ARTIFICIAL INTELLIGENCE

1. HIGH THROUGHPUT COMPUTING

Where is Moore’s Law when you need it?

Moore’s law states that processing speed will double every two years. In theory, over the last 10 years, run times should be 32 times as fast. This is far removed from the experience in most industries.

The hardware alone won’t do it

Benefits have not been realised because new programming techniques are needed to exploit the abilities of the latest generation of hardware. Doing the same things the old way with faster computers will not circumvent bottlenecks. There are two ways to transform performance:

.NET optimization

GPU acceleration

Both of which are complex and generally beyond the scope of regular programming studies.

.NET Optimization

The .NET platform is massive and geared towards enabling most programmers. Sections of the framework that accelarate speed come at the expense of ease and convenience. In addition many new enhancements improve performance but most existing code was written (and most programmers were trained) before they existed. Analycat keeps track of the latest changes in .NET to speed up your code and trains your team to exploit the full benefit of the platform.

GPU – Graphic Processing Units

While not a new technology, graphic processing is coming of age and transforming. There is already a graphics card with 5000 processors in one unit. This card alone has the same processing power as the world’s largest supercomputer in the year 2000. GPUs are now outpacing CPUs dramatically.

It needs to be done carefully

Analycat has a complete set of strategies to accelerate your soware in small, safe steps and to train your existing staff and upskill them where possible.

You keep control and ownership

Unlike other consultancies, Analycat will let you own the source code to optimisations which are done and enjoy the benefit of our training to maintain it.

Free hands-on introduction

Contact Analycat now for a free 3-hour seminar on GPU training in your premises. The first hour accommodates management as well as programmers while the platform is introduced conceptually and explained in layman’s terms. For the last 2 hours only the programming team remain for a tour of the platform, guided installation and instruction up until they have built their first application and debugged it.

NVIDIA-square
2. UP2TOM

The ability to rapidly create and deploy virtual experts

TOM is a technology that allows experts to transfer their knowledge to computers. This is really the ability to replicate the decisioning performance of the top experts in an organisation. These virtual experts are created using the Merlynn TOM (Tacit Object Modeller) technology. Virtual experts have near limitless capacity and can be integrated and deployed anywhere and in any operational environment.

Real time decisioning control

UP2TOM is an application that houses the individual TOM models and enables their practical utilisation. It is built in 3 easy steps:

1. Specify what the decision is and what questions you would ask to make the decision.

2. Teach TOM. TOM will ask you numerous configurations of your questions to find your underlying rules.

3. Distribute and learn. Distribute the model throughout your organisation so everyone can benefit from your judgement. Learn the hidden rules that guide ‘expert opinion’ in your organisation.

Copy of TOM Cat 2016_no Pricing