Welcome!

Microservices Expo Authors: John Worthington, Liz McMillan, Elizabeth White, Stackify Blog, Pat Romanski

Related Topics: Containers Expo Blog, Java IoT, Microservices Expo, @CloudExpo, @DXWorldExpo, SDN Journal

Containers Expo Blog: Article

SSD vs. HDD – The Battle for the Future of Storage

Look to the latest economically sensible storage solutions that are answering today’s performance and capacity calls

I have been working for technology companies for 33 years now, so I don't know why I'm always surprised at the technology myths that proliferate. For example, there is a popular notion running around that Solid State Drives (SSDs) will replace Hard Disk Drives (HDDs) as the dominant storage media. So, which will win? Let me give you my answer upfront before I take you through my arguments: both win, and the market size for both continues to grow.

As I offer up this data and this perspective, I'll repeat some advice I received early in my career. A mentor once told me, when searching for the truth in business and technology "look to the economics" (maybe my version of "go to the mattresses" for you "Godfather" fans).

Here is what I read and hear on a regular basis regarding the battle of HDDs and SSDs:

  • SSDs run circles around HDDs for performance
  • SSDs will soon replace HDDs
  • The improving density of SSDs will collapse the current cost premium to HDDs

Sound familiar? Well, as with most claims of this ilk, there are some elements of truth to them but each of these statements goes too far when viewed in light of some data.

Performance
Yes, it's true that SSDs are fundamentally faster than HDDs. This is particularly true of random I/O data reading (as with databases). With large block sequential data (i.e., rich media like video), however, this difference tends to be very small.

Replacing HDD?
Here's a claim where the data just doesn't add up. While the attractiveness of the SSD technology would lead one to believe this, the economics really don't hold up. I should probably point more specifically to NAND flash here (the "solid state" in SSDs) because another solid state technology might actually achieve this, but that appears to be years and tens of billions of dollars away.

Let's talk for a minute about the world's demand for storage. Last year approximately 2500 exabytes of data was created and/or replicated. And...this is doubling about every two years. This needs to be serviced largely by HDDs and SSDs. Last year, the NAND flash industry produced somewhere between 30 and 40 exabytes of storage; with somewhere around 3.5 exabytes finding its way into SSDs (the balance in phones, tablets, cameras, etc.). What is that, 1.4% of our total storage need? But production is ramping up... at a capital cost of about $1.5 billion per Exabyte (semiconductor fabs are expensive)! So, from an available supply perspective, SSDs replacing HDDs seems implausible.

Improving Technology Makes SSDs Cheaper
Agree. The NAND flash technology point is at 21nm line widths with plans to move to 19, then 16 nm. Storage density is further improving with the use of multi-level cell (MLC) capabilities vs single-level cell (SLC). This is bringing down the cost of solid state storage in much the same way the areal density increases seen in HDDs brought down the cost of hard drive storage.

Notice how I said "brought" down (i.e., past tense)? Because a real density growth in the hard drive industry has slowed to a crawl, the rapid erosion of cost/Gb has also slowed to a crawl. Now, the HDD industry needs to move to its next technology (HAMR?) to continue to take cost out and HAMR is a number of years away. HAMR will require significant capital investment by the HDD companies. Significant capital investment will be required anyway to keep up with storage demand (even more so with slowing areal density growth). This all spells out a flattening of HDD costs for the foreseeable future. Some would speculate that a renewed interest in improving the utilization of HDD capacity is an artifact of these economics.

Here's what's preventing the complete collapse of the price difference between the two technologies. The NAND flash suppliers have a similar problem as the HDD manufacturers... the implications of shrinking technology geometries. As the line widths shrink, the ability of flash to sustain multiple write-erase cycles declines. To make up for this deficiency, sophisticated error-correction algorithms and "brute force" overprovisioning (to allow some cells to wear out) are being used. This and the above mentioned semiconductor fab costs tend to mitigate the progress SSDs are making in closing the cost gap to HDDs.

Having made my arguments that SSDs won't take over the world, I have to say that IT solutions need this technology. I think you can see that SSDs aren't the answer to all our storage needs, but they allow us to address a crying storage requirement. At a high level, storage is called on to produce two key deliverables:

  • Make data available to an application or user in an appropriate timeframe (i.e., performance) and in today's environment this need is growing.
  • Store data reliably (i.e., capacity) and in today's environment this need is growing

By and large, to date, systems with hard drives have been architected to deliver both of these capabilities. It's hard to argue that hard drives haven't done a good job of delivering affordable capacity. But to deliver against the performance requirements of IT solutions, hard drives:

  • Have been developed with higher speeds, but arguably little progress has been made in the last ten years
  • Have been "short stroked" (a technique limiting the stroke of the actuator to improve performance)
  • Have been grouped together to allow striping of data across a large number of drives to aggregate their performance.

