IDEAS Insights

CloudSizer: Out of Beta with Two New Versions

CloudSizer sheds the beta moniker today and unveils two new versions, CloudSizer Pro and CloudSizer Express. Some features of CloudSizer have not changed. It is still powered by RPE2, the industry-recognized server performance estimate. However, the latest release has many exciting new features:

 

CloudSizer	Express	Pro
Usage	Evaluation	Analysis
Physical Server Selection	Approximate Product Level	More Accurate Configuration Level
Cloud Performance Estimates	Smallest Instance	ALL Instances
Access Term	2 Weeks	3 Months
Price	FREE	$299 AUD
Go to
	More information

Buy Some Supercomputing with Your Credit Card on Amazon EC2

Forty-two – the answer to the ultimate question of life, the universe, and everything – is now also the ranking of Amazon EC2's public compute cloud in the November 2011 list of the LINPACK-based Top 500 Supercomputer Sites (Top500). The impressive result utilized EC2's new Cluster Compute Eight Extra Large instance, which has 16 Intel Xeon cores and 60 GB of RAM. The total resources used for the benchmark result were 17,024 cores and 66,000 GB of memory. Depending on whether you divide by cores or memory, that represents either 1,064 or 1,100 EC2 instances. Amazon EC2 has not submitted many official benchmark results, which makes the new results even more exciting (in past Top500 lists, it ranked 233rd and 451st). Hopefully, with all the buzz that the new result is receiving, Amazon will now see the value of participating in benchmarks.

NetApp’s Smart Decision to Target Midsize Customers

NetApp recently refreshed its FAS2000 series of entry-level unified storage systems, introducing the new FAS2240 (FAS2240-2 and FAS2240-4). The launch, which NetApp called the “Smart Decisions Launch,” focused on midsize customers and channel partners. The FAS2240 is designed to appeal primarily to midsize customers, including medium-sized businesses, state and local governments, and educational institutions. The FAS2240 also has channel-friendly packaging and pricing that was clearly set to maintain NetApp’s traditional affinity with resellers and integrators.

Oracle’s High-End Path to Public Cloud

When Oracle released Exalogic a few years ago, it was billed as a private cloud-in-a-box. At OpenWorld, Oracle doubled down on the plug-and-play features of Exalogic and Exadata by having these systems serve as the backbone for its new public cloud offering. Recently, IDEAS published a blog post offering a broader definition of mainframes as an integrated solution stack. By this definition, Oracle is eschewing commodity solutions, the architecture on which many existing cloud services are based, in favor of a high-end solution as the basis for its public cloud (see graphic below). Computing as a service has been successfully monetized (recently confirmed), but by taking the high-end approach, Oracle is trading larger potential sales volumes for a niche customer base. The strategy may prove to be as profitable, but it is not without risk.

How to Model Storage Growth



RPE2, the unique server performance methodology from IDEAS, has made server performance modeling easier. However, predicting storage growth that results from switching servers continues to pose a challenge. End users often ask IDEAS to model storage; this modeling could be interpreted in two different ways: developing a storage performance estimate analogous to RPE2, or correlating storage capacity with changing server performance.

We previously published a blog post that described how to take the initial steps to correlate the variables in the SPC-2 storage benchmark, but more research will be required to develop a storage performance estimate that is as robust as RPE2. In the meanwhile, the latter approach of the two approaches above – correlating storage capacity with changing server performance – may be the most practical. Here, we present some steps to derive equations that project changes in storage needs as server performance grows.

Image source: Jon Delacruz/Wikicommons

'Small Step' for HP Could Be 'Giant Leap' for Hyperscale Servers

HP has launched a multiyear public effort code-named Project Moonshot to enable low-power servers for multiple-node server computing. As part of this announcement, HP will release an evaluation platform using a new ARM-based chip from processor startup Calxeda. This one small step by HP may eventually become a giant leap for hyperscale servers. HP is smartly positioning the announced server as an effort to get ARM off the ground, rather than claiming it’s a product ready for deployment in the datacenter today. Ultimately, whether this announcement is successful depends on whether software developers choose to create or port server applications to ARM. Let’s take a closer look at what was announced.