UPDATED 23:42 EDT / APRIL 03 2014

MySQL + flash, and laser-focused KPIs | #PerconaLive

Fusion-io, Gary Orenstein, theCUBE, PerconaLive 2014Broadcasting from PerconaLive : MySQL Conference and Expo, theCUBE co-hosts John Furrier and Jeff Kelly invite Gary Orenstein, Fusion-io CMO, who takes the opportunity to announce some new features around his company’s flashware applications.

Orenstein begins with some background on Fusion-io, describing how it started with a basic question: If we want to optimize the entire stack, what’s the path? Fusion-io believes that path begins with optimizations around flash memory. With that mantra in mind, perhaps the most notable of Fusion-io’s announcements this week is the new NMV Compression, a flash-aware interface that can double the usable capacity of Fusion-ioMemory while eliminating the performance impact associated with disk-era compression algorithms common in flash SSDs. Orenstein says NMV can deliver up to 4x more flash endurance by streamlining commands to optimize MySQL databases for persistent flash memory architectures when integrated with Atomic Writes interface. The end result is a diminishing overhead for using compression with MySQL.

 

Hyperscale + MySQL market

 
Furrier asked Orenstein how the MySQL marketplace was reacting the enterprise data challenges. Orenstein has this to say:

““To focus a little bit on the MySQL community in regards to that question, we’re seeing across our customer base tremendous usage of MySQL, Percona, MariaDB to solve web application challenges. And that could be anything from online games to messaging applications, which of course we know are all the rage, to online text message voting applications, you name it. We’ve even seen MySQL being used very widely in areas for analytics, advertising analytics, and consumer insight. And across the board what we’ve found is that people really like the familiarity and the relational nature of MySQL, but have had challenges where they’ve needed to, perhaps, scale out a little bit more than they wanted to or a little bit more than their data center could handle.”

Using flash memory, enterprises are able to:

  1. Shrink their footprint
  2. Consolidate the number of servers
  3. Improve efficiency in terms of number of transactions delivered

“Essentially, you’re able to build applications that couldn’t be built at a certain economic number previously,” says Orenstein.

WebScaleSQL, the scalable and customizable MySQL database

 
 .
Another big news item the hosts asked Orenstein about was WebScaleSQL. This week Google, Facebook, LinkedIn, and Twitter launched project WebScaleSQL, in which engineers of four companies will come together to solve the problems of information processing in massive databases. As the name implies, WebScaleSQL is a fork of MySQL (right now, that’s MySQL 5.6), optimized for large scale databases. WebScaleSQL will broaden existing activities in the MySQL community. Orenstein quotes in their words, “A branch of MySQL that will be optimized for the type of web transactions that they’re working on.”

An interesting takeaway from the announcement for Orenstein is that these are four companies we inherently trust with our data. Additionally, they didn’t go and build something net-new, but instead built on improving an already existent technology. A quarter of multi-DMSs already run on MySQL, Orenstein notes, reminding that we’ve already seen very successful models for sharding, or partitioning the database to grow it. So MySQL has proved it can scale, it just scales differently.

Companies are becoming increasingly reliant on their ability to capture, process, and present information. In essence, ingest what you can ingest, run analytics and queries, then deliver actionable insights.


 

Webscale + Hyperscale for the enterprise = The future

 
With non volatile memory compression in mind, Orenstein shares this scenario:

“When you look at that data stack or that transaction stack, how do I go from the piece of data that relies on some kind of storage media out to the end user? In between there you’ve got some kind of storage infrastructure, you’ve got a server infrastructure, you’ve got operating system, database, application and then end user on their new fangled mobile device. Ever since Fusion-io began, we took a different track. We said, as this new memory was coming out… how can we use this to the best advantage, and how can we be different from the architectures that may be tying us to the past?”

One thing Fusion-io did was instead of using storage protocols on flash memory, it exposed flash natively to the CPU in the operation system so that it could perform quickly, as opposed to protocols designed for rotating spinning media. That philosophy of thinking about new memory in terms of its potential for the future as opposed to just thinking about it as how it would fit with the past was a huge differentiator for Fusion-io.

Applications and databases that rely on storage, previously needed disc drives and DRAM. But now, flash memory is a great new memory tier that is in between, according to Orenstein. It takes the best of disc (persistent) and combines it with the best of DRAM (really, really fast).

Fusion-io took a look at compression, recognizing its business appeal. MySQL compression was a no brainer. But why wasn’t the community using it? In talking with Fusion-io clients, Orenstein says the feedback they got was that the overhead and cost of a slight performance penalty was turning people off. Fusion-io knew they had a win that was just waiting to be tackled, compressing data and gaining more usable space.

So Fusion-io went to work. By exposing APIs from flash memory to MySQL, and with a few minor tweaks to MySQL, Fusion-io has been able to essentially eliminate the overhead impact. Its clients are now able to utilize the compression that’s built into MySQL. Users are seeing a huge gain in usable capacity.

KPIs for new modern era in computing

 
Orenstein shares a case study from eBay:

“In December 2013, eBay shared some key performance indicators (KPIs) – eBay is measuring URLs served per kilowatt. What an interesting idea, that they’ve understood that as a company they’re serving web pages… we all know how much we refresh when we’re doing an eBay auction. Their serving web pages is the most important thing they do, and their biggest cost component to that equation is electricity, so they’re optimizing for URLs per kilowatt.”

Fusion-io takes a very workload specific view. That means that with topline performance, they want to get as close to the end user as far as that performance is possible. In their research, they’re noticing that latency consistency is fast becoming a performance must. Orenstein stresses that companies need to seek out what works for them. Topline performance, dollar costs, space costs, power costs…companies should come up with the right equation that works for their specific workload.

Kelly chimed in that it sounds a lot like mapping business requirements to the technology. Orenstein and Furrier agree — in some cases flash is going to be relevant, same for SQL and MySQL. But it’s a workload specific question and there is no silver bullet.

“The good news is that while the scale of those problems might be different between one of these web giants and a smaller company, the nature of the problems is not that different,” says Orenstein.


A message from John Furrier, co-founder of SiliconANGLE:

Your vote of support is important to us and it helps us keep the content FREE.

One click below supports our mission to provide free, deep, and relevant content.  

Join our community on YouTube

Join the community that includes more than 15,000 #CubeAlumni experts, including Amazon.com CEO Andy Jassy, Dell Technologies founder and CEO Michael Dell, Intel CEO Pat Gelsinger, and many more luminaries and experts.

“TheCUBE is an important partner to the industry. You guys really are a part of our events and we really appreciate you coming and I know people appreciate the content you create as well” – Andy Jassy

THANK YOU