With each of these approaches the user suffers from higher power requirements, and with the last two approaches their system has been overprovisioned (leaving stranded capacity) to deliver performance. This all adds up to significant system cost that can be avoided with new storage architectures.

Here's where SSDs come in. SSDs are proving to be a technology answer to the new generation of storage needs, both the growing performance and the growing capacity requirements. Here's the simple, logical way to think about SSDs and HDDs and their role in storage solutions. Use the right tool for the job. That is, take advantage of SSDs for performance (particularly small block, random I/O) and HDDs for capacity.

You might challenge me now and ask, "Does adding SSDs (Flash) to IT solutions make economic sense?" SSDs are expensive, but, used appropriately, SSDs can also minimize the number of HDDs required in a given solution. The secret is in optimizing the use of both, that is, avoid overprovisioning of both. Thin provisioning of capacity has become popular as a cost saver. Similarly, thin provisioning of performance is a similar cost saver. (This is a benefit of virtualized caching or tiering capabilities of newer storage solutions.)

I'll close with a few proof points. Look to the latest economically sensible storage solutions that are answering today's performance and capacity calls. The vast majority of affordable "Ultrabooks" incorporate both flash and hard drive technology, as does Apple's latest "Fusion" drive.

More Stories By Tom Major

Tom Major is President of Starboard Storage. He joined Starboard from Seagate Technology, where he was senior vice president of Product Management and Business Operations. Before Seagate, Major worked at LeftHand Networks as chief strategy officer. Previous positions included vice president and general manager, Disk Business Unit, at StorageTek and vice president of Network Storage Marketing at HP, where he spent 21 years.

Comments (0)

Share your thoughts on this story.

Add your comment
You must be signed in to add a comment. Sign-in | Register

In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.


