Transactional databases are one of the most important applications run in data centers today. They can be used by any industry that conducts transactions -- e-commerce sites, banks, stock exchanges, retail stores, to name a few. Transactional workloads rely heavily on the underlying storage to read or write individual rows of content very quickly1. If the storage deployed consists of 24G SAS SSDs, data bandwidth will be significantly better than SATA SSDs. This is largely due to the 24G SAS interface that can transfer data at up to 22.5 gigabits2 per second (Gb/s), and is many times faster than the current SATA Revision 3.0 interface at up to 6 Gb/s of data transfer. In comparison with SATA SSDs, 24G SAS SSDs can significantly improve database transactions per minute (TPM), as well as responses to incoming database queries.
Our lab wanted to validate the performance improvements that can be achieved using 24G SAS SSDs when compared with SATA SSDs, and set-up a test comparison. We used a Microsoft® SQL Server™ database to simulate the transactional workloads and HammerDB3 test software to compare our KIOXIA PM7-V Series 24G SAS SSDs with SATA SSDs from a leading provider. We tested TPM, read/write throughput and read/write latency. We also included a CPU usage comparison as higher processing core utilization leads to higher TPM.
Huge 24G SAS SSD Performance Boosts
We were very pleased that our testing showed the KIOXIA PM7-V Series 24G SAS SSDs were able to handle 243% more database transactions when compared with SATA SSDs (see results below). They also delivered 281% higher read throughput and 223% higher write throughput.
We also thought that latency would be a key test metric as slow storage can have a drastic impact on transaction queueing. We were also happy to see that the KIOXIA PM7-V Series 24G SAS SSDs delivered an 8.9x improvement in read latency and a 109.2x improvement in write latency when compared with SATA SSDs. For servers experiencing slow or bottlenecked performance from deployed SATA SSDs, these transactional databases can garner a significant latency boost from 24G SAS SSDs. This outstanding latency boost from 24G SAS SSDs can also improve performance in other primary data center applications, such as cloud services, data analytics, AI/ML and financial services.
With the increased performance delivered by 24G SAS SSDs, we also noticed that the CPU usage increased and the test server was able to use its processing cores efficiently to service more transactions. The KIOXIA PM7-V Series 24G SAS SSDs demonstrated over 53% higher CPU usage when compared with SATA SSDs. Better CPU utilization provides more effective use of the server and application.
Easily Replace SATA SSDs with 24G SAS SSDs
Data centers can take advantage of these enhanced performance and CPU utilization improvements today as most servers ship with a 24G SAS infrastructure and backplane, even if they are purchased with SATA SSDs. This capability enables SAS and SATA drives to be used in the same drive bay enabling slow SATA SSDs to be easily replaced with higher performing and larger capacity 24G SAS SSDs.
Check out all of the results compiled from our lab testing in the full performance brief available here.
NOTES:
1 Read and write speed may vary depending on the host device, read and write conditions, and file size.
2 Definition of capacity - KIOXIA Corporation defines a megabyte (MB) as 1,000,000 bytes, a gigabyte (GB) as 1,000,000,000 bytes and a terabyte (TB) as 1,000,000,000,000 bytes. A computer operating system, however, reports storage capacity using powers of 2 for the definition of 1Gb = 230 bits = 1,073,741,824 bits, 1GB = 230 bytes = 1,073,741,824 bytes and 1TB = 240 bytes = 1,099,511,627,776 bytes and therefore shows less storage capacity. Available storage capacity (including examples of various media files) will vary based on file size, formatting, settings, software and operating system, and/or pre-installed software applications, or media content. Actual formatted capacity may vary.
3 HammerDB is benchmarking and load testing software that is used to test popular databases. It simulates the stored workloads of multiple virtual users against specific databases to identify transactional scenarios and derive meaningful information about the data environment, such as performance comparisons.
TRADEMARKS:
Microsoft and SQL Server are registered trademarks or trademarks of Microsoft Corporation in the United States and/or other countries. All other company names, product names and service names may be trademarks of third-party companies.
DISCLAIMERS:
KIOXIA America, Inc. may make changes to specifications and product descriptions at any time. The information presented in this blog is for informational purposes only and may contain technical inaccuracies, omissions and typographical errors. Any performance tests and ratings are measured using systems that reflect the approximate performance of KIOXIA America, Inc. products as measured by those tests. In no event will KIOXIA America, Inc. be liable to any person for any direct, indirect, special or other consequential damages arising from the use of any information contained herein, even if KIOXIA America, Inc. are advised of the possibility of such damages.