Transcript
Matching Hard Drives to Cloud Storage Workloads Kelsie J. Betsch, PhD
Seagate Point of View
The Effect of Drive Performance on System-Level Performance and the Impact on IT Decisions Introduction Understanding workloads and drive capabilities allows cloud storage providers to make informed choices about what hard drives to use in different situations. To show how important it is to understand how different drives perform with a given workload, this paper presents results from object storage performance tests using two kinds of hard drives: Seagate® Enterprise Capacity 3.5 and Seagate Terascale®. Enterprise Capacity 3.5 HDDs have higher reliability design and better performance, however Terascale HDDs cost less and consume less power. By reviewing the results of the tests, it will be possible to see how the drives’ performance affects system-level performance and how that may factor into IT decisions.
Methodology Setup for testing consists of a small object storage setup with one proxy and one object storage server. The cluster was provisioned with OpenStack Swift, a type of cloud object storage software used to store and retrieve data with an API. Then, the system was tested with a series of workloads varying in size, worker count, and get:put ratio. One set of tests was run using 60 Enterprise Capacity 3.5 HDDs, the other set of tests was run using 60 Terascale HDDs. For more detailed information regarding methodology, refer to Advanced Methodology for Benchmarking Object Storage Solutions.
Matching Hard Drives to Cloud Storage Workloads Benchmark Performance is determined by finding out the average bandwidth achieved for each workload. Section a of Figure 1 shows the get (read) bandwidth achieved for each of the workloads when the object storage server is filled with Enterprise Capacity 3.5 (gray) and Terascale (orange) HDDs. Section b of Figure 1 shows the put (write) bandwidth. The measured bandwidth approaches the theoretical network maximum (10GbE = 1250MB/s; horizontal red line) for many of the largeobject size workloads when using either drive type. However, the Enterprise Capacity 3.5 HDDs afford a larger throughput for a number of workloads that are not constrained by the network bottleneck. Side by side, the Enterprise Capacity 3.5 HDDs perform better in 140 out of the 160 workloads by an average margin of 18.9%. The results are similar for puts: the Enterprise Capacity 3.5 HDDs perform better in 115 out of 160 workloads by an average margin of 15.3%. For the remaining workloads, the performance difference is within the experimental uncertainty of ±7%.
Figure 1. Read and write bandwidth. Note: The horizontal red line indicates the 10GbE network limit.
Conclusion By testing different drives, it is easy to see which drive offers the best performance for a particular workload. This allows cloud storage providers to make informed choices when weighing cost, power consumption, capacity, performance and reliability.
www.seagate.com AMERICAS ASIA/PACIFIC EUROPE, MIDDLE EAST AND AFRICA
Seagate Technology LLC 10200 South De Anza Boulevard, Cupertino, California 95014, United States, 408-658-1000 Seagate Singapore International Headquarters Pte. Ltd. 7000 Ang Mo Kio Avenue 5, Singapore 569877, 65-6485-3888 Seagate Technology SAS 16–18, rue du Dôme, 92100 Boulogne-Billancourt, France, 33 1-4186 10 00
©2014 Seagate Technology LLC. All rights reserved. Printed in USA. Seagate, Seagate Technology and the Wave logo are registered trademarks of Seagate Technology LLC in the United States and/or other countries. Terascale is either a trademark or registered trademarks of Seagate Technology LLC or one of its affiliated companies in the United States and/or other countries. All other trademarks or registered trademarks are the property of their respective owners. When referring to drive capacity, one gigabyte, or GB, equals one billion bytes and one terabyte, or TB, equals one trillion bytes. Your computer’s operating system may use a different standard of measurement and report a lower capacity. In addition, some of the listed capacity is used for formatting and other functions, and thus will not be available for data storage. Seagate reserves the right to change, without notice, product offerings or specifications. PV0029.1-1410US, October 2014
