The Oracle NoSQL Database Blog covers all things Oracle NoSQL Database. On-Prem, Cloud and more.

  • November 10, 2014

Is your NoSQL ready for SSD - benchmark results from SanDisk

It’s well known that the killer NoSQL use case is
scalability for large data sets under high concurrency.   However, what is less talked about is the
fact that performance is often the key driver. Given a high performance
profile, NoSQL then provides the ability to scale that performance out linearly
by adding more servers. If not for the high performance profile to
begin with, often times an alternative technology like a distributed cache
would be used instead, fronting a more traditional database technology.

Given that performance is often a key driver, it’s no
surprise to find that flash based storage devices are increasingly common for
NoSQL deployments. In fact, popular
cloud providers of NoSQL services (Amazon
and Google) provide those
services exclusively on SSD backed infrastructure. The SSD form factor gives not only the best
performance, but also more manageable service level agreements, by taking disk
I/O bottlenecks out of the equation. At
Oracle, we are also seeing customers deploy the Oracle
NoSQL Database
on servers backed by SSD storage.

Unlike Oracle NoSQL Database, not all NoSQL Databases are
capable of taking advantage of SSD’s. Probably the biggest offenders are those
aligned with the Hadoop stack, based on HDFS which is using a large block based
scanning technique that is ill suited to SSD optimization. At Oracle, we’ve seen customer’s benchmark
against technologies like Hbase only to find that with large clusters, the cost
of deploying on non-SSD backed storage is quite simply cost prohibitive. In one customer benchmark for a mobile profiling
application, we saw a customer predicting the need for dozens of Hbase servers
to accommodate the same workloads that could be achieved on a standard high availability
configuration (3 servers) for Oracle NoSQL when backed by SSD. Forget about the cost of software folks
(Oracle NoSQL is practically free, providing annual subscription based support),
imagine having to buy 50 servers to stand up your NoSQL cluster just to run
what many see as “the Hadoop” software eco system. Seems to me people ought to be seriously re-thinking
their Big Data software choice when it comes to Big Data solutions that require
both low latency NoSQL and batch based Hadoop MapReduce.

SanDisk has recently benchmarked Oracle NoSQL Database using
the standard Yahoo
Server Scalability Benchmark
on its latest generation SanDisk CloudSpeed
SATA SSD’s. This makes for an
interesting read and highlights the advantages of using a NoSQL Database that
is designed to work with this latest generation of storage technology.


Be the first to comment

Comments ( 0 )
Please enter your name.Please provide a valid email address.Please enter a comment.CAPTCHA challenge response provided was incorrect. Please try again.