@MicroservicesExpo Stories
identify the sources of event storms and performance anomalies will require automated, real-time root-cause analysis. I think Enterprise Management Associates said it well: “The data and metrics collected at instrumentation points across the application ecosystem are essential to performance monitoring and root cause analysis. However, analytics capable of transforming data and metrics into an application-focused report or dashboards are what separates actual application monitoring from relat...
Agile has finally jumped the technology shark, expanding outside the software world. Enterprises are now increasingly adopting Agile practices across their organizations in order to successfully navigate the disruptive waters that threaten to drown them. In our quest for establishing change as a core competency in our organizations, this business-centric notion of Agile is an essential component of Agile Digital Transformation. In the years since the publication of the Agile Manifesto, the conn...
"Opsani helps the enterprise adopt containers, help them move their infrastructure into this modern world of DevOps, accelerate the delivery of new features into production, and really get them going on the container path," explained Ross Schibler, CEO of Opsani, and Peter Nickolov, CTO of Opsani, in this SYS-CON.tv interview at DevOps Summit at 21st Cloud Expo, held Oct 31 – Nov 2, 2017, at the Santa Clara Convention Center in Santa Clara, CA.
Enterprises are adopting Kubernetes to accelerate the development and the delivery of cloud-native applications. However, sharing a Kubernetes cluster between members of the same team can be challenging. And, sharing clusters across multiple teams is even harder. Kubernetes offers several constructs to help implement segmentation and isolation. However, these primitives can be complex to understand and apply. As a result, it’s becoming common for enterprises to end up with several clusters. Thi...
Let's do a visualization exercise. Imagine it's December 31, 2018, and you're ringing in the New Year with your friends and family. You think back on everything that you accomplished in the last year: your company's revenue is through the roof thanks to the success of your product, and you were promoted to Lead Developer. 2019 is poised to be an even bigger year for your company because you have the tools and insight to scale as quickly as demand requires. You're a happy human, and it's not just...
Cavirin Systems has just announced C2, a SaaS offering designed to bring continuous security assessment and remediation to hybrid environments, containers, and data centers. Cavirin C2 is deployed within Amazon Web Services (AWS) and features a flexible licensing model for easy scalability and clear pay-as-you-go pricing. Although native to AWS, it also supports assessment and remediation of virtual or container instances within Microsoft Azure, Google Cloud Platform (GCP), or on-premise. By dr...
While some developers care passionately about how data centers and clouds are architected, for most, it is only the end result that matters. To the majority of companies, technology exists to solve a business problem, and only delivers value when it is solving that problem. 2017 brings the mainstream adoption of containers for production workloads. In his session at 21st Cloud Expo, Ben McCormack, VP of Operations at Evernote, discussed how data centers of the future will be managed, how the p...
The benefits of automation are well documented; it increases productivity, cuts cost and minimizes errors. It eliminates repetitive manual tasks, freeing us up to be more innovative. By that logic, surely, we should automate everything possible, right? So, is attempting to automate everything a sensible - even feasible - goal? In a word: no. Consider this your short guide as to what to automate and what not to automate.
DevOps teams have more on their plate than ever. As infrastructure needs grow, so does the time required to ensure that everything's running smoothly. This makes automation crucial - especially in the server and network monitoring world. Server monitoring tools can save teams time by automating server management and providing real-time performance updates. As budgets reset for the New Year, there is no better time to implement a new server monitoring tool (or re-evaluate your current solution)....
"Codigm is based on the cloud and we are here to explore marketing opportunities in America. Our mission is to make an ecosystem of the SW environment that anyone can understand, learn, teach, and develop the SW on the cloud," explained Sung Tae Ryu, CEO of Codigm, in this SYS-CON.tv interview at 21st Cloud Expo, held Oct 31 – Nov 2, 2017, at the Santa Clara Convention Center in Santa Clara, CA.
The nature of test environments is inherently temporary—you set up an environment, run through an automated test suite, and then tear down the environment. If you can reduce the cycle time for this process down to hours or minutes, then you may be able to cut your test environment budgets considerably. The impact of cloud adoption on test environments is a valuable advancement in both cost savings and agility. The on-demand model takes advantage of public cloud APIs requiring only payment for t...
Many enterprise and government IT organizations are realizing the benefits of cloud computing by extending IT delivery and management processes across private and public cloud services. But they are often challenged with balancing the need for centralized cloud governance without stifling user-driven innovation. This strategy requires an approach that fundamentally reshapes how IT is delivered today, shifting the focus from infrastructure to services aggregation, and mixing and matching the bes...
"CA has been doing a lot of things in the area of DevOps. Now we have a complete set of tool sets in order to enable customers to go all the way from planning to development to testing down to release into the operations," explained Aruna Ravichandran, Vice President of Global Marketing and Strategy at CA Technologies, in this SYS-CON.tv interview at DevOps Summit at 21st Cloud Expo, held Oct 31 – Nov 2, 2017, at the Santa Clara Convention Center in Santa Clara, CA.
High-velocity engineering teams are applying not only continuous delivery processes, but also lessons in experimentation from established leaders like Amazon, Netflix, and Facebook. These companies have made experimentation a foundation for their release processes, allowing them to try out major feature releases and redesigns within smaller groups before making them broadly available. In his session at 21st Cloud Expo, Brian Lucas, Senior Staff Engineer at Optimizely, discussed how by using ne...
We just came off of a review of a product that handles both containers and virtual machines in the same interface. Under the covers, implementation of containers defaults to LXC, though recently Docker support was added. When reading online, or searching for information, increasingly we see “Container Management” products listed as competitors to Docker, when in reality things like Rocket, LXC/LXD, and Virtualization are Dockers competitors. After doing some looking around, we have decided tha...
It’s “time to move on from DevOps and continuous delivery.” This was the provocative title of a recent article in ZDNet, in which Kelsey Hightower, staff developer advocate at Google Cloud Platform, suggested that “software shops should have put these concepts into action years ago.” Reading articles like this or listening to talks at most DevOps conferences might make you think that we’re entering a post-DevOps world. But vast numbers of organizations still struggle to start and drive transfo...
While we understand Agile as a means to accelerate innovation, manage uncertainty and cope with ambiguity, many are inclined to think that it conflicts with the objectives of traditional engineering projects, such as building a highway, skyscraper or power plant. These are plan-driven and predictive projects that seek to avoid any uncertainty. This type of thinking, however, is short-sighted. Agile approaches are valuable in controlling uncertainty because they constrain the complexity that ste...
"This all sounds great. But it's just not realistic." This is what a group of five senior IT executives told me during a workshop I held not long ago. We were working through an exercise on the organizational characteristics necessary to successfully execute a digital transformation, and the group was doing their ‘readout.' The executives loved everything we discussed and agreed that if such an environment existed, it would make transformation much easier. They just didn't believe it was reali...
"We're developing a software that is based on the cloud environment and we are providing those services to corporations and the general public," explained Seungmin Kim, CEO/CTO of SM Systems Inc., in this SYS-CON.tv interview at 21st Cloud Expo, held Oct 31 – Nov 2, 2017, at the Santa Clara Convention Center in Santa Clara, CA.
The cloud revolution in enterprises has very clearly crossed the phase of proof-of-concepts into a truly mainstream adoption. One of most popular enterprise-wide initiatives currently going on are “cloud migration” programs of some kind or another. Finding business value for these programs is not hard to fathom – they include hyperelasticity in infrastructure consumption, subscription based models, and agility derived from rapid speed of deployment of applications. These factors will continue to...