Solidigm D5-P5336 61.44TB SSD Review - Size Matters

TweakTown's Rating: 95%

The Bottom Line

Truly capacity is king, and Solidigm's D5-P5336 is the king of capacity.

Pros

• + Pure read workloads
• + Random read
• + Sequential read

Cons

• - Writes

Should you buy it?

AvoidConsiderShortlistBuy

Buy at Amazon

Buy at Newegg

Introduction and Drive Details

As of late, we've seen a trend toward higher and higher capacity SSDs in the enterprise space. Most of the top suppliers have at least one version of their datacenter series SSDs available at 30.72TB. In fact, Micron's newest datacenter SSD is only available in 30.72TB. This is a testament that density has become the primary consideration for most enterprise storage roles.

This makes perfect sense for several reasons beginning with footprint. If your server can hold double the data while maintaining the same footprint and a similar power draw, TCO falls dramatically. TCO, or Total Cost of Ownership, is really the bottom line for any sizeable enterprise storage application. Additionally, modern workloads are becoming more data-hungry much of which is AI-driven. In fact, there is no end in sight for growth in data services and applications. So, in a nutshell, density is king.

So what we need is the ability to store and access massive amounts of data no matter where it lives. This is how Solidigm sees the future of big data and is responding accordingly with the world's highest capacity PCIe SSD, the 61.44TB D5-P5336. Solidigm's D5-P5336 is part of Solidigm's third-generation QLC SSDs for the datacenter, delivering an industry-leading blend of capacity, read-optimized performance, and value-to-read and data-intensive workloads.

Solidigm describes its newest datacenter SSD as follows:

Combining read performance exceeding that of some cost-optimized TLC SSDs and capacities up to 61.44 TB built on industry-leading NAND density, the D5-P5336 has been architected to efficiently accelerate and scale the increasingly massive datasets found in widely deployed, modern read-intensive workloads while increasing storage density, reducing total cost of ownership (TCO), and enabling a more sustainable storage infrastructure.

The modern datacenter has evolved into more of a consumer-like data usage pattern in that the vast majority of data is read frequently, write infrequently. This is exactly what our QLC arrayed test subject is designed to do. Store massive datasets and provide access to that data at latencies rivaling that of lower capacity, higher cost TLC SSDs. The direct TCO savings benefits the 61.44TB D5-P5336 has on offer over TLC arrays include the following: Up to 2x fewer servers, up to 20% lower five-year energy cost, and up to 17% lower overall TCO.

Solidigm's D5-P5336 at 61.44TB sounds especially compelling to us. Now let's see exactly what kind of performance the world's highest capacity PCIe SSD has on offer.

Specs/Comparison Products

34

34

Solidigm offers its D5-P5336 series at capacity points ranging from 7.68TB -61.44TB across three form factors, including 2.5-inch U.2, E3.S, and E1.L 9.5mm.

Solidigm D5-P5336 61.44TB NVMe PCIe Gen4 x4 U.2 SSD

34

The label on our engineering sample says Intel, but the drive reads correctly as Solidigm. Notice we've written more than 1.38 petabytes to the drive, and it still shows 99% health remaining.

Enterprise Testing Methodology

TweakTown strictly adheres to industry-accepted Enterprise Solid State Storage testing procedures. Each test we perform repeats the same sequence of the following four steps:

1. Secure Erase SSD
2. Write the entire capacity of SSD a minimum of 2x with 128KB sequential write data, seamlessly transition to the next step
3. Precondition SSD at maximum QD measured (QD32 for SATA, QD256 for PCIe) with the test-specific workload for a sufficient amount of time to reach a constant steady-state, seamlessly transition to the next step
4. Run test-specific workload for 5-minutes at each measured Queue Depth, and record results

Solidigm D5-P5430 15.36TB SSD

	Today	Yesterday	7 days ago	30 days ago
	$1013.78	$1013.78	$1013.78	$1013.08	Buy
	-	-	-	-	Buy
* Prices last scanned on 9/30/2023 at 1:24 am CDT - prices may not be accurate, click links above for the latest price. We may earn an affiliate commission.

Benchmarks - Random and Sequential

4K Random Write/Read

34

We precondition the drive for 160,000 seconds, receiving performance data every 10 seconds. We plot this data to observe the test subject's descent into steady-state and to verify steady-state is in effect as we seamlessly transition into testing at queue depth. Steady-state is achieved at 120,000 seconds of preconditioning. The average steady-state write performance at QD256 is approximately 48K IOPS.

34

34

Keeping in mind that this is a read-exclusive focused SSD, we are just fine with the 48K IOPS our test subject is delivering. Full speed at QD1 is something we like, as is the fact that we are getting better than the drives up to random write spec and further that we are doing so with 4K aligned data as opposed to the quoted 43K IOPS for 16K aligned data.

34

34

Our test subject delivers more across the board than its recent predecessor, the D5-P5430, and ends up tying the output of Micron's 7450 Pro at QD256. Can our QLC contender deliver equivalent 4K random read performance to some of its TLC competition? Yep.

