Late last month I gave you a sneak peak at our newest EC2 instance type, the I2. These instance types are available today, in four sizes across seven AWS regions.
The I2 instance type was designed to host I/O intensive workloads typically generated by relational databases, NoSQL databases, and transactional systems. The largest I2 instance type can deliver over 365K random reads per second and over 315K random writes per second, both measured with a block size of 4 KB. With four instance sizes, you can start small and scale up as your storage and I/O needs grow.
This is our second generation High I/O instance type, picking up where the HI1 instance left off. In comparison to the HI1 instance type, members of the I2 family offer faster processors, three additional instance sizes, a doubling of the amount of memory per vCPU, and 56% more SSD-based instance storage.
Here are the instance sizes and the associated specs:
|Instance Name||vCPU Count||RAM
||Instance Storage (SSD)||Price/Hour|
|i2.xlarge||4||30.5 GiB||1 x 800 GB||$0.85|
|i2.2xlarge||8||61 GiB||2 x 800 GB||$1.71|
|i2.4xlarge||16||122 GiB||4 x 800 GB||$3.41|
|i2.8xlarge||32||244 GiB||8 x 800 GB||$6.82|
The prices shown above are for On-Demand instances in the US East (Northern Virginia) and US West (Oregon) regions; see the EC2 pricing page for full information.
The instances are available in On-Demand and Reserved form in the US East (Northern Virginia), US West (Oregon), US West (Northern California), EU (Ireland), Asia Pacific (Singapore), Asia Pacific (Tokyo), and Asia Pacific (Sydney) regions.
I2 instances support Hardware Virtualization (HVM) AMIs only. In order to obtain the best I/O performance from these instances, you should use the Amazon Linux AMI 2013.09.02 or any Linux AMI with a version 3.8 or newer kernel. Older versions of the kernel will exhibit lower I/O performance when used with I2 instances.
NUMA (Non-Uniform Memory Access) speeds access to main memory by optimizing for workloads where the majority of requests for a particular block of memory come from a single processor. By enabling processor affinity (asking the scheduler to tie a particular thread to one of the processors) and taking care to manage memory allocation according to prescribed rules, substantial performance gains are possible.
All four sizes of the I2 instance type benefit from our new Enhanced Networking feature. When you launch these instances inside of a Virtual Private Cloud (VPC), you will enjoy low latency, low jitter, and the ability to move a very large number of packets per second (PPS). In order to take advantage of this important feature, you will need to use an HVM AMI with the proper drivers installed (read our documentation on Enabling Enhanced Networking to learn more).
The three smallest instance types also support EBS Optimization, with dedicated network throughput from the instance to Amazon EBS.
As you can see from the table above, the I2 instances include a copious amount of SSD storage, ranging from 800 gigabytes on the i2.xlarge all the way up to 6.4 terabytes on the i2.8xlarge.
The SSD storage now supports TRIM functionality, which improves performance when the SSD handles a series of successive write operations.
Go For Launch
As I mentioned earlier, these instance types are available now in seven AWS regions and you can start to use them right now.