Exploiting Hardware Differences in Cloud Data Centers

Any large cloud vendor would have several data centers located in many different parts of the world. Between all of these data centers, there would be tens of thousands of disks, CPUs, RAM chips and networking devices. All of this equipment would be of differing vintage. Even if all data centers started off with identical equipment (which is seldom the case since they do not get built all together), over time, they would be subject to differing schedules of hardware maintenance and replacement of unserviceable components. Therefore, with the passage of time, even the most strictly managed data centers will tend to become heterogeneous in nature. The end result of this variation in equipment is that differences in performances begin to appear for different cloud instances. Since cloud usage is typically charged over the hour (per minute charging is also emerging now), if you are able to move your application to a cluster with higher performance, you can achieve significant savings. The question is how to find clusters that are faster and more efficient. Recent research at Aalto University Finland and Deutsch Telekom Laboratories, Germany focused on understanding hardware variations between data centers of the same vendor and studied the performance variation between various instances. While they studied a global vendor offering PaaS services, the results would be applicable to all large vendors. (It is possible that small cloud computing vendors may be able to have a greater degree of similarity of hardware).

The researchers used standard operating system commands to obtain the CPU details of their clusters and verified the output using other system calls. After considerable effort, they determined that the vendor had the following different types of CPUs –

S No	CPU Name	% availability in 2011	% availability in 2012
1	E5507	58%	40%
2	E5430	29%	17%
3	E5645	5%	42%
4	2218HE	4%	1%
5	270	4%	0%

  The trend is very clearly visible – the vendor is slowly phasing out the processors at serial numbers 1,2,4 and 5 of the table above and is switching over to E5645 (serial number 3). When the list of CPUs was broken down by availability zones, there were some zones where the newer E5645 processor was barely present (<10%) and in some others, the numbers had risen to nearly 90%. Obviously, new data centers had the most modern processors. When the output of the different processors was compared, it was seen that there were differences amongst all and that the new ones were delivering 1.25 times the throughput of the next best ones. Compared with the earliest CPUs, there was a performance difference of more than 1.6 times. Armed with these results, the research team began to calculate costs. After a fair amount of calculations, liberally laced with probability and calculus, some very interesting answers emerged. If a large task took 100 server instances and took a year to complete, then by switching the instances from old processors to new ones (with a 1.6 times performance advantage), there was a net saving of $40,664 (the per hour rate of the original instance was $0.34). If you are a small business hoping to make a major impact in your domain, this kind of saving is very significant. If users take some care to select those data centers where the infrastructure is new, they would be automatically getting the better processors and other hardware. Users who are aware of these issues can select instances where they getter better performance. In times such as this, where every dollar needs to be squeezed for maximum value, this research offers companies a simple way to further reduce expenditure.

Be Part of Our Cloud Conversation

Our articles are written to provide you with tools and information to meet your IT and cloud solution needs. Join us on Facebook and Twitter.   About the Guest Author: Sanjay Srivastava has been active in computing infrastructure and has participated in major projects on cloud computing, networking, VoIP and in creation of applications running over distributed databases. Due to a military background, his focus has always been on stability and availability of infrastructure. Sanjay was the Director of Information Technology in a major enterprise and managed the transition from legacy software to fully networked operations using private cloud infrastructure. He now writes extensively on cloud computing and networking and is about to move to his farm in Central India where he plans to use cloud computing and modern technology to improve the lives of rural folk in India.