8K Random Write/Read

34

We precondition the drive for 160,000 seconds, receiving performance data every 10 seconds. We plot this data to observe the test subject's descent into steady-state and to verify steady-state is in effect as we seamlessly transition into testing at queue depth. Steady-state is achieved at 80,000 seconds of preconditioning. The average steady-state write performance at QD256 is approximately 38K IOPS.

34

34

We expect 8K random to track pretty much the same as 4K random here, just at a lower rate because it's moving twice the amount of data. Again we are fine with the results here for the same reason stated previously.

34

34

Okay, we find the results here very impressive. Intel QLC SSDs have shown a real affinity for reading 8K aligned data, and our test subject is the best example of this so far. When reading 8K aligned data, the D5-P5336 does it better than Micron's TLC arrayed ION 6500 and 7450 Pro at QD1-2 & QD128-256.

128K Sequential Write/Read

34

We precondition the drive for 65,000 seconds, receiving performance data every 10 seconds. Steady-state for this test kicks in at 0 seconds. The average steady-state sequential write performance at QD256 is approximately 3,360 MB/s.

34

34

Full speed at QD1 is what we like to see from any SSD. Additionally, the drive outperforms its predecessor, the QLC arrayed D5-P5430, across the board. Better than expected, especially from a drive with a 16K alignment.

34

34

Outstanding performance here, as we see it. At QD1, our capacious contender is the third best-performing SSD in our test pool. At queue depths of up to 8, it is the best-performing of the Solidigm SSDs that appear in our test pool, including the TLC arrayed D7-P5520. At queue depths of 32 or more, our test subject outperforms the TLC arrayed ION 6500. In fact, at queue depths of 32 or more, the D5-P5336 is dishing up performance that is as good as it gets for a PCIe Gen4 x4 SSD. Impressive.

Benchmarks - Workloads

4K 7030

4K 7030 is a commonly quoted workload performance metric for Enterprise SSDs.

34

We precondition the drive for 160,000 seconds, receiving performance data every 10 seconds. We plot this data to observe the test subject's descent into steady-state and to verify steady-state is in effect as we seamlessly transition into testing at queue depth. Steady-state is achieved at 80,000 seconds of preconditioning. The average steady-state performance at QD256 is approximately 149K IOPS.

34

34

4K aligned writes are the nemesis of our test subject, as demonstrated by its performance curve here. Performance at QD1 is decent, but otherwise, for this workload mix, the D5-P5336 is not the best solution unless capacity considerations happen to outweigh performance considerations.

Email Server

Our Email Server workload is a demanding 8K test with a 50 percent R/W distribution. This application gives a good indication of how well a drive will perform in a write-heavy workload environment.

34

We precondition the drive for 160,000 seconds, receiving performance data every 10 seconds. We plot this data to observe the test subject's descent into steady-state and to verify steady-state is in effect as we seamlessly transition into testing at queue depth. Steady-state is achieved at 80,000 seconds of preconditioning. The average steady-state performance at QD256 is approximately 74K IOPS.

34

34

As we saw previously, our test subject digests 8K aligned data significantly better than 4K aligned. Because of this, our test subject is able to deliver better QD1-2 performance than its rival, the ION 6500.

OLTP/Database Server

Our On-Line Transaction Processing (OLTP) / Database workload is a demanding 8K test with a 66/33 percent R/W distribution. OLTP is online processing of financial transactions and high-frequency trading.

34

We precondition the drive for 160,000 seconds, receiving performance data every 10 seconds. We plot this data to observe the test subject's descent into steady-state and to verify steady-state is in effect as we seamlessly transition into testing at queue depth. Steady-state is achieved at 80,000 seconds of preconditioning. The average steady-state performance at QD256 is approximately 106K IOPS.

34

34

Again, thanks to its affinity for 8K aligned data, our test subject is able to deliver better QD1-2 performance than its rival, the ION 6500.

Web Server

Our Web Server workload is a pure random read test with a wide range of file sizes, ranging from 512B to 512KB at varying percentage rates per file size.

34

We precondition the drive for 160,000 seconds, receiving performance data every 10 seconds. We plot this data to observe the test subject's descent into steady-state and to verify steady-state is in effect as we seamlessly transition into testing at queue depth. We precondition for this test with an inverted (all-write) workload, so no relevant information can be gleaned from this preconditioning other than verification of steady-state.

34

34

We consider this test to be the most taxing test we run, and as such, we are quite impressed with what our test subject is delivering here. Our read-focused contender handles this 100% read test like a champ, this time outperforming about half of the SSDs in our test pool at queue depths of 64 or more. This is, as we see it, absolute verification that Solidigm's contention that its QLC-arrayed D5-P5336 compares favorably with TLC-arrayed SSDs in a read-intensive environment is factually accurate.

Final Thoughts

As we see it, even more so than in the consumer space, capacity is king in the datacenter, or even at the edge. This is why there is a massive upward trend in capacity as of late. Big Data needs big capacity not only for the huge datasets being generated every second of every day at an exponential rate but for sustainability as well. At some point, capacity, more often than not, takes precedence over performance. We are at that point and have been there for some time now. This is why Solidigm's D5-P5336 61.44TB is such an important milestone for the solid state storage industry.

34

With double the capacity of its next closest rival and read performance rivaling that of TLC, Solidigm's QLC arrayed D5-P5336 has earned our highest award. Editor's Choice.