Fuzzball: HPC-2.0

Hybrid Performance Computing Platform as a Service
Improving Software Infrastructure for Everyone

HPC2.0 – The Next Generation of High Performance Computing, a Cloud Native Federated Computing Platform

What is Fuzzball?

Imagine a computing environment so powerful that it can orchestrate workflows, services, and data globally across data centers while maintaining supply chain integrity from on premise, to cloud, and to the edge. 

Include the ability to support multiple geographic locations and clouds, all unified and federated into a virtual cloud depending on architecture availability, cost, and data management policies.

Interconnects

Fuzzball is HPC-2.0, and we couldn’t call ourselves HPC if we didn’t optimally support fast interconnects.  We are proven on Infiniband but we have tested with PCIe Interconnect and we can actually compose with PCIe Interconnect solutions on the market today and some that aren’t yet.

Scaling

Capable of efficiently scaling resources up or down as needed to meet demand while controlling costs in the clouds today and in the future we can do this on-premise as well with our entire stack.

Clouds

Fuzzball can deploy to a single CSP within minutes and we can connect to multiple geographies in that CSP while expanding to other CSPs as needed.  All the major CSPs are covered already.  We have a tried and tested process to include any cloud needed with a few weeks notice.

Distributed HPC  

Geographically dispersed on-premise supercomputers, Fuzzball is designed as a cloud native and hybrid, federated computing platform for unification of geographically distributed HPC instances whether on-premise or cloud.

Federated Hybrid Orchestration

Orchestration across architectures and resources of workflows and data.  Scheduling based on cost of compute, data & availability of compute.  Data (including: Locality, Mobility, Gravity, and Security) management to align applications with data on the resources for optimization.

Cloud Native HPC

HPC resources can be on-premise, multi-premise, cloud, multi-cloud, and federated.  Cloud based resources are elastic based on jobs, resource policies & data.  Our Fuzzball platform supports all major clouds natively, Kubernetes, as well as custom cloud resources.

A Unified Workload and Resource Platform

Unifying the end-user and administrator experience no matter where you utilize the platform is a design principle.  A user interface (UI) application for the submission, tracking, and management of HPC jobs and workflows which is completely API driven with a command line interface as well as a GUI interface. 

Federated Hybrid Orchestration

Orchestration across architectures and resources of workflows and data.  Scheduling based on cost of compute, data & availability of compute.  Data (including: Locality, Mobility, Gravity, and Security) management to align applications with data on the resources for optimization.

Cloud Native HPC

HPC resources can be on-premise, multi-premise, cloud, multi-cloud, and federated.  Cloud based resources are elastic based on jobs, resource policies & data.  Our Fuzzball platform supports all major clouds natively, Kubernetes, as well as custom cloud resources.

A Unified Workload and Resource Platform

Unifying the end-user and administrator experience no matter where you utilize the platform is a design principle.  A user interface (UI) application for the submission, tracking, and management of HPC jobs and workflows which is completely API driven with a command line interface as well as a GUI interface. 

Welcome to HPC-2.0

Fuzzball is the cloud-native, cloud-hybrid, meta-orchestration platform for all performance intensive workloads and data.

Fuzzball Availability

Fuzzball will be released late Q2 / early Q3 in 2022 but we are currently doing active proof of concepts and granting early access today.

If you are interested in seeing a demo of Fuzzball, or learning more about how Fuzzball can help to accelerate your computational needs, reach out to us to schedule a time to talk.


Flexible support plans for your needs.

We support people, not sockets or cores. People.

Why? Because we believe in empowering people to do amazing things. When people succeed, teams succeed, when teams succeed, organizations succeed. We start where it matters most: with the people.

With this model, we don’t count systems, entitlements, cloud instances, virtual machines, or containers because they don’t count. It’s the people who count.

Standard

  • User Effort
  • Self Help Full Access
  • 6:00 AM – 6:00 PM PT
  • Email, Chat, Phone
  • Response Time < 6 Hours
  • System Administration Effort

Advanced

  • User Effort
  • Self Help Full Access
  • 24 x 7
  • Email, Chat, Phone
  • Response Time < 2 Hours
  • System Administration Effort
  • Engineering/ Development Effort

Latest Information