Page 1:Seagate's 600 Pro SSD: Enterprise On A Budget
Page 2:Inside Seagate's 600 Pro SSD
Page 3:Test Setup, Benchmarks, And Methodology
Page 4:Results: Write Endurance
Page 5:Results: 4 KB Random Performance And Latency
Page 6:Results: Performance Consistency
Page 7:Results: Enterprise Workload Performance
Page 8:Results: Sequential Performance
Page 9:Results: Enterprise Video Streaming Performance
Page 10:Can An Old-World Storage Vendor Compete In The SSD Space?
Results: Performance Consistency
Increasingly, we pay close attention to the performance consistency of enterprise-class SSDs. This is what separates a good drive from a great one when all of the corner case testing seems equal. Over the past year, we measured this in terms of large block transfers in our Enterprise Video Streaming section. Armed with this data and our exclusive analysis, the peaks, valleys, and frequency of each become clear. If you look at the information for long enough, you start to see fingerprints for each drive.
We started with large block transfers because, in enterprise video applications, if you don't buffer or write data fast enough, you can lose it completely. Random 4 KB transfers are slightly more academic, but they also match database transfers more closely. With this sort of workload, you might not lose data, but the system will slow down.
For the following tests, we subjected Seagate's 600 Pro, along with Micron's P400m and Intel's SSD DC S3700, to 25 hours of continuous random 4 KB writes across each entire drive. We recorded the IOPS every second, giving us 90,000 data points. We then zoomed in to the last 60 minutes to more coherently visualize the results.
As you can see, the P400m and SSD DC S3700 are stellar performers, which we already knew going in. The 600 Pro's results surprised us a little, though. We didn't necessarily expect Seagate's new SSD to match the higher-priced competition. However, the amount of variance it demonstrates is quite high.
More positively, the 600 Pro posted at least 40,000 IOPS (0.8 ms or less response time) on 25% of its transfers. But, as the histogram also shows us, almost 10% of the drive's transfers took more than 3 ms (less than 11,000 IOPS). Some of the other SSD vendors limit raw I/O performance to maintain more consistency; this does not appear to the case here, though. If you watch Iometer as the test runs, its variability is apparent. The value of the last update jumps between 8,000 and 40,000+ IOPS.
We did notice that if you limit your write area to 100 GB, manually over-provisioning, in essence, the variability disappears. This makes sense because, as the number of spare cells increases, the controller manages their use more effectively. This also tells us that the amount of spare capacity and the memory's performance is responsible for the ups and downs, rather than an issue with the controller architecture. Interestingly, Seagate rates its 240 GB 600 Pro 11,000 4 KB random write IOPS, which is almost exactly where that top distribution centers around. Unfortunately, since we only have the 200 GB version to test, we couldn't double-check those numbers.
Now, to put all of this information into perspective. If you're looking at worst-case scenarios, Intel's SSD DC S3700 is the way to go; even Micron's P400m gives you great consistency. At its worst, the 200 GB 600 Pro is a 10,000-IOPS drive. But at best, expect to see more than 40,000 IOPS. If your application writes in bursts, giving the 600 Pro time to recover, it is an excellent option.
- Seagate's 600 Pro SSD: Enterprise On A Budget
- Inside Seagate's 600 Pro SSD
- Test Setup, Benchmarks, And Methodology
- Results: Write Endurance
- Results: 4 KB Random Performance And Latency
- Results: Performance Consistency
- Results: Enterprise Workload Performance
- Results: Sequential Performance
- Results: Enterprise Video Streaming Performance
- Can An Old-World Storage Vendor Compete In The SSD Space?