next features youd love to see. Please feel free to email spot-instance-feedback@amazon.com if you have more feedback. Alternatively, to learn more about Spot, please visit the Spot Instance page for more details.

Jeff;

New Docs: AWS Getting Started Guides for Linux and Microsoft Windows

We’ve created three new documents to make it even easier for you to get started with AWS:

The first two documents (Getting Started Guide: AWS Web Application Hosting for Linux and Getting Started Guide: AWS Web Application Hosting for Microsoft Windows) are designed to help you create scalable, robust web applications that handle sophisticated demands and workloads using AWS. It provides an example architecture diagram of a web application hosted on AWS and a step-by-step walkthrough of how to deploy your web application using AWS services and follow best practices.

The guides walk you through each step of the process. You’ll sign up for the services and install the command-line tools. Then you will create an Elastic Load Balancer, EC2 Security Group, and a Key Pair. Next, you will use Auto Scaling to launch a load-balanced array of Amazon EC2 instances and set up a CloudWatch alarm to drive the Auto Scaling process. You will add database capabilities by launching an Amazon RDS DB Instance along with the associated DB Security Group. With the infrastructure in place, you will install and launch your web application.Finally, you will use the CloudFormer tool to capture your setup as a reusable CloudFormation template. The guide also covers the use of Route 53 for DNS hosting and CloudFront for content distribution.

We also have a brand new Microsoft Windows Guide. This guide contains conceptual information about Amazon EC2, as well as information about how to use the service to create new web applications on Windows instances. Separate sections describe how to program with the command line interface (CLI) and the Query API.

— Jeff;

 

Amazon Linux AMI – General Availability and New Features

We introduced the Amazon Linux AMI in beta form about a year ago with the goal of providing a simple, stable, and secure Linux environment for server-focused workloads. We’ve been really happy with the adoption we’ve seen so far, and we continue to improve the product and further integrate it with other Amazon Web Services tools.

Today we are zapping the “beta” tag from the Amazon Linux AMI, and moving it to full production status. We are also releasing a new version (2011.09) of the AMI with some important new features. Here’s a summary:

  • The Message of the Day now tells you when updates to installed packages are available.
  • While the AMIs default configuration is set to provide a smooth upgrade path from release-to-release, you can now lock the update repositories to a specific version to inhibit automatic updates to newer releases.
  • Security updates are automatically applied on the initial boot of the AMI.  This behavior can be modified by passing user data into the AMI with cloud-init.
  • There’s a new Amazon Linux AMI Security Center.
  • Puppet has been added to the repositories and is available for system configuration management.
  • Access to the Extra Packages for Enterprise Linux (EPEL) repository is configured, though not enabled by default. EPEL provides additional packages beyond those shipped in the Amazon Linux AMI repositories, but these third party packages are not supported.
  • The cfn-init daemon is installed by default to simplify CloudFormation configuration.
  • A total of 50 new packages are available including the command line tools for AWS, Dash, Dracut, Facter, Pssh, and Varnish. 227 other packages have been updated and 9 have been removed. For a full list of changes, refer to the Amazon Linux AMI Release Notes.

Users of existing Amazon Linux AMIs can either upgrade to the latest release with yum or launch new 2011.09 AMIs. The new AMIs are available in all AWS regions.

— Jeff;

PS – We have some open positions for those with deep Linux systems experience:

AWS Summer Startups: Discovr

Over the summer months, we’d like to share a few stories from startups around the world: what are they working on and how they are using the cloud to get things done. Today, we’re profiling Filter Squad from Perth, Australia!

Discovr team


In one of Werner Vogels‘ many travels through Australia this summer, he tweeted about a lean startup he had met, Filter Squad. Lean startups, not to be confused with bootstrapped startups, are built by adapting agile software development methodologies to business. Some of the concepts include building a minimum viable product, testing business assumptions with real market data, measuring results, and then quickly iterating or pivoting depending on what that data shows you. The concept of a “lean startup” was coined by entrepreneur & author Eric Ries

I reached out and spoke to Stuart Hall, CTO of Filter Squad, and creator of, among others, the Discovr Apps and Discovr Music for the iPhone & iPad.

