Real Results: Seagate Pulsar XT.2
Why You Need a Reliable Storage Benchmark
What's this
All of this theory is well and good, but perhaps the best way to illustrate what the SPC provides to the world is to examine some of its results. One of the most recent products tested by SPC analysts is Seagate’s SLC-based Pulsar XT.2 solid state drive. This is a 400 GB model using a 6 Gb/s SAS interface. To illustrate, we’ll highlight some of the key points from the drive’s 64-page SPC-1C test report.
After a couple of preliminary letters from the SPC and the Test Sponsor (Seagate in this case), the report gets down to business with an executive summary of its findings.

In a nutshell, we see that the Pulsar XT.2 delivered a consistent 20,0008.82 IOPS under 100% load. The total storage area written during the test amounted to 399.931 GB, and the SPC mandates that there must be either no unused storage or that the utilized space is within 1 GB of 50% of the drive’s total capacity. (Adding in overhead, the used total reached the drive’s maximum of 400.088 GB.) Fifty percent utilization would likely indicate a short-stroked drive, which can help improve performance but will obviously alter the drive’s IOPS per dollar ratio.
A single drive does not utilize any measures for data protection in the event of failure, such as a RAID configuration. Because SPC-1C is targeted at single devices, but it also accommodates arrays. SPC-1C amendments are on the table that would include more complex RAID protection levels.[41]
The price for the tested storage subsystem, including the Pulsar, was $6,312.60. This price doesn’t include the motherboard, graphics card, or any other host components. The SPC report goes on to provide line items for the drive, SAS controller, cable, and adapter, and it verifies that there were no differences between the tested storage configuration and that shown in its pricing breakdown.

The SPC requires that drives not exceed a 30 ms response time and deliver a linear response distribution free from significant spikes or dips. The above chart clearly shows this. You can also see how the drive responds to increasing traffic load, which is an observation often missing in conventional benchmarking results.
The report next delves into several pages covering the host system configuration and SPC-1C test dynamics. Next, the specific tests run are itemized.
Primary Metrics Test
Sustainability Test Phase and Test Run
IOPS Test Phase and Test Run
Response Time Ramp Test Phase
95% of IOPS Test Run
90% of IOPS Test Run
80% of IOPS Test Run
50% of IOPS Test Run
10% of IOPS Test Run (LRT)
Repeatability Test
Repeatability Test Phase 1
10% of IOPS Test Run (LRT)
IOPS Test Run
Repeatability Test Phase 2
10% of IOPS Test Run (LRT)
IOPS Test Run
Data Persistence Test
Data Persistence Test Run 1
Data Persistence Test Run 2
The SPC provides all four hours of data rate readings (plus a six-minute ramp-up) broken down into one-minute measurements followed by a composite graph:
