Are you ready to take full advantage of an all-NVMe* system?
The advantages of all-NVMe storage systems are being discovered now. Elimination of the storage subsystem as a bottleneck represents a paradox shift. If your HCI vendor qualifies all NVMe drives, but still uses legacy protocols and underpowered CPUs, you are leaving performance on the table.
Hyperconverged solutions are being evaluated to host mission-critical workloads, but results vary greatly. The level of performance and responsiveness needed by mission-critical workloads have eluded many HCI solutions until now.
In order to achieve the desired performance and responsiveness, a total new generation of hardware and software is needed.
Cisco’s next generation HyperFlex* system enables use of all NVMe SSDs, with Intel® Optane™ DC SSDs to fully utilize all the resources available. And it takes the performance to a whole new level.
But what does it mean to have all this performance? It means you have more design choices.
You can achieve your performance target with few nodes in the solution, or spread the load across a fewer number of cost effective nodes. You can already see Cisco HyperFlex leadership when compared to other HCI all-Flash solutions.
Check out how a Microsoft SQL Server* online transaction processing workload behaves in an all-FLASH HyperFlex solutions.
SQL OLTP Workload – Aggregate Testing IOPS – HyperFlex ALL FLASH
You can see that total IOPS on Cisco HyperFlex significantly outperform competition on similar hardware.
Configuration data used for test above (see source in footnote 1):
But what if we can take that hyperdrive even higher, and do it with lower core count and a smaller memory footprint?
That’s the efficiency of all-NVMe designs plus Intel Optane DC SSD with HyperFlex 4.0.
This is the result for the same test – SQL Aggregate Testing IOPS- but with our new all-NVMe + Intel Optane DC SSD platform.
SQL Server OLTP Workload—Aggregate Testing IOPS
Configuration data used for test above (see source in footnote 2):
So what does this hyperdrive performance mean?
For starters, you can have less nodes to achieve a certain minimum performance. Let’s say you have a solution target of 500K IOPS. With previous all-FLASH solutions, you would need to add at least one node. With all-NVMe and Intel Optane DC SSDs, that is no longer necessary. You can actually meet that target with the same four nodes with performance to spare. And, you already saved in extra hardware and software cost.
So…are you ready to take full advantage of your all-NVMe system?
See other ways in which Intel® Optane™ DC SSDs are making an impact in the data center.
References and Footnotes
1 Source – Enterprise Strategy Group: Mission-critical Workload Performance Testing of Different Hyperconverged Approaches on the Cisco Unified Computing System Platform (UCS). https://www.cisco.com/c/dam/en/us/products/collateral/hyperconverged-infrastructure/hyperflex-hx-series/esg-hyperflex-compl-work.pdf
2 Source – Enterprise Strategy Group: Mission-critical Hyperconverged Workload Performance Testing on Cisco HyperFlex All NVMe with Intel Optane DC SSD. http://cs.co/esgnvme
Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such as SYSmark and MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more complete information visit www.intel.com/benchmarks.
*Other names and brands may be claimed as the property of others.
© Intel Corporation. All rights reserved.