GigaIO’s game-changing FabreX™ technology unlocks the next generation of data center rack-scale solutions, where the ability to bring massive compute capability to bear on a problem, harnessing all the possible performance from the system, and where keeping costs – both CapEx and OpEx – in check are essential. The following are some of the solutions areas where FabreX helps customers do more for less.
The digitization of the real world opens the door to endless opportunities for helping to make everyday life better for all mankind through the incredible advances in Artificial Intelligence in all its forms. The data sets are immense and making sense of the data requires building accurate neural networks that are constantly being tuned. You need access to banks of GPUs and FPGAs and the fastest connections between them and storage. However, this is an incredibly fast evolving environment and you need an infrastructure that is exceptionally agile, as you are one algorithm change away from having static hardware becoming obsolete. FabreX gives you the performance you need and the agility that is essential. Pursue the leading edge in AI and we will help propel you there.
Data Analytics and Data Science workflows have traditionally been slow and cumbersome as they rely on server CPUs to load, filter and manipulate data for training and deployment models. Today data scientists are rapidly adopting Machine Learning software with GPUs and/or FPGAs to substantially reduce infrastructure costs and to provide superior performance for end-to-end data science workflows. The ability to mix several types of GPUs or FPGAs in a single workflow with FabreX also dramatically increases performance and efficiency. A FabreX-optimized data science workflow increases productivity by reducing the time spent waiting, allows you to accomplish more by accelerating machine learning training, and reduces infrastructure costs.
Bioinformatics & Genomics
The world of Bioinformatics and Genomics is going through a transformation. The good news is that critical applications are now supporting GPUs and FPGAs, in addition to CPU-only computing, thus dramatically lowering the time to solution per case. More cases can now be run with the same infrastructure. Now the question is how to deploy these expensive resources efficiently so that all users can gain access to these application accelerators, within your budget constraints. Only FabreX enables you to both share resources between groups, AND to combine the compute power of your servers and all your accelerators to gain even more processing throughput when necessary.
High Performance Computing (HPC/Simulation)
HPC is about bringing a whole new understanding to the workings of the physical world to better explain the world around us, and to help us to better predict the future. Geophysical modeling, weather forecasting, bioinformatics, physical simulations, the workings of the financial markets all use advanced heterogeneous computation architectures to enable faster execution of compute-intensive tasks.
Artificial Intelligence is playing a larger role in interpreting the data and in helping to bring meaning. FabreX creates a flexible, performant architecture for your dynamically changing workloads, combining sub-microsecond latency and high-bandwidth networking between all resources and servers in your HPC systems. Slurm is the standard in this market and now you can submit your job as you have always done, and we do all the work to create an optimized infrastructure to run your job. It’s easier, and faster than ever to get results when FabreX works for you.
Driven by the vast volumes of data generated by the rapidly growing number of sensors in IoT and IIoT, there is an increasing need to move compute where the data is being generated. To move processing to the edge, your challenge is to make the most efficient use of resources to deal with space and power constraints. FabreX, with half the power consumption of Infiniband or Ethernet, and the ability to dramatically increase the utilization rate of your resources, enables you to deploy less equipment, gaining space, power and thermal efficiency.
Visualization fuels innovation – from the high-end workstations to servers to the data center rack and cloud – delivering unparalleled performance and capabilities to give insight to cutting edge product development, media and entertainment, or aiding in the search for natural resources, climate change or simply virtual reality. Artificial Intelligence is playing a role in visualization to fix or enhance images and accelerate the rendering of images. The ability to mix several types of GPU or FPGA in a single workflow with FabreX also dramatically increases performance and efficiency.
In a game where a nanosecond can be the difference between millions of dollars lost or made, trading platforms hunger for the lowest possible latency. FabreX gives you a performance advantage both at the network and the messaging levels. With the absolute lowest latency available of any computer network, FabreX is ideal for building your next generation order management system.
Whether in public or private cloud, optimizing resource utilization through virtualization is key. Your challenge is to move beyond virtualizing only storage, to also include accelerators, storage-class memory, and even CPUs and memory. FabreX gives you the sub-microsecond latency you need to build entire racks where every resource can be dynamically composed. Together with our Ready-to-Run software solutions, your on-premise architecture can move beyond enterprise-class, to Cloud-Class™: as agile and flexible as the public cloud, but under your control and at a fraction of the cost.