The right way to use SSDs in your storage environment

The increasing availability of cheap, fast SSDs (solid-state drives) in the enterprise primary storage market is exciting, to say the least. SSDs offer tremendous transactional throughput and extremely low latency — neatly solving two problems that plague many storage environments. However, they still come at a premium over traditional spinning disks and don’t yet offer the same enterprise-ready high-density capacity. Knowing where to apply SSDs in your environment is critical to a cost-effective deployment.

The most common application for high-performance flash-based SSDs is in OLTP (online transaction processing) environments and high-density VDI farms. These types of workloads generally consist of extremely large amounts of relatively small, entirely random reads and writes. This mix is very difficult for traditional spinning disk to keep up with.

The randomized nature of OLTP workloads renders most controller and on-disk caches useless, which fully exposes the relatively high seek time of the mechanical disk heads that are being used. The most common solution to this problem is to use a large number of high-performance disks to satisfy the workload. By increasing the number of disk spindles that the data set is spread across, the storage infrastructure can increase the overall number of transactions it can process.

Avoiding inefficiency

However, this approach can be tremendously wasteful. Let’s say, for instance, that you have a 1TB database that must be able to respond to 10,000 disk transactions per second in order to meet your users’ demands. To deliver that kind of performance, you’d need about 60 active 15,000-rpm SAS or Fibre Channel disks dedicated to that workload. Assuming that fairly common 300GB disks are used, you’ve just effectively deployed 18TB of raw disk to host 1TB of data — a massive capacity inefficiency. Despite its wastefulness, this solution is so common that it actually has a name: short-stroking.

Using SSDs can effectively do away with the need to short-stroke large amounts of spinning disk. Instead of deploying lots of traditional disks and only using a small percentage of their capacity, you can deploy a fairly small number of SSDs, get the performance you need and not waste capacity (not to mention money, power, and rack space).

Striking a balance

However, not every workload is this cut-and-dried. Most real-world storage environments don’t just cater to a single high-performance workload. Instead, they often must serve the needs of many smaller applications with varying workloads and capacity requirements. While you could blindly deploy SSDs for everything, you’d likely end up buying a significant amount of unneeded transactional performance just to meet your capacity requirements. Essentially, this is no better than short-stroking.

For as long as SSD lags behind traditional disk in cost per gigabyte of capacity, it will be necessary to balance transactional performance requirements, capacity requirements, and cost against each other to determine the best fit for a given set of workloads.

This balance is a bit difficult to see at first glance, but some quick spreadsheet work can illustrate the balance as it relates to cost. I did this by comparing one enterprise vendor’s street prices for SATA, 10K SAS, 15K SAS, and SLC SSD SAN arrays against their respective capacities and expected performance. In this instance, I found that cost per raw gigabyte of capacity ranges anywhere from around $1.50 for SATA to $28 for SSD. Likewise, cost per I/O per second ranges from $29 for SATA to $5 with SSD. In the comparison, SAS/Fibre Channel ends up somewhere in the middle of the capacity-versus-performance spectrum, though much closer to SATA than to SSD.

If you’ve properly tiered data sets with similar capacity and performance characteristics together, determining whether SSD is a good fit for any given tier is fairly straightforward.

The hybrid option

If past history is any guide, SSD will likely overtake traditional spinning disk as the most popular primary storage media in the long term. However, in the near term there will be an increasing number of hybridized SSD/disk solutions that will attempt to bridge the capacity-versus-performance gap that lies in between the two.

One great example of this is Oracle’s Sun Open Storage NAS platform. For the past few years, Sun (and now Oracle) has offered this traditional disk storage platform with a write-biased SSD called a Logzilla that stores the system’s write-intensive ZIL (ZFS Intent Log). Likewise, a read-biased SSD can act as an enormous extension of the controller’s read cache. Though not yet common, this hybridized SSD and disk architecture can significantly magnify spinning disk performance without the high cost of a full SSD deployment.

Another type of hybridization is starting to happen at the level of the individual disk. Seagate recently announced the Momentus XT solid-state hybrid drive, which combines 4GB of SLC NAND memory and a traditional SATA hard disk in the same disk package. By using the SSD space to store the most-often-used data, the disk can appear to dramatically outperform a standard SATA disk while costing only marginally more. Although this product is targeted at the consumer space, you can expect this type of hybridization to find its way into some enterprise products as well.

Putting it all together

While SSDs have some amazing advantages over spinning disk, capacity and cost are still significant limiting factors to widespread adoption. Just remember to do the math to ensure that they’re really the best choice for your workloads before you commit yourself to deploying them.

This article, “The right way to use SSDs in your storage environment,” originally appeared at InfoWorld.com. Read more of Matt Prigge’s Information Overload blog and follow the latest developments in data storage and information management at InfoWorld.com.