Meet Filter Squad
Filter Squad is a startup focused on building apps that find what you like, according to CTO Stuart Hall. They began with a #1 selling iPad/iPhone app called Discovr Music in January 2011 and expanded the discovery product suite to include Discovr App in June 2011, which has been a #1 category application in 17 countries. As the name implies, Discovr Music makes it easy for users to find music they like based on their preferences, while Discovr App recommends apps the user might like based on the ones you’re already using. We have been extremely happy with AWS and we also plan to use it for our future products. We are big fans of products such as Amazon RDS and the Elastic Load Balancer to give a complete app scaling solution with Amazon EC2, says Stuart.  
Take a look at the Discovr Music app review from Fox News:

AWS & Lean Startups
Because we are a small, lean team, we were looking for a hosting solution that was going to be easy for us to setup, be reliable, and be easy to scale up and down throughout our product iterations. We looked at a large number of providers, but AWS stood out immediately for a number of reasons:
  • Low maintenance
  • Easy to scale
  • Simple to setup
  • Provided good redundancy
We couldn’t find anyone else who could match the AWS products and price. The number of other large, successful companies also using the service was very reassuring.
Building a Native iPhone/iPad App on AWS
Native mobile apps often need server-side components to create a rich user experience. For our Discovr Apps and Discover Music apps, we have used the following AWS products: 
  • Amazon EC2 – because we had no idea of the market reaction to the application when we launched, flexibility in adding and removing virtual servers based on demand was key. 
  • Amazon RDS – we needed a database that would also be easy to scale and be easy to maintain. Amazon RDS provided easy scaling, easy replication for slave instances, and a system where minor software updates are handled entirely by AWS.
  • Amazon S3 – S3 provides a great and cheap way to host static resources, one with which we had worked before and found ideal for our use case.
  • Amazon Elastic Load Balancer –  the load balancer is provided straight out of the box: it doesn’t require any installing and it needs very little configuration. The load balancer provides built in health checks and takes out instances that are not behaving. Elastic load balancing has been faultless since we launched.
  • Caching: the only thing missing was a caching solution, which AWS has since launched and we will be soon moving to. This was also a big consideration, the pace at which AWS are iterating and improving their service matches our philosophy to application development.
We are also big fans of New Relic for monitoring our AWS instance performance. 

 

Scaling up Ruby on Rails with AWS
We use Ruby on Rails server side, Objective C, and Java for client side. More details of our stack, including our architecture and test data, can be seen detailed on our blog

Words of Wisdom for Other Startups:
Understand that you can do it from anywhere, you dont have to be based in Silicon Valley, or even a big city. With the help of the internet and web services such as the AWS cloud, anyone can deliver great products from anywhere in the world.
For example were based in Perth, Australia. Its a five hour flight to Sydney and our hometown is most definitely not the tech capital of the world! To sum up:
  • Build a great product, then don’t forget to market it!
  • Treat your customers like precious gold.
  • Make it easy for your customers to talk to you and listen to what they say.
  • Cross-promote your app with other apps that youve also built.

We’ve collected some lessons learned on our blog: how we got 250k downloads in 4 days.

——————————————————

8 Days Left to Enter Your Startup in the AWS Start-up Challenge!
This year’s AWS Start-up Challenge is a worldwide competition with prizes at all levels, including up to $100,000 in cash, AWS credits, and more for the grand prize winner. Learn more and enter today!

You can also follow @AWSStartups on Twitter for startup-related updates.

 -rodica

Facebook Developer Update: Meet RootMusic, Funzio, and 50Cubes

In honor of today’s Facebook Developer Conference, I’d like to recognize the success of our existing Facebook app developers and invite even more developers to kick-start their next Facebook app project with Amazon Web Services.

Quick Numbers
We crunched some numbers and found out that 70% of the 50 most popular Facebook apps leverage one or more AWS services. Many of their developers rely on AWS to provide them with compute, network, storage, database and messaging services on a pay-as-you-go basis. In addition to Zyngas popular FarmVille and CafeWorld, or games from Playfish and Wooga, many of the most exciting and popular Facebook apps are also running on AWS.

Here are a few examples:

RootMusic‘s BandPage app (currently the #1 Music App on Facebook, and #8 overall app on Facebook) helps bands and musicians build fan pages that will attract and hold the interest of an audience. RootMusic enables artists to tap into the passion their fans feel for their art and keep them engaged with an interactive experience. More than 250,000 bands of all shapes and sizes, from Rihanna and Arctic Monkeys, to bands you haven’t heard of yet but may soon discover, have already made RootMusics BandPage their central online space for connecting with their fans. Artists use it to share music, release special edition songs/albums here, share photos, and list events/shows. BandPage now supports 30 million monthly active users from all over the world. Behind all the capabilities that ignite BandPages music fan communities lies a well-thought out, highly-distributed and highly-scalable backend, powered by Amazon Web Services:

In 20 seconds, we can double our server capacity. In a high-growth environment like ours, it’s very important for us to trust that we have the best support to give to the music community around the world. Five years ago, we would have crashed and been down without knowing when we would be back. Now, because of Amazons continued innovation, we can provide the best technology and scale to serve music communities needs around the world, Christopher Tholen, RootMusic CTO.


Funzio‘s Crime City is #7 in the top 10 Facebook apps, and its the highest rated Facebook game to reach 1 million daily users with an average user rating of 4.9 out of 5. Crime City currently has 5.5 million monthly active users, with 10 million monthly active users at its peak. The iPhone version was recently listed among the top 5 games in the Apple Appstore and #1 free game in 11 countries and counting. Crime City sports modern, 3D-like graphics that look great on both Facebook and iPhone, and has a collection of hundreds of virtual items that players can collect.

Powering this incredibly rich user experience across multiple platforms is their business acumen in promoting the app, as well as a strong backend that leverages many AWS products to serve their viral and highly active user base. Funzio uses Amazon EC2 to quickly scale up and down based on demand, Amazon RDS to store game and current state information. They use Amazon CloudFront to optimize the delivery to a global, widely-distributed audience and to meet Facebook’s SSL certificate requirements.

At Funzio, we use AWS exclusively to host the infrastructure for our games. When developing social games, you need to be ready for that traffic burst for a hit game in a moment’s notice. AWS provides us with the flexibility to quickly and efficiently scale our applications at all layers, from increasing database capacity in RDS, to adding more application or caching servers within minutes in EC2. Amazon’s cloud services allow us to focus our efforts on developing quality games and not on worrying about managing our technology operations. – Ram Gudavalli, Funzio CTO.


50Cubes, the creator of Mall World, is a startup that has developed one of the most highly-regarded and longer-running successful female focused social game on Facebook. With over 5 million monthly active users, Mall World has a track record of being not only one of the first but also the top game of its kind for the past 1.5 years and continues to entice users world-wide.

50Cubes powers Mall World and other games they developed with a suite of AWS products. Out of these, they value the Amazon Auto-scaling and EBS features the most these products helps them effortlessly scale up and down their exclusive use of Amazon EC2 instances with user demand. Their database clusters are a mix of MySQL and other key value storage databases, all hosted and managed by the team on Amazon EC2 using EBS for Cloud Storage.

One thing that impresses me the most about AWS services is that they have rapidly iterated and improved their products and services over the past year and half, executing almost like a startup of our scale.” – Fred Jin, 50cubes CTO.

Get Started: Your Facebook App, Powered by AWS
Doug Purdy, Director of Developer Relations at Facebook, said:

AWS is great for Facebook developers you can start small, test and prove your ideas. As your app grows, you can easily scale up your resources to keep your users engaged and connected. AWS allows developers to build highly-available, highly-scalable, cost-efficient apps that provide the type of rich and responsive user experiences that our global audience has grown to expect.

To make it as easy as possible for you to get started, we’ve updated our Building Facebook Apps on AWS page. We have also improved and refreshed our Facebook App AMI. The new AMI uses AWS CloudFormation to install the latest versions of the Facebook PHP SDK and the AWS SDK for PHP at startup time. If you want to learn more about developing AWS applications in PHP, feel free to check out the free chapters of my AWS programming book (or buy a complete copy).

— Jeff;