Co-funded by the Horizon 2020 Framework Programme of the European Union
Login
Twitter
Mikelangelo - Horizon 2020 Project on Virtualization, Cloud Computing, and HPC
Mikelangelo – Horizon 2020 Project on Virtualization, Cloud Computing, and HPC
  • Blog
  • Partners
    • XLAB
    • HLRS
    • IBM
    • Intel
    • GWDG
    • Pipistrel
    • Scylla
    • Ben-Gurion University
    • Huawei
  • Technology
    • == Whitepapers ==
    • IOcm – IO Core Manager
    • Full-Stack Monitoring
    • Unikernel Application Management
    • The importance of security in cloud and HPC environments is no scam
    • UNCLOT – UNikernel Cross Level cOmmunication opTimisation
    • LEET – Lightweight Execution Environment Toolbox
    • vTorque – virtualization support for Torque
    • vRDMA – Virtio Remote Direct Memory Access
    • ZeCoRx – Zero Copy Receive
    • Universal Unikernel – OSv
    • Seastar library
  • Use Cases
    • A Virtualised Big-Data Software Stack
    • Cancellous Bone Simulation
    • Cloud Bursting
    • OpenFOAM
  • Releases
  • Reports
  • Media
  • Collaboration
Menu back  

Vision

The vision of MIKELANGELO project is to make virtual infrastructures ready to run big data, high performance computing, and I/O intensive applications in production.

Cloud and HPC architectures are inefficient. Layers, introduced by legacy and compatibility requirements, result in complexity of software execution, set-up, management and security. MIKELANGELO is targeting the sweet-spot between efficiency, stability and security.

We focus on lowest layers of virtual infrastructure – on virtual IO in the hypervisor (KVM), guest (unikernel – OSv) and between them (virtual RDMA – vRDMA).


Building on these pillars, supporting infrastructure is built:

  • OSv Application Management (link);
  • Seastar system library for improved parallel performance (link)
  • KVM extensions for IO improvement (link)
  • vRDMA extension (link)
  • vTorque (extensions for PBS Torque)
  • SCAM (link)
  • Snap, instrumentation and supporting framework (link)


Download different releases of our components using our Releases webpage. You can speed-up only your KVM; alternatively, just use OSv under KVM; plug everything under OpenStack (we are compatible) or deploy OSv using vTorque (backwards compatible with Torque).

Components can be used in typical use-cases falling in Cloud, HPC and HPC-Cloud categories. Mix and match them to fit your use case.

 

Consortium

Blog Articles

Running OSv-based Apache Spark Workers with Kubernetes
January 31, 2018
MIKELANGELO at Linux Plumbers Conference 2017 in Los Angeles
November 14, 2017
The Art of Preparing OSv Packages
August 21, 2017
Tweets by @mikelangelo_eu

Case Studies

A Virtualised Big-Data Stack

Cancellous Bone Simulation

Cloud Bursting

OpenFOAM

Reports

  • D2.1 First Cancellous bone simulation Use Case Implementation strategy
  • D2.4 First Cloud-Bursting Use Case Implementation Strategy
  • D2.7 The First Virtualized Big Data Use Case Implementation Strategy
  • D2.10 The First Aerodynamic Map Use Case Implementation Strategy
  • D2.13 The first sKVM hypervisor architecture
  • D2.19 The first MIKELANGELO architecture
  • D7.16 The initial Data Management Plan
Mikelangelo - Horizon 2020 Project on Virtualization, Cloud Computing, and HPC
  • Home
  • Blog
  • Partners
  • Technology
  • Use Cases
  • Deliverables
  • Imprint
Secondary Menu