The TidalScale Blog

Why 2018 is the Year of the Software-Defined Server


With data volumes growing at a whopping 62 percent CAGR, it’s easy to see why some organizations are worried about keeping up. They’re trying to process and analyze bigger and more complicated problems, which increasingly stresses their computing resources.  Eventually, you hit the limits of what your systems can handle.

The solution to this hasn’t come easily. Data scientists and others know that big problems run a lot faster in memory—1,000 times faster, in fact, than flash storage. So more memory leads to faster results.  But memory is expensive, not just in terms of hardware, but also when it comes to software licensing costs.

Add to this the simple truth that businesses today can only succeed if they adapt quickly to market conditions, and this ties directly into their ability to gain insight for ever-growing mountains of data.

Today’s companies, then, face a harsh calculus: Either invest a lot of money to arrive at results sooner, or accept slower insights that could threaten your competitiveness.

This year, more organizations will find that their solution to the limits of hardware isn’t hardware at all. Their solution is the Software-Defined Server.

A Software-Defined Server combines all the resources of multiple commodity servers into one or more virtual servers.  Applications and operating systems need no modification, and they think they’re looking at an actual single hardware system.  It’s almost ludicrous in its simplicity.  But as Timothy Pickett Morgan noted in an article in TheNextPlatform, many worthy brains have worked for years to achieve “a big ole flat memory space that is as easy to program as a PC but brings to bear all that compute, memory and I/O of a cluster as a single system image.”   That is, until TidalScale and the Software-Defined Server came along.

I’ll argue here that 2018 promises to be the year of the Software-Defined Server.  Here’s why.

Users Need Results Sooner

Rapid insight is a business imperative. This means organizations don’t have time to wait for the next big system to be aquired or for applications to be rewritten so they can run across clusters.

With TidalScale WaveRunner, our point-and-click control panel, virtually any user can combine resources from a pool of commodity servers – memory, cores, storage and network – into one or more systems that are right-sized to a particular workload. You can boot a new system in just a few minutes, getting you productive right away. Think about that: In as little as five minutes, you can configure and boot up a completely new system that five minutes earlier didn’t even exist in your datacenter. And you can do this again and again as your workloads intensify or change. To save even more time, there is no recoding, no sharding of data.  

And because applications run entirely in memory, you’ll get your results sooner. How much sooner? Depending on the application, performance on TidalScale can improve by a factor of 300X.

IT Needs to Lower Costs

The traditional scale-up model of buying new hardware just doesn’t work for many organizations. For one thing, a lot of them can’t afford it. And even if you can, there’s the often futile exercise of trying to anticipate your needs three years from now so you can overprovision today in anticipation of having enough iron to throw at the problems you think you’ll face in 2011. But who can possibly predict what they’ll face in a year, let alone three?

To make matters worse, traditional software licensing is usually based on cores. That causes organizations running memory-intensive applications to essentially overpay for cores they’re not even using just to access the amount of memory they need for their problem. (Even if you buy a large scale-up server, hardware vendors typically tie large memory with high processor core counts, along with their associated low operating frequencies—a combination that is not always optimal. So you end up overpaying twice.)

TidalScale turns these old economics on their ear by enabling memory scaling with low core-count servers. The result?  TidalScale reduces license costs for common core-based software licensing models.

To further derive the greatest ROI from their investment and to drive down operating costs, many IT organizations are implementing a more composable infrastructure approach, with an eye toward creating a Software-Defined Data Center (SDDC). Software-Defined Servers are taking a more prominent role in this trend, increasingly seen as the missing piece of the SDDC. And thanks to our open source RESTful API, TidalScale WaveRunner is also an excellent control panel for monitoring and managing not just right-sized servers, but all software-defined resources in the data center.

Building on Momentum

This will also be a big year because Software-Defined Servers entered 2018 on a roll, riding some impressive momentum from the past year.  

We earned broad industry recognition by analyst firms and publications:

We also extended the market reach of Software-Defined Servers via strategic partnerships with several industry leaders and innovators.

  • Our October announcement with Oracle Cloud Infrastructure launched an alliance to enable the world’s largest cloud servers. Oracle Cloud customers can now combine Oracle bare metal servers into single system images of virtually any size.
  • Two industry giants – SAP and InfoSys – have partnered with us to bring to organizations around the world the breakthrough ability to right-size resources on the fly.
  • On-demand infrastructure powered by TidalScale is now available from OrionVM’s Infrastructure as a Service (IaaS) offering and IBM’s BlueMix cloud platform.

Add to this our expansion to Asia, and TidalScale’s Software-Defined Servers really hit their stride last year.

We expect only greater things in 2018. By addressing two crucial needs – delivering results faster and driving down costs – Software-Defined Servers are helping organizations push past the physical limitations of their own data centers and helping them achieve things they once believed were impossible.

Watch us for more news and more breakthroughs as we and our cusotmers make 2018 the year of the Software-Defined Server.

Topics: software-defined data center, software-defined server, software-defined storage, Oracle Cloud Infrastructure, OrionVM, IDC, Gartner, TidalScale WaveRunner

Sign up for weekly updates!

Recent Posts:

gartner_cool_vendor_2017
IDC-Innovator-Logo-2017
e-week-promo
red-herring-nm-winner
cube-interview-graphic