Information Technology News.


Let's have a look at Cisco's HyperFlex converged solutions

Share on Twitter.

Sponsered ad: Get a Linux Enterprise server with 92 Gigs of RAM, 16 CPUs and 8 TB of storage at our liquidation sale. Only one left in stock.

Sponsered ad: Order the best SMTP service for your business. Guaranteed or your money back.

March 7, 2017

Today, we're taking a good look at Cisco's new HyperFlex converged solutions that Cisco says can support more than twice the number of virtual machines than other HCI systems from competing vendors, while still maintaining high performance, according to various tests recently performed by ESG.

The benchmark tests used HCI apparatus and Oracle's VD-Bench tool, with the HCI products running VMs.

It demonstrates to the IT industry that the extensive testing was executed using a stringent methodology including several months of baselining and iterative testing, says HCI.

Overall, several industry benchmarks were run for long periods of time to observe the overall performance as it would occur in a customer's environment.

Additionally, tests were run many times, never back-to-back but separated by several hours and even days and the overall results were a mixed bag.

"Also, testing was done using data sets large enough to ensure that the data didn't remain in a cache, but leveraged the back-end disk across each cluster," HCI asserted.

The ESG testers compared a 2U, four-node HyperFlex HX-220c hybrid flash and disk array system with three other competing products from vendors A, B, and C. We're told these are leading hyperconverged infrastructure vendors and "included two software-only systems from leading vendors that leveraged standard x86 based servers, and a proprietary system from a single vendor based on its own hardware and partially integrated with its own software."

These comparative HCI products were also in the 2U form factor, four-node systems with similar configurations, although we were not given the CPU and DRAM details.

The overall testing was performed using various read/write profiles and block sizes, with 100 percent random data. One test looked at overall cluster scalability with a synthetic workload designed to emulate a typical OLTP I/O mix, 70 percent read, 100 percent random with a per-VM target of 800 IOPS.

The test was run across 140 VMs in each cluster for three to four hours with a goal of remaining at or below 5 milliseconds write latency.

HyperFlex managed 140 VMs at sub-5ms write latency. ESG's paper states: "Vendor A successfully supported 70 VMs at 4.65 ms average response time, Vendor B passed running 36 VMs with 5.37 ms average response times, and Vendor C supported 48 VMs at sub-5.02 ms response times."

ESG also looked at read and write latency and SQL Server I/O patterns and checked out all-flash configurations of HyperFlex and a Vendor B.

The all-flash HX-220c had one 400 GB cache SSD and six 960 GB SSDs. Vendor B's system was again a four-node cluster like the HX-220c, fitted with the same CPU and memory as the the HX-220c, 2 x 480 GB cache SSD and 6 x 960 GB SSDs.

Testing looked at IOPS and latency for the virtual machines and for a SQL Server overall I/O pattern.

The results were mixed. In the SQL test "Cisco HyperFlex cluster more than tripled the IOPS of Vendor B with an average response time of about 5.3 ms.

The average vendor B response times were about 30.58 ms due to an extremely high write response time of 99.84 ms throughout the test.

That test was run several times on multiple days with consistent results. ESG also looked at the IOPS per VM with 140 VMs in each cluster, and found a marked difference.

The ESG testers found that the HyperFlex hybrid cluster can support more than twice the number of VMs as rivals while maintaining low latency, and deliver 2X-8X the IOPS for 140 VMs in a cluster, using an OLTP workload.

With a SQL workload, the hybrid HyperFlex delivered significantly more IOPS and lower latency than other products.

For all-flash testing, HyperFlex delivered higher IOPS and lower latency, and consistent high performance across all VMs, ESG asserted.

However, that is Cisco marketing material that ESG paid to run the tests, so take this with a grain of salt...

We are not told who the three competing vendors are and the test results are partial. Your mileage may vary and you should run your own tests on your own workloads.

With that in mind, the HyperFlex systems look interesting and appear to avoid suffering from storage hotspots.

Cisco now says it has a little over 1,000 HyperFlex customers after nine months of selling. Thirty percent of these are new to Cisco.

The company says you can add extra compute to a HyperFlex cluster without having to pay any extra software licensing or usage fee.

Cisco asserts that its HyperFlex solutions work with its other server offerings as well, like converged systems such as FlexPods, and are managed with the same UCS Director.

Source: Cisco.


Sponsered ad: Get a Linux Enterprise server with 92 Gigs of RAM, 16 CPUs and 8 TB of storage at our liquidation sale. Only one left in stock.

Sponsered ad: Order the best SMTP service for your business. Guaranteed or your money back.

Share on Twitter.

IT News Archives | Site Search | Advertise on IT Direction | Contact | Home

All logos, trade marks or service marks on this site are the property of their respective owners.

Sponsored by Sure Mail™, Avantex and
by Montreal Server Colocation.

       © IT Direction. All rights reserved.