GET STARTED
by subbu on April 5, 2015

Welcome to Bitfusion

Many key areas—pharmaceuticals, data analytics, deep learning, financial services—require complex computations to improve turn-around time to speed-up time-to-market and boost company profits. The volume of data and demand for compute continues to mount and the only only recourse has been to keep scaling: adding CPU-based server nodes. Scale-out definitely helps, but it also means ever lower efficiency as more nodes are added, increased complexity from managing many nodes, increased expenses, and ultimately higher response times overall.

Better hardware options exist. There are GPUs with thousands of cores that can be used to parallelize a variety of applications and reconfigurable chips like Field Programmable Gate Arrays (FPGAs) that can be reconfigured on the fly for a specific application at hand. Implementing software on top of more specialized hardware has been proven to result in large performance and efficiency gains. However, the time and expertise that’s required to take advantage of these architectures has kept these gains largely out of scope for a vast number of users and developers.

This is where Bitfusion comes in.

Our software layer abstracts and exposes the benefits of all available hardware like FPGAs and GPUs to software people enabling them to interact with this hardware just like they interact with CPUs. Instead of starting with performance, imagine starting with “zero friction” and moving towards increased performance, and at no point compromising the user experience. The user sees increased performance with absolutely no intervention.

At Bitfusion we developed a software layer between the user and any type of hardware to provide an order of magnitude performance increase. Our solution is to leverage hardware acceleration to greatly boost application performance in any vertical, without adding complexity or cost – transparently, automatically and without any application rewrite. We believe in performance density or scale-up over just normal scale-out of compute nodes. The benefits are immense: better response times, easier infrastructure management, cheaper, and more power efficient.

The flood of data is not about to recede anytime soon. More and more companies now depend on data collation and analytics to guide just about every kind of business initiative, which means all those organizations, will need supercomputing-class performance.

Gaining supercomputing-level performance won’t just help companies to do the same things faster and better. It will fuel innovation by enabling new capabilities—pursuing new markets with new products and services. That’s why we believe this is so important.

The flood of data is not about to recede anytime soon. More and more companies now depend on data collation and analytics to guide just about every kind of business initiative, which means all those organizations, will need supercomputing-class performance.

Our vision is to make the power of supercomputing accessible to anyone by unleashing hardware acceleration and fundamentally solving the performance problem via performance density and scale-up. No longer does exotic accelerator hardware require highly specialized experience or years of R&D. With Bit Fusion, supercomputing will now be available on demand to software developers everywhere.

Topics: Blog

Search

New Call-to-action

Stay Up-to-Date!

Get our regular deep learning and AI news, insights, tutorials, and more.