At NEXT 2018 this last September, John Mcgee, our VP of Portfolio Marketing hosted a session on What’s NEXT: Innovation at Hitachi. In this session he explained how Hitachi thinks about innovation, how Hitachi is bringing new solutions to market, and introduced some of the things that we are looking at further down the road. One of the areas he emphasized was the need for data driven insights.
John pointed out that every application is becoming an analytics app. How do you get more insights to drive business outcomes? Machine learning apps are getting updated and refreshed. Every customer interaction is informed by the latest data and the latest model. It’s not like the old days when you develop an app and put it into production, now it is about continual development and learning. Data is no longer processed in the back room by data scientists working on batch data. The data must be streaming in real time, where we inject it with meta data, so we know the context of the data and can curate it to provide the right data to the right consumer for business purposes.
To assist with data driven insights, one of the areas for future development is in the area of Field Programable Gate Arrays (FPGAs). FPGAs are semiconductor devices that are based around a matrix of configurable logic blocks (CLBs) connected via programmable interconnects. FPGAs can be reprogrammed to desired application or functionality requirements after manufacturing. More importantly FPGAs are wired just like electronic circuits, so you can have truly parallel circuits. Unlike micro-processors that process code sequentially or jumps from one piece of code to another to simulate parallelism, FPGAs can process multiple code in multiple parallel circuits, which is ideal for accelerating tasks like graphics processing or machine learning at a much lower cost than micro-processors. If you are not familiar with the benefits of FPGAs, take a few minutes to view this YouTube Video
Hitachi has been working with FPGA technology for some time with thousands of man hours invested in research and development, producing over 90 patents. The most visible use of FPGA’s in Hitachi products is our high-performance Hitachi NAS which enables us to run 4 parallel processing streams where traditional NAS controllers run one. What is exciting will be the use of FPGAs in accelerating a variety of analytics applications with Pentaho analytics being a first use case and Proof of Concept. Pentaho engineers have demonstrated 10 to 100 time faster analytics with much less space, much less resources, and at a fraction of the cost to deploy.
The parallelism of FPGAs makes them very fast, but they have been very specific to use case and workload. So, what the researcher are working on is a software defined FPGA accelerator that can use a common FPGA platform on which we can develop algorithms that are much more transferable across workloads. The benefit will be the acceleration of insights on many analytic opportunities, many different application types, and brings things out faster to market. In this way we hope to crunch those massive data repositories and deliver faster business outcomes and solve social innovation problems. It also means that as we see data gravity pull more compute to the edge, we can vastly accelerate what we can do in edge devices with less physical hardware because of the massive compute and focused resources that we can apply with FPGAs.
Other topics covered in this presentation were Edge to Cloud data services, Neural Network Storage Research, Digital Twins, Machine-learning Model Management, and Optimizing Operations with Video Analytics.
If you missed this presentation at our NEXT2018 event you can see a video of this presentation at our NEXT website which contains most all the presentations at this event. This particular presentation is under General session Day 2.
from Hitachi Vantara Community : All Content - Hu's Place https://ift.tt/2CC1JGR
via https://ifttt.com/ IFTTT
No comments:
Post a Comment