Cisco and MapR post their 1st Big Data TPCx-HS benchmark
Share on Twitter.
Get the most reliable SMTP service for your business. You wished you got it sooner!
March 5, 2015
This morning, Cisco and MapR have posted their first big data TPCx-HS benchmark, but for now, there's nothing
to actually compare it with yet.
The TPC Express HS (TPCx-HS) benchmark result was based on a few clustered Cisco UCS servers
running Red Hat Enterprise Linux and MapR Hadoop.
TPC is the Transaction Processing Council, which helps produce auditable benchmarks in the IT industry.
Overall, the TPCx-HS specification suggests the test "stresses both hardware and software, including
the Hadoop run-time, the Hadoop Filesystem API compatible systems and MapReduce layers".
The scale factors available run from 1 TB through 3 TB, 10 TB, 30 TB, 100 TB, 300 TB, 1,000 TB, 3,000 TB and 10,000 TB,
which measure Hadoop architectures across several dimensions.
This also includes the overall performance and cost on an equal footing at the level of scale that
is appropriate for each unique deployment.
For its part, MapR ran a series of tests at 1 TB, 3 TB and 10 TB. The performance metric is the
HSph score, reflecting the TPCxHS throughput or processing power, at each capacity level.
A TPCxHS result is only comparable with other TPCxHS results at the same scale factor.
The MapR/UCS/Red Hat scores, each run with a cluster of 16 x 2 x E5-2660 – 2.2 GHz servers, were:
1 TB – 5.07 HSph at $121,23/HSph
3 TB – 5.10 HSph at $120,519/HSph
10 TB – 5.77 HSph at $106,524/HSph
You're invited to come on Cloudera and Hortonworks, step up to the TPCx-HS benchmark and post
results, if you're team is interested.
Cisco resells the MapR software and its channel partners can select MapR SKUs from Cisco's suggested resale price
Get the most dependable SMTP server for your company. You will congratulate yourself!
Share on Twitter.