Rethink BI : Business Insights over Business Intelligence
The purpose of this business insights thought leadership blog is to share the HP point of view on industry trends such as Big Data and Real Time Analytics, and provide updates on key innovations and solutions.

Big Data architecture in the New Style of IT—Part 5

This is the final post in our series with Greg Battas, CTO of Business Intelligence Solutions for HP Converged Systems. Last time, we debunked some mythology around Software Defined Storage. In this post, we conclude with salient reasons why hardware designed specifically for Big Data is an efficient, cost-effective solution. 


Is there currently hardware specifically designed for Big Data?

THPProLiantMoonshot_server1_Ctcm2451393756_Ttcm245108560332_F.jpgGB: Yes and no. Hopefully HP Moonshot is becoming a common answer to some of these server issues. There’s no doubt that Moonshot can be optimized to handle specific application issues. For example, a single Moonshot with onboard DSP could handle voice processing in a way that it used to take racks and racks of servers to accomplish. It’s that efficient. But I feel the significance of Moonshot goes beyond it’s current functionality. It’s taught us the power of efficiency.


Learn more about HP Moonshot.  


For example, there’s this notion floating around out there about something called “dark silicon.” I know it sounds like something out of Star Wars, but the idea is that we can now fit more transistors on a chip than we can power. A few years ago, if you would have suggested this, IT engineers would have looked at you like you had a third eyeball. But it turns out, as a chip grows it’s surface area grows faster than it’s edges, so eventually it’s hard to get enough power onto the chip. We also know that we can power specific regions of the chip, which opens up the door for all kinds of processing possibilities.


I expect we will see SOC’s that have specialized acceleration for certain workloads and these extra functions can be powered on depending on the use case. Because it is very cost-effective to add function to an existing SOC, we lower the barrier to accelerate a lot of databases and filesystems. So we see a lot of chip vendors experimenting with what they can put on the piece of silicon that will actually help the database work better. That’s what I mean by optimized hardware. And open source opens the door for all of us — chip set makers, SOCs, NoSQL vendors and hardware providers — to work together to optimize solutions that accommodate the demands of Big Data.


Why are we working on Big Data?

GB: Big Data poses a number of challenges, it’s true. But it also poses a series of unique possibilities and opportunities. There are numerous benefits to converging common infrastructures with Big Data clusters. When you converge these two systems, you can leverage shared resources for multiple Big Data environments. For example, when you combine several Hadoop clusters you can provision resources more quickly, and allow for rapid elasticity without moving the data. It also allows customers to store the data one time, while opening the door to operating on it with different types of compute resources, maybe even optimized for different types of work.


We also want to bring Big Data software together into a common framework. At HP we have started to do this with HAVEn bringing together Hadoop, Autonomy, Vertica and our Enterprise Security solutions. The “n” in HAVEn stands for other partners to be a part of that converged system. So our customers can use their own products or partner products together in a solution that’s optimized for their specific Big Data infrastructure. Of course all this data would be aligned around a common distributed file system that is HFDS-compliant.


Learn more about HAVEn.


We’re also commited to assisting our ISV partners and customers in this move toward a NoSQL and NewSQL architecture. Frankly, HP has a lot of intellectual property in database management in our labs and with our existing products, such as Vertica, that we feel can assist our ISV partners in this move. Today, many of the ISVs are doing hand-coding to do this work, and we have solutions that we think can really help them eliminate this step.


Finally, we hope to use HP Moonshot to leverage this shift toward hardware optimized for Big Data. We have resources in hardware innovation, systems design, silicon manufacturing, software — you name it. At HP, we have the opportunity to get all these folks into a room together and focus on an end-to-end solution for a business problem.


Greg_Battas_badge_176x304_tcm245_1428057_tcm245_1422290_32_tcm245-1428057.pngAbout Greg Battas

Greg’s background in solving business problems for customers — in particular those in the retail, telecommunications and financial services sectors — and in product development for relational database management systems, has played a critical role in helping bridge the gap between the viewpoints of IT and business decision-makers to explain how to use technology to solve challenging organizational issues.

Leave a Comment

We encourage you to share your comments on this post. Comments are moderated and will be reviewed
and posted as promptly as possible during regular business hours

To ensure your comment is published, be sure to follow the community guidelines.

Be sure to enter a unique name. You can't reuse a name that's already in use.
Be sure to enter a unique email address. You can't reuse an email address that's already in use.
Type the characters you see in the picture above.Type the words you hear.
Showing results for 
Search instead for 
Do you mean 
About the Author

Follow Us
The opinions expressed above are the personal opinions of the authors, not of HP. By using this site, you accept the Terms of Use and Rules of Participation.