Rethinking your server storage
Between the creation of the earth and 2003, humans produced five extabytes of data, total. By 2013, we were creating five extabytes each day.
Unfortunately, many companies continue to store their data as if it were still being produced at the same rate it was 10 years ago, when it was just a fraction of what it is now. That poses a problem.
At this point, a lot of companies still use three-tier hardware, which separates out the compute, network, and storage so that they can be managed in pieces. This approach is comfortable, and has been the norm for decades. But now there are potentially better storage options available.
As the explosion of data continues, it might be time to shift the focus off of three-tiered hardware, and explore some of the newer options that could better fit your needs.
Server storage options for every business
Three additional kinds of server storage have been developed to suit the needs of modern data: converged, hyperconverged, and software-defined storage (SDS). Which option you choose will depend on your needs, but in many cases, moving away from three-tiered could add efficiency, scalability, and flexibility.
First, let’s take a look at what each type of storage offers:
- Traditional three-tier hardware: The compute, storage networking, and the storage itself each run independently of the other two. If your system needs one specific function more than another, such as ingesting a lot of information into deep storage without much need to access it, three-tiered can allow for the storage to grow faster than the other tiers. But this is the exception rather than the rule in most cases.
- Converged: Practically, this means that all three tiers are purchased at once in a rack. From an operational standpoint, they work the same as traditional three-tier storage. However, from a purchasing standpoint, they are “converged.” This makes it easier to implement and integrate than buying three separate components and can make sense for larger organizations purchasing in larger units.
- Hyperconverged: By combining the three tiers into a single unit, they can be managed together from one point, rather than three, reducing the personnel you need to operate them. If you’re more comfortable with a common toolset to manage your storage, hyperconverged can offer that. It can be more expensive to start up, but with typically lower maintenance costs, more scalability, and more efficiency, it has a better cost of ownership over its life.
- SDS: SDS is virtualized storage that’s similar to hyperconverged, except the compute lives where it does traditionally, instead of in the same place as the storage. The hardware itself is general-purpose, which means the storage can be flexible depending on the needs of your company. That can lower costs and streamline procedures.
Because of the way they combine different functions in the same space, hyperconverged and SDS are easier to scale and manage than converged or regular three-tier hardware. They use general-purpose X86 hardware and disk drives in combination with smart software, which greatly simplifies and centralizes the entire storage process.
Determine your storage based on your needs
Hyperconverged and SDS offer so much scalability and efficiency in the face of the vast amounts of data being produced that instead of asking why you should use them, it may be a better idea to ask why you shouldn’t use them.
Of course, each situation is unique and you might have solid reasons for choosing the older three-tiered or converged. However, in most cases, hyperconverged and SDS would be an improvement and a better investment.
As always, consider all your options and consult with your account executive before deciding on what kind of server storage is best for you.
Joe Murphy, Lenovo Senior Systems Engineer, contributed to this post. Joe supports Lenovo’s complete data center portfolio, including hyperconverged offerings. He joined Lenovo in 2014 after more than 25 years at IBM and has been supporting SHI for over a year.