Amazon Web Services unveiled a new instance type this week that expands the types of workloads that IT shops can...
By submitting your personal information, you agree that TechTarget and its partners may contact you regarding relevant content, products and special offers.
run in the cloud.
The instance, dubbed the "High Memory Cluster Eight Extra Large," includes two Intel E5-2670 processors running at 2.6 GHz with Intel Turbo Boost and NUMA support, 244 GB of RAM, two 120 GB solid-state drives (SSDs) for storage, and 10 Gigabit Ethernet networking with support for cluster placement groups.
The ability to argue that the virtualized servers cannot meet the performance and scalability requirements of enterprise systems is dissolving.
CIO, Robert Half International Inc.
The high-powered instance nullifies arguments for not virtualizing or "cloudifying" systems, said Sean Perry, CIO at Robert Half International Inc., based in Menlo Park, Calif. "This seems like a machine that could easily run most IT database workloads," Perry said. "So, the ability to argue that the virtualized servers cannot meet the performance and scalability requirements of enterprise systems is dissolving."
Some expected this AWS IaaS instance following Amazon's Redshift data warehouse service announcement in November.
Though his company doesn't have any workloads that require this kind of horsepower, Perry said Software as a Service providers will be happy to have this type of server. The High Memory Cluster Eight Extra Large could also be a good platform for consolidating Citrix XenApp workloads, he said.
Lack of larger instance types is a common concern that clients and prospects have, according to Kris Bliesner, AWS partner and CEO at 2nd Watch, a cloud computing consultancy and systems integrator in Liberty Lake, Wash., in an email. "We see this need in analytics and distributed node functions, such as Hadoop clusters, that could benefit from a larger node," he said.
These new instance types also offer a way to meet the demand of a large single-instance workload, Bliesner said. "AWS' pay-as-you-go infrastructure makes these kinds of new instances good for transient workloads," he said. "The advantage here is that you aren't locked into buying an expensive server if you only need the horsepower part of the time."