Carrier networks are undergoing their biggest transformation since the beginning of the Internet. The ability to get to market quickly and to respond to constantly changing network demands and location-based demographics is the key to success and growth. At the same time, video, mobile, and cloud usage is driving huge growth in traffic and overwhelming current networks. To meet these challenges and thrive, you must find new ways to architect and build secure network infrastructures that can improve agility and reduce operating costs. One approach is to apply successful IT technologies, such as commercial off-the-shelf (COTS) hardware and mass virtualization and cloud deployment of applications. Technologies such as OpenStack and Kernel-based Virtual Machine (KVM) have emerged as de facto standards for cloud and virtualization. Combined with data plane acceleration technologies and virtual machine (VM) management services, they are defining a new class of network virtualization platform. FitOS is the industry’s highest-performing portfolio of network virtualization platforms.
Fault tolerance to single and multiple software and hardware faults.
Support for a variety of redundancy models, including 1:1, 1:N load-shared, N:1, and N:M, ensuring that a single fault cannot impact service. Automatic VM recovery on failure of a host compute node (node failure detection in seconds rather than minutes).
Automatic VM recovery on failure of a VM (60 times faster than standard IT grade).
Fast live migration of VMs—even those using DPDK.
Overlay on top of OpenStack cloud VM management, providing six-nines availability.
Implement software upgrades and rollbacks, patch management, support for live patching and hitless upgrade.
VM and application graceful shutdown. VM management, fast and easy VM definition and creation.
High-availability management of applications, and carrier grade fault management, isolation, and recovery.
Predictable performance through validated and restricted resource assignment to VMs.
Automatic resource scaling, increasing or decreasing a VM’s resources without requiring a restart.
Small two-node configuration ideal for CPE and appliance virtualization.
High-performance VM-to-VM communication by optimizing internal stream lookup algorithms.
Guest network abstraction (logical versus physical).
High-performance DPDK-based accelerated virtual switch for highest packet performance. Support for SR-IOV and PCI pass-through.
Support for VM access to high-performance hardware encryption and compression accelerators.
Accelerated distributed virtual router (DVR), scalable accelerated routing with no single point of failure.
