Oracle, AWS and Cloud

It’s been a long gap for my blogging activity. For past 2 years I’ve been shifting my focus on core oracle technologies and started working on cloud technologies more. It’s been an interesting journey for me for past couple years. I’ll start blogging on new technologies that I’ve been working on and start posting more work logs over here.

-Ravi.M

EBS-Optimization inconsistencies

Note: Below issue has been fixed by AWS sometime between 6/15 and 6/23. Ref:http://web.archive.org/web/20140623182841/http://aws.amazon.com/ec2/previous-generation/

Today I’ve observed that since introducing new EC2 instances and announcing the present-generation and previous generation instances, AWS made some changes to their EC2 instances. One of the notable change is the EBS Optimization characteristics of previous generation’s instances.

Below is the screenshot of this information as on 6/15/2014 5:40 PM EST.

As you can see there are 2 major changes… m2.2xlarge and m2.4xlarge instance types do not have EBS-Optimized option anymore. However, this information is contradicting with other pages with in AWS website.

By referring to EBS optimization page in AWS EC2 documentation at http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/EBSOptimized.html, it is clearly mentioned that, these 2 instance types are having EBS optimization capability of 500 MBPS and 1000 MBPS respectively.

Another notable change is with instance type cr1.8xlarge. As per the table, network performance on this instance type is “High”. However, it used to “10 Gigabit”. Same was confirmed using the original blogpost from its launch back in Jan 2013 @ http://aws.amazon.com/blogs/aws/ec2-for-in-memory-computing-the-high-memory-cluster-eight-extra-large/

I did another quick check in AWS console to make sure that backend AWS engine is not modified to decrease this performance and I could see that it’s just the EC2 instance page that was modified incorrectly.

Hope AWS folks will correct this pretty soon.