My daily readings 08/03/2010

  • tags: EC2

    • As long as you’re not at your instance limit (which used to be 20 instances per account, unless they raise it for you), you can certainly start a new instance in parallel, which I recommend if you want the change-over to be as smooth as possible. Like Allen says, if you want, you can create an AMI from your image to save state (except for what’s on /mnt, which you need to backup yourself). Once you boot  your new image, you could set up an NFS mount to copy the data in /mnt over directly, or use some other means to move it to S3 in between.


  • tags: EC2

    • As you can see the spot price for the Amazon Linux small EC2 instance has hovered around $0.03/hr since inception, only once did it spike to $0.045/hr, and that too for a very short period.
      • This means the spot price rate is the same rate as that being charged for a dedicated instance after you make a large payment ($227.50 for 1 year or $350 for 3 years). So lets compare how much it would have cost me for 1 year of a small EC2 instance for each of the three different payment options based on current prices.

        • On-Demand Instance : $0.085 * 24 * 365 = $744.60
        • Dedicated Instance : ($0.03 * 24 * 365) + $227.50 = $490.3
        • Spot Instance : $0.03 * 24 * 365 = $262.8
    • Reserved Instances are easy to use and require no change to how you use EC2. When computing your bill, our system will automatically apply Reserved Instance rates first to minimize your costs. An instance hour will only be charged at the On-Demand rate when your total quantity of instances running that hour exceeds the number of applicable Reserved Instances you own.
    • Many applications require servers to be available at all times. Reserved Instances could provide your business substantial savings if you currently own your own hardware or exclusively use On-Demand instances for your steady state applications.
    • Reserved Instances can also provide cost savings even if your applications run less than 100% of the time. Unlike traditional dedicated capacity, Reserved Instances allow you to avoid usage charges when you don’t need the instance. 3-year term Reserved Instances breakeven with On-Demand rates at 24% utilization or greater. The table below shows the possible savings over On-Demand rates depending on utilization. The higher your utilization, the more savings Reserved Instances provide.
    • For example, a web application maintains the same average usage level most of the year, but sees traffic spike occasionally. They purchase 5 three-year term Reserved Instances to handle their everyday levels of usage and leverage additional On-Demand instances to meet spikes in demand. If they had wanted to have the same flexibility with self-owned hardware, they would have needed to purchase 8 servers to meet their peak demand. With Reserved Instances, their costs average to an effective hourly rate 40% lower than the On-Demand rate.
    • Customers using both Reserved and On-Demand Instances will have Reserved Instance rates applied first to minimize costs. Each hour, if the number of running instances is less than or equal to the number of applicable Reserved Instances you have, all running instances will be charged at the Reserved Instance hourly rate. When your total quantity of running instances exceeds the number of applicable Reserved Instances you own, you will be charged the On-Demand rate.

      For example, if you own 3 Reserved Instances with the same instance type and Availability Zone, the billing system checks each hour to see how many total instances you have running that match those parameters. If it is 3 or less, you will be charged the Reserved Instance Rate for each instance running that hour. If more than 3 are running, you will be charged the On-Demand rate for the additional instances.

      If you purchase a Reserved Instance for an Availability Zone where you have a running instance, the lower hourly charge will automatically be applied to that instance moving forward with no need to restart.

  • tags: EC2

    • Amazon EC2 instances are grouped into three families: Standard, High-Memory and High-CPU. Standard Instances have memory to CPU ratios suitable for most general purpose applications; High-Memory instances offer larger memory sizes for high throughput applications, including database and memory caching applications; and High-CPU instances have proportionally more CPU resources than memory (RAM) and are well suited for compute-intensive applications. Cluster Compute instances provide a very large amount of CPU coupled with increased network performance making them well suited for High Performance Compute (HPC) applications and other demanding network-bound applications. When choosing instance types, you should consider the characteristics of your application with regards to resource utilization and select the optimal instance family and size.

      One of the advantages of EC2 is that you pay by the instance hour, which makes it convenient and inexpensive to test the performance of your application on different instance families and types. One good way to determine the most appropriate instance family and instance type is to launch test instances and benchmark your application.

    • Amazon EC2 uses a variety of measures to provide each instance with a consistent and predictable amount of CPU capacity. In order to make it easy for developers to compare CPU capacity between different instance types, we have defined an Amazon EC2 Compute Unit. The amount of CPU that is allocated to a particular instance is expressed in terms of these EC2 Compute Units. We use several benchmarks and tests to manage the consistency and predictability of the performance of an EC2 Compute Unit. One EC2 Compute Unit provides the equivalent CPU capacity of a 1.0-1.2 GHz 2007 Opteron or 2007 Xeon processor. This is also the equivalent to an early-2006 1.7 GHz Xeon processor referenced in our original documentation. Over time, we may add or substitute measures that go into the definition of an EC2 Compute Unit, if we find metrics that will give you a clearer picture of compute capacity.

      To find out which instance will work best for your application, the best thing to do is to launch an instance and benchmark your own application. One of the advantages of EC2 is that you pay by the hour which makes it convenient and inexpensive to test the performance of your application on different instance types.

    • Amazon EC2 provides virtualized server instances. While some resources like CPU, memory and instance storage are dedicated to a particular instance, other resources like the network and the disk subsystem are shared among instances. If each instance on a physical host tries to use as much of one of these shared resources as possible, each will receive an equal share of that resource. However, when a resource is under-utilized you will often be able to consume a higher share of that resource while it is available.
  • tags: Startup Organization design

  • tags: iPad development

  • tags: Gmail plugin

    • I don’t think it will change anytime soon.

      1, The marketplace sucks compared to the app store in many ways. There are few ways to feature your apps on the marketplace vs apple which has the “what hot”, “new and noteworthy”, “top paid”, “top grossing” categories. In comparison, the android marketplace doesn’t.

      2, On the apple app store, you are given more promotion room on your app page itself for more descriptive text and screenshots.

      3, The android app store is FILLED with crap and spam apps, just go to the “whats new” section of your android marketplace and its filled with 95% crap-ware and those ringtone apps. There is no clear way for consumers to find good apps vs these garbage apps. This destroys the user’s experience in the marketplace.

      4, There is a culture of “free” and “opensource” in the entire Android ecosystem and most apps are released for free or are ad-supported. Google seem to love this since if you use their ad system, they get a cut and therefore they don’t seem to have an incentive to push paid apps to have wider availability.

    • Google showcased a ton of this at Google I/O.

      Android ties in really well to their webapps, Google TV (If that succeeds), etc.

      I remember one example was that you can go to and click on a phone icon, and it just sends it straight to your phone which opens a map in the same place. That’s really useful integration. I forget the rest of the examples, but they were all pretty ‘wow’.

      I think iTunes is an example of how not to do things. The iPhone is reliant on iTunes completely. With android, it stands on its own and can integrate with other things when you want it to.

    • Unfortunately, the Android Marketplace as well as the SDK still sucks for us developers. Buggy and inconsistent behavior across different devices; Bad performance (MotionEvent anyone?); The lack of paid apps in different countries and the prevalence of spammy crapware apps on the unregulated marketplace. I wonder if google will do anything to address these issues as Android gain a wider audience. Otherwise I can see this as a hindrance to the adoption of Android.
  • tags: sales android

    • Sales of Google Android phones in the U.S. are rising so quickly, the devices have outsold Apple handsets for the first time on record. New smartphone subscribers choosing Google phones accounted for 27 percent of U.S. smartphone sales, the Nielsen Company will announce this morning, nudging past the 23 percent share held by Apple. But Android isn’t just a U.S. phenomenon: Canalys today estimates that Android handset sales grew 886 percent worldwide from the year ago quarter.
      • The warehouse address is repeated in every record that refers to a part stored in that
      • If the address of the warehouse changes, every record referring to a part stored in that
        warehouse must be updated.
      • Because of the redundancy, the data might become inconsistent, with different records
        showing different addresses for the same warehouse.
      • If at some point in time there are no parts stored in the warehouse, there may be no
        record in which to keep the warehouse’s addres
      • Good reasons

Posted from Diigo. The rest of my favorite links are here.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: