Excellent news. This may come handy. In our data center we have a few servers for just two applications. These applications are just run for 2 or 3 days a month and then the rest of the time all servers in rack just sit idle. It is a waste of servers, time, energy and resources. This is a good use-cases for on-demand high I/O server(s), where I need low-latency and are an exceptionally good host for NoSQL databases such as MongoDB.
The most important part here is local ssd storage. From the blog post:
The SSD storage is local to the instance. Using PVM virtualization, you can expect 120,000 random read IOPS (Input/Output Operations Per Second) and between 10,000 and 85,000 random write IOPS, both with 4K blocks...
Why use on demand high I/O server?
- Spin up instances on-demand for high I/O intensive job.
- I can test a new app or upgrades without any deployments.
- May result into cost savings.
New Amazon high I/O EC2 instances specs
The specs are as follows for new family is the High I/O Quadruple Extra Large (hi1.4xlarge in the EC2 API) (taken from the blog post):
- 8 virtual cores, clocking in at a total of 35 ECU (EC2 Compute Units).
- HVM and PVM virtualization.
- 60.5 GB of RAM.
- 10 Gigabit Ethernet connectivity with support for cluster placement groups.
- 2 TB of local SSD-backed storage, visible to you as a pair of 1 TB volumes.
You can get High I/O Quadruple Extra Large instances in US East (Northern Virginia) and EU West (Ireland) today, at an On-Demand cost of $3.10 and $3.41, respectively. So it comes around (approximate cost) $12k per/year for reserved instance.
Check out related media
For more information see: New High I/O EC2 Instance Type - hi1.4xlarge - 2 TB of SSD-Backed Storage.