SSDs vs. HDDs: Ten considerations


Like a phoenix rising from the ashes, solid-state disk (SSD) drives have re-appeared 20 years after their initial rise and fall. And if enthusiasm in the storage community is any indicator, it seems that SSDs are here to stay this time. For users who are deciding whether to buy SSDs, here are 10 key considerations:

1. SSDs reduce energy costs and support 'green' data centers.

Because SSDs do not have any moving parts, they use less energy than hard disk drives (HDDs). SSDs produce less heat and, therefore, require less cooling. Accordingly, SSDs make a valuable contribution to energy savings and green IT practices, although today the price per capacity is still much higher than HDDs. There are also variations in energy consumption among SSDs, depending on how the interface technology was implemented and what other components, such as DRAM cache, are used.

2. When it comes to price/capacity, HDDs are still first choice.

Though capacities go up and prices go down for all types of storage devices, SATA drives are still cheapest in terms of $/GB—and the point in time where SSDs will achieve price-per-capacity parity is still not in sight.

3. When it comes to price/IOPS, SSDs could be the best choice.

With no moving parts, latency on SSDs is very low. In an environment with random access to large data sets, SSDs perform very well. They can also deliver blazing read throughput; however, writing to an SSD can be relatively slow because of the way SSDs operate—they have to erase the content of a cell before writing new content. Enter- prise-class SSDs speed up write operations by using DRAM as a cache, and typically perform faster than HDDs. Though they are still expensive, SSD $/IOPS ratios are better than HDDs.

4. For SSDs, form factor loses importance.

Today, flash chips are built into disk frames to support the same form factor as hard disks, so they can be used in current server and RAID array designs. Like HDDs, SSDs are available in either 2.5- or 3.5-inch form factors. Unlike HDDs, where form factors have a direct effect on price and available capacity, this is not a very important criterion for SSDs.

5. Write operations on SSDs are complex and, therefore, slower.

While the read performance of a flash chip is typically faster than that of HDDs, write performance is often slower. This is because on a flash chip, you do not just write one block; the block first has to be erased, and you cannot erase single blocks—only a number of blocks at the same time. This group of blocks is called the "erase block," and it can be very large and may contain data that is still needed, which makes the write operation complex.

Enterprise-class SSDs overcome this limitation in a number of ways (e.g., by using a cache to aggregate write operations). As changing all blocks in an erase block is almost as fast as changing one, using a cache can help significantly; however, if the blocks are spread throughout different erase blocks, caching may not solve the problem. Another way to improve write performance is to address a number of flash chips in parallel, which also means writing to different erase blocks at the same time, converting sequential write operations into parallel write operations.

When it comes to performance, write operations are the Achilles' heel of flash drives, and this is where the difference between consumer-class and enterprise-class SSDs is most apparent.

6. There are limits to the durability of SSDs.

A block can be written on a flash chip typically no more than 10,000 to 100,000 times, depending on the type of chip. This is not much for a storage system that must deliver high I/O loads, especially as failing blocks impact the lifetime of the entire flash chip. So, SSDs have to make sure that all blocks are used in a balanced way through a process called "wear levelling" (writing to the blocks that have been used least). Another strategy to increase the life of flash chips is to provide overhead capacity sufficient to replace defective blocks, which requires an intelligent way of detecting and managing defective blocks.

7. There are important differences between MLC and SLC chips.

Single-level cell (SLC) chips store one bit per cell, and multi-level cell (MLC) chips store more (typically three). The negative aspects of MLC are that if one bit fails in a three-bit cell, all three stop working. MLCs are also slower than SLCs in both read and write operations, and they use more power. But MLCs offer more capacity at almost the same manufacturing cost as SLCs, and MLC chips have better $/capacity ratios.

8. SSDs reduce the demand for physical storage extensions, but won't replace external arrays.

To achieve high IOPS with HDDs, you need a lot of them—more spindles equals more speed. As servers cannot physically house numerous spindles, the use of an external RAID array makes sense; however, now that two SSDs running in a RAID-1 configuration can replace several HDD spindles, and because two drives fit easily into any server, it may appear as though RAID arrays for high-performance systems will become obsolete because of SSDs. But RAID will not become obsolete because external arrays still offer the benefits of consolidation; many servers can access them, and consolidation dramatically decreases unused spare capacity. Also, many applications require RAID-protected storage, and they can’t run on internal or JBOD storage. Though SSDs will reduce the demand for simple physical storage extensions (i.e., JBOD arrays), their weakness is still in price per capacity.

9. SSDs enable storage systems with much higher density.

A flash chip is small, and though today it is typically built into a disk frame, this won't necessarily be the case in the future. And because flash chips use little power, they don't get hot and, therefore, need less cooling than HDDs. So, it is possible to imagine storage devices with new form factors and a tremendous increase in density. Capacity has been roughly doubling each year while the price per GB has declined. Flash technology is good for enabling small storage devices with high capacity and tremendous performance that, in the future, will look completely different from today's RAID systems.

10. The choice between SSDs and other storage media is usage-dependent.

As usual in storage environments, the ideal infrastructure depends on the specific scenario. If the primary aim is to store huge amounts of data that is rarely accessed, SSDs are not the best choice and will not be in the near future. If speed is the priority, SSDs could be the right option, although write performance is still a critical factor. So, where an application primarily requires fast reads, consider SSDs. If there is a large amount of writes, SSDs could be too slow.


MARCUS SCHNEIDER is on the Storage Networking Industry Association Europe (SNIA-E; www.snia-europe.org) board of directors, and is the European liaison to the SNIA Solid State Storage Initiative.

This article was originally published on April 01, 2009