Do You Prefer Your IOPS Without Extra Latency?

This blog is part of a series on Data ONTAP 8 clustering, and provides more analysis of a recent SPC-1 performance benchmark.

 

Data ONTAP 8.1 blog series:

1.       Why Scale-Out & What’s New

2.       Unified Scale-Out – What’s Supported & Why It’s Unique

3.       Nondisruptive Operations – What Does it Mean?

4.       Enterprise Ready Data ONTAP 8 Cluster-Mode

5.       What’s Included In A Cluster Configuration

6.       Data ONTAP 8 Clustering – Multitenancy Designed In

7.       ESG Lab Validation of Data ONTAP 8.1 Unified Clustering

8.       Storage Monitoring Made Easy:  A Video Demonstration of Data ONTAP 8 Clustering

9.       Data ONTAP Unified (SAN and NAS) Cluster Benchmark Performance

10.    Announcing Data ONTAP 8.1.1 operating in Cluster-Mode

11.    PeakColo Guarantees 100% SLA for the Cloud with Data ONTAP 8 Clustering                               

12.    Data ONTAP #1 Storage Operating System

 

I posted another blog here that discussed our recent NAS and SAN cluster benchmark performance.  This analysis delves into the SPC-1 results in a little more detail. 

We performed an analysis of SPC-1 results to compare multiple disk-based systems—highly reliable, general-purpose systems that can provide high performance, low latency and high capacity—based on a response time threshold of approximately 3 milliseconds. This was done because, for a great majority of database workloads, very low I/O latencies are vastly preferred to higher latencies. The analysis shows that the NetApp SPC-1 results are among the best for enterprise disk-based systems, given the low latency delivered for the IOPS provided. See the figure 1 table below for details. 

Figure 1

 

A similar analysis of SPECsfs results showed that unlike scale-out solutions optimized specifically for sequential throughput, a NetApp storage cluster running the NFS protocol provides a predictable latency behavior that makes it suitable for a wide range of workloads, including databases and highly virtualized environments.

 

Mike McNamara