Public cloud market leader Amazon Web Services (AWS) today announced that its six types of I3 virtual machine (VM) instances are now available in 15 out of AWS’ 16 data center regions around the world. AWS first announced the I3 series of instances at its re:invent conference in November.
These instances backed by solid-state drives (SSDs) are designed for high storage throughput and input-output operations per second (IOPS); the largest instance can handle as much as 16GB per second of sequential disk throughput, AWS chief evangelist Jeff Barr wrote in a blog post. They will work well for relational and not-just-relational databases, among other things, Barr wrote.
At the low end of the range, the i3.large offers 2 vCPUs, 15.25GiB (gibibytes) of RAM, and 0.475TB of storage; at the high range, the i3.16xlarge carries 64 vCPUs, 488GiB of RAM, and 15.2TB of storage spread across eight SSDs. The i3.large costs 15 cents per hour in AWS’ US East (Northern Virginia) region, while the i3.16xlarge goes for $4.99 per hour in US East.
AWS’ closest competitor, Microsoft Azure, offers the L-Series of storage-optimized instances. At the high end, the L32 instance offers 32 cores, 256GiB of RAM, and 5.63TB of storage, for the price of $7.782 per hour. Google Cloud Platform currently does not have storage-optimized instances, although it does let customers create customized instances.
AWS made the predecessor i2 series instances available in